Document 06
RNAi-based techniques, accelerated
breeding and CRISPR-Cas: basics
and application in plant breeding
Impressum
Eigentümer, Herausgeber und Verleger
Bundesministerium für Gesundheit und Frauen (BMGF)
Radetzkystraße 2, 1030 Wien
Autorinnen und Autoren
Dr.in Julia Hilscher
Univ. Prof. Dr. Hermann Bürstmayr
Department für Nutzplanzenwissenschaften und Department für Agrarbiotechnologie, BOKU Wien
Univ. Prof.in Dr.in Eva Stöger
Department für Angewandte Genetik und Zellbiologie, BOKU Wien
Der Bericht steht zum Download auf der Website des BMGF unte
r www.bmgf.gv.at im Bereich
Gentechnik zur Verfügung.
Erscheinungsdatum März 2017
ISBN 978-3-903099-18-0
link to page 7 link to page 8 link to page 9 link to page 9 link to page 9 link to page 9 link to page 10 link to page 10 link to page 10 link to page 10 link to page 11 link to page 12 link to page 12 link to page 14 link to page 15 link to page 15 link to page 17 link to page 19 link to page 21 link to page 23 link to page 23 link to page 23 link to page 24 link to page 25 link to page 27 link to page 27 link to page 27 link to page 29
Table of Contents
1 Introduction ............................................................................................................................................... 1
1.1
Structure of the literature-based study and questions addressed ................................................... 2
1.1.1
Definitions and explanatory notes to chapters ......................................................................... 3
1.1.1.1 Definition of terms used in this study ................................................................................... 3
1.1.1.1.1 Intended and unintended effects and safety considerations ......................................... 3
1.1.1.1.2 Intermediate organism – resulting organism .................................................................. 3
1.1.2
Explanatory notes to chapters ................................................................................................... 4
1.1.2.1 Intended and unintended effects .......................................................................................... 4
1.1.2.2 Safety aspects ........................................................................................................................ 4
1.1.2.3 Aspects relating to GMO classification .................................................................................. 4
1.1.2.4 Detection and identification .................................................................................................. 5
1.2
Interaction with stakeholders ........................................................................................................... 6
1.3
Participation at GARNet/OpenPlant CRISPR-Cas Workshop ............................................................. 6
1.4
Recommendations (“Handlungsempfehlungen”) ............................................................................. 8
2 CRISPR-Cas ................................................................................................................................................. 9
2.1
Introduction ....................................................................................................................................... 9
2.1.1
CRISPR-Cas9 mediated genome editing: underlying processes .............................................. 11
2.1.2
Production processes of CRISPR-Cas9 genome edited plants ................................................. 13
2.1.3
Techniques (SDN1,2,3) ............................................................................................................ 15
2.2
Application in plant breeding .......................................................................................................... 17
2.2.1
Potential applications of SDN1 ................................................................................................ 17
2.2.1.1.1 Elimination of unwanted compounds ........................................................................... 17
2.2.1.1.2 Increasing production of desired compounds ............................................................... 18
2.2.1.1.3 Engineering pathogen resistance by targeting recessive resistance genes .................. 19
2.2.2
Potential applications of SDN2 ................................................................................................ 21
2.2.3
Potential applications of SDN3 ................................................................................................ 21
2.2.4
Applications other than genome editing ................................................................................. 21
2.3
State of research and development in plants ................................................................................. 23
link to page 29 link to page 30 link to page 30 link to page 31 link to page 32 link to page 34 link to page 34 link to page 35 link to page 35 link to page 35 link to page 35 link to page 37 link to page 37 link to page 37 link to page 38 link to page 38 link to page 38 link to page 39 link to page 39 link to page 41 link to page 41 link to page 43 link to page 47 link to page 47 link to page 49 link to page 50 link to page 50 link to page 51 link to page 52 link to page 54
2.3.1
Transferability of the system to plant species ......................................................................... 23
2.3.2
Techniques (SDN1, 2, 3) ........................................................................................................... 24
2.3.3
Delivery methods ..................................................................................................................... 24
2.3.4
Types of mutations generated by SDN1 technique ................................................................. 25
2.3.5
Off-target activity .................................................................................................................... 26
2.3.6
Limiting off-target effects ........................................................................................................ 28
2.4
Intended and unintended effects of CRISPR-Cas9 in genome editing ............................................ 28
2.5
Safety considerations ...................................................................................................................... 29
2.5.1
SDN1 technique in genome modification of plants ................................................................. 29
2.5.1.1 Comparison of CRISPR-Cas9 and conventional mutagenesis techniques in relation to
mutational load and type of modifications ......................................................................................... 29
2.5.1.2 Safety considerations in respect to CRISPR-Cas9 transgene retention, background
mutations caused by transformation procedures and the use of viral vectors .................................. 31
2.5.2
SDN2 technique in genome modification of plants ................................................................. 31
2.5.3
SDN3 technique in genome modification of plants ................................................................. 32
2.6
Detection and identification ............................................................................................................ 32
2.6.1
Detection and identification of SDN1 and SDN2 genome editing ........................................... 32
2.6.2
Detection and identification of SDN3 genome editing............................................................ 33
2.7
Aspects of GMO classification of CRISPR-Cas9 genome edited plants ............................................ 33
2.7.1
Evaluation of ZFN and related genome editing techniques by the German expert commission
ZKBS
35
2.8
Tables ............................................................................................................................................... 37
3 Accelerated breeding – rapid cycle breeding .......................................................................................... 41
3.1
Introduction ..................................................................................................................................... 41
3.2
Potential applications in plant breeding ......................................................................................... 43
3.3
State of development ...................................................................................................................... 44
3.3.1
Species of interest and genes tested for precocious flower induction ................................... 44
3.3.2
Experimental systems to induce precocious flower induction................................................ 45
3.3.3
Current rapid-cycle breeding programmes ............................................................................. 46
3.3.4
Establishing infrastructure for rapid-cycle breeding programmes.......................................... 48
link to page 55 link to page 56 link to page 57 link to page 57 link to page 58 link to page 60 link to page 66 link to page 66 link to page 67 link to page 69 link to page 70 link to page 71 link to page 73 link to page 75 link to page 78 link to page 78 link to page 79 link to page 82 link to page 83 link to page 83 link to page 84 link to page 90 link to page 92 link to page 107 link to page 107 link to page 107 link to page 107 link to page 108 link to page 108 link to page 110
3.4
Intended and unintended effects .................................................................................................... 49
3.5
Safety considerations ...................................................................................................................... 50
3.6
Identification and detection ............................................................................................................ 51
3.7
Aspects of GMO classification ......................................................................................................... 51
3.7.1
Evaluation of a related breeding practise by the German expert commission ZKBS .............. 52
3.8
Tables ............................................................................................................................................... 54
4 Small RNA-directed techniques ............................................................................................................... 60
4.1
Introduction ..................................................................................................................................... 60
4.1.1
miRNAs .................................................................................................................................... 61
4.1.2
siRNAs ...................................................................................................................................... 63
4.2
Application of RNAi approaches in plant breeding ......................................................................... 64
4.2.1
Applications based on targeting plant endogenous genes ..................................................... 65
4.2.2
Applications by targeting RNA expressed by plant pathogens ................................................ 67
4.3
State of development ...................................................................................................................... 69
4.4
Intended and unintended effects .................................................................................................... 72
4.5
Safety considerations ...................................................................................................................... 72
4.5.1
EFSA workshop on risk assessment considerations for RNAi-based GM plants ..................... 73
4.5.2
EFSA call on literature review to support risk assessment of RNAi-based GM plants ............ 76
4.6
Detection and identification ............................................................................................................ 77
4.7
Aspects of GMO classification ......................................................................................................... 77
4.8
Table ................................................................................................................................................ 78
5 Abbreviations ........................................................................................................................................... 84
6 References ............................................................................................................................................... 86
7 Appendix ................................................................................................................................................ 101
7.1
Literature Search ........................................................................................................................... 101
7.1.1
CRISPR-Cas ............................................................................................................................. 101
7.1.2
Rapid cycle breeding .............................................................................................................. 101
7.2
Definition of GMO according to EU Directive 2001/18/EC on the deliberate release into the
environment of genetically modified organisms ....................................................................................... 102
7.3
Tables ............................................................................................................................................. 104
Introduction
1 Introduction
The aim of plant breeding is to develop and select plants adapted to human needs [1]: breeding
objectives include abiotic and biotic stress tolerance, increased yield and/or yield stability, but also
for example the development of value-added crops with increased protein content or altered fatty
acid composition.
After being dependent on naturally occurring variation in plants for domestication and later for
breeding, the 20th century brought techniques to support the breeding process and cultivar
development. Mutation breeding is a method of artificially inducing mutations, which form the
genotypic basis of differing traits. Polyploidy induction, i.e. doubling chromosome sets, may lead to
cultivars with higher biomass. Other techniques facilitate re-combining (nuclear and/or cytoplasmic)
genomes, like protoplast fusion. Finally, in the 1980ies transformation of plants with selected
additional genetic material became possible. These and other biotechnological techniques increase
genotypic variation in a given gene pool, which can be utilized directly or as basis for further breeding
material.
Directive 2001/18/EC regulates the deliberate release of genetically modified organisms (GMO) and
Regulation (EC) 1829/2003 the food and feed use of GMOs. GMOs falling under these regulations and
exemptions are defined in Directive 2001/18/EC. Since formulation of the legal definition of a GMO,
progress in research and development brought questions from stakeholders to competent
authorities in European Union Member States on whether certain techniques lead to such regulated
GMO´s. A working group (WG) was established in 2007 to identify and discuss so called “new plant
breeding techniques” (NPBT) in relation to the definition of a GMO and in light of the most recent
available scientific data [2]. The techniques under scrutiny contained (1) zinc finger nuclease (ZFN)
technology, (2) oligonucleotide directed mutagenesis (ODM), (3) cisgenesis and intragenesis, (4) RNA-
dependent DNA methylation (RdDM), (5) Grafting (on GM rootstock), (6) reverse breeding, (7) agro-
infiltration, and (8) synthetic genomics [2].
CRISPR-Cas and accelerated breeding are covered in this report under the light of the above. Based
on modules of the CRISPR-Cas system a genome editing technique was developed, the most recent
addition to site directed nuclease (SDN) techniques, joining ZFNs. Accelerated breeding is a breeding
strategy that uses a GMO to accelerate individual breeding cycles; the resulting plants, though, do
not carry the early flowering transgene. In this sense, using a GMO intermediate in a breeding
process, accelerated breeding has parallels to reverse breeding. The eight NPBT have been covered
in studies conducted by AGES [3, 4]; CRISPR-Cas and accelerated breeding have come into focus very
recently, thus the coverage in this report.
1
link to page 107
Introduction
RNAi-based plants are plants falling under the definition of GMO in Directive 2001/18/EC. They
express a transgene transcribed into an RNA molecule that downregulates a third gene and so
confers the desired phenotype. RNAi-based GM plants have been among the very first commercially
developed GM plants (FlavrSavrTM), however, to date the majority of genetically modified plants
(GMP) authorized in the EU are based on expression of one or more transgenes expressing proteins
that confer the desired phenotype. RNAi-based GM plants have again come into focus for example
because of their potential for engineering pathogen resistance traits. There is an ongoing process in
the EU to evaluate whether the risk assessment implemented for GMPs in general may be specifically
adapted to RNAi-based GM plants. RNAi-based GM plants and the ongoing activities on questions in
relation to adaptation of risk assessment are covered in this report. The sub-category of RNAi-based
GM plants functioning through RNA-dependent DNA methylation (RdDM) has been covered by AGES
[4] and is not further covered in this study.
1.1 Structure of the literature-based study and questions addressed
CRISPR-Cas9 is a novel site directed nuclease technique and accelerated breeding a relatively novel
concept integrated in conventional breeding strategies. For these two, a literature search was
undertaken to collect available primary research publications. Details on the search strategy in
scientific literature databases can be found in Appendi
x 7.1.
Basic research into RNAi based pathways goes back to the early 1990ies. Description of the RNAi-
based techniques, current state of application and development were guided by the most recent
peer reviewed secondary literature present, and where informative to the focus of this study,
expanded by data from primary research publications. Furthermore, publicly available documents by
EFSA informing on ongoing developments on risk assessment evaluation are included.
Literature search ended March 2016.
Following a general description, (i) applications in plant breeding, (ii) the state of development in
plant systems, (i i) intended and unintended effects upon application, (iv) safety considerations, (v)
detection and identification, and (vi) aspects of GMO classification, are addressed for each of the
techniques.
2
Introduction
1.1.1 Definitions and explanatory notes to chapters
1.1.1.1 Definition of terms used in this study
1.1.1.1.1 Intended and unintended effects and safety considerations
GM risk assessment is focused on identifying and characterizing potential adverse effects on human
and animal health and on the environment, both of intended and possible unintended effects caused
by GM-based plants.
The term intended and unintended effects was defined in the “Scientific Opinion on Guidance on the
environmental risk assessment of genetically modified plants”, EFSA Journal 2010 [5]:
“Intended effects are those that are designed to occur and which fulfil the original objectives of the
genetic modification. Alterations in the phenotype may be identified through a comparative analysis
of growth performance, yield, pest and disease resistance, etc. Intended alterations in the
composition of a GM plant compared to its appropriate comparator, may be identified by
measurements of single compounds.
Unintended effects of the genetic modification are considered to be consistent (non-transient)
differences between the GM plant and its appropriate comparator, which go beyond the primary
intended effect(s) of introducing the transgene(s). […] these unintended effects are event-specific,
applicants must supply data on the specific event. Sources of data that may reveal such effects are: 1.
Molecular characterization […]. 2. Compositional analysis […]. 3. Agronomic and phenotypic
characterization […]. 4. GM plant-environment interactions […].”. [5]
In this report, intended and potential unintended effects on the plant genome and derived safety
considerations are specified and discussed for the application of the particular techniques, based on
the current state of the science.
1.1.1.1.2 Intermediate organism – resulting organism
The terms intermediate and resulting organism are used in this study in the chapters covering
CRISPR-Cas and accelerated breeding. In this report the following terms are used as defined in the
NTWG (New Techniques Working Group) final report of 2011; the report was never officially
published but can be accessed via a link in [6].
Resulting organism was therein defined as
“… an organism that results after having gone through al the steps of the particular technique. This
could be a plant or seed intended for deliberate release or placing on the market or a microorganism
intended for contained use.”
and intermediate organism as
“…any organism that is generated in the steps leading to the resulting organism.”
3
Introduction
The NTWG was composed of national experts nominated by the Competent Authorities of EU
Member States in 2008. Their objective was to analyse whether specific biotechnological methods,
including ZFN and related techniques, lead to resulting organisms falling under the definition of a
GMO Directive 2001/18/EC [2].
1.1.2 Explanatory notes to chapters
1.1.2.1 Intended and unintended effects
Intended and potential unintended effects on the plant genome are specified and discussed due to
the application of the particular techniques based on the current state of the science.
1.1.2.2 Safety aspects
Directive 2001/18/EC explicitly excludes plants generated by conventional mutagenesis breeding and
plants generated by cell or protoplast fusion, as well as does not consider plants generated by
polyploidy induction fal ing under the GMO definition; plants generated by these techniques are
exempted from the risk assessment and regulatory procedure established by Directive 2001/18/EC
that – based on the precautionary principle – has the objective to protect human health and
environment.
This is based on the grounds of considerations given in recital 18 of Directive 2001/18/EC which
reads that the “Directive should not apply to organisms obtained through certain techniques of
genetic modification which have conventionally been used in a number of applications and have a
long safety record.”
The Directive therefore implicitly states that the risks associated arising from intended and
unintended mutations by the exempted techniques, mutagenesis breeding, cel culture methods and
bringing together related genomes or multiplication of genomes, are considered to be manageable
outside the regulatory procedure of Directive 2001/18/EC, that is, by the breeding practices
implemented by breeders.
Therefore, unintended effects on the genome arising due to application of these exempted
techniques that may be applied during the production process of CRISPR-Cas9-based genome edited
plants or during rapid-cycle breeding are treated the same in this report.
1.1.2.3 Aspects relating to GMO classification
Directive 2001/18/EC and Regulation EC/1829/2003 provide authorization procedures for deliberate
release and placing on the market of genetically modified organisms (GMO) as well as for food and
feed derived from GMOs. In Directive 2001/18/EC a definition of organisms falling under the
4
link to page 11
Introduction
authorization procedure is given and exemptions are specified (Articles 2 and 3 and Annex IA, IB; see
excerpt in Appendix 7.2 ).
8 NPBTs were assessed by the NTWG, for whether they generate organisms fal ing under the GMO
definition in Directive 2001/18/EC. Similarly, the ZKBS (Zentrale Kommission für die Biologische
Sicherheit), established under the scope of the German Gene Technology Act, published a position
statement [7] on new plant breeding techniques.
Information in this report relating to CRISPR-Cas and to accelerated breeding may be used to
interpret organisms in relation to relevant paragraphs of the legal GMO definition in Directive
2001/18/EC. In this chapter, thus, the techniques wil be described in regard to the different steps
involved in carrying out the techniques and the generated intermediate and resulting organisms.
Where applicable, ZKBS expert opinions on analogous techniques are reported.
1.1.2.4 Detection and identification
To date, most commercialised genetically modified (GM) plants and all GM plants listed in the
European Union GMO register (Regulation EC 1829/2003) are based on integration of transgenes
containing one or more non-plant derived sequences, for example the cauliflower mosaic virus
(CaMV) 35S promoter or bacterial herbicide tolerance conferring phosphinotricin-
N-acetyltransferase
sequences (pat, bar) [8]. Detection of (unauthorized) GMOs uses the common occurrence of these
signature sequences (element and/or construct specific) in various GM plant lines; a platform (JRC-
GMO-Matrix [9], storing information on known GM events) supports in deciding of an optimal
screening strategy for a given sample. A first screening step detecting element and/or construct
specific sequences establishes GM presence or absence (detection). In case GM presence is detected,
validated analyses to identify event-specific sequences are carried out in order to unequivocally
identify unique GM plant lines (identification). An event-specific detection method is an integral part
of an application dossier for any GMO authorization in the European Union. Event-specific markers
span the junction between the transgene insertion site and the genomic target site. Polymerase
chain reaction (PCR) derived methods for detection, identification and quantification are commonly
used (real time PCR using hybridization probes; see European Reference Laboratory for GM Food and
Feed
1).
In this chapter the possibility of detection and identification of intermediate and resulting organisms
of the covered techniques CRISPR-Cas and accelerated breeding will be described.
1 http://gmo-crl.jrc.ec.europa.eu/gmomethods/
5
Introduction
1.2 Interaction with stakeholders
Information material was collected in the course of this study to be used for research education, e.g.
within the program for the “Long Night of Research”, where genome editing was explained to the
general public.
Information gathered within the study was also used for an article in the Austrian journal “Der
Pflanzenarzt” (Neue Züchtungsmethoden: Gentechnik – oder doch keine Gentechnik?, 4/2016, p 24-
27) in several talks held at meetings of breeder´s associations and other stakeholder associations.
• Vereinigung österreichischer Pflanzenzüchter, June 2015
• Klausur der Saatbau Linz, November 2015
• Saatgutgipfel der AGES; April 2016
• Interne Diskussion in der LKÖ zum Thema „Neue Methoden der Gentechnik“, April 2016
1.3 Participation at GARNet/OpenPlant CRISPR-Cas Workshop
Overal , the GARNet/OpenPlant Workshop at the John Innes Centre, UK (September 2015) provided
an excellent environment to meet researchers working with CRISPR in crop species. It gave an update
on state of the art of CRISPR-Cas applications in plants and made aware of where to look for current
and future developments in the highly active field of CRISPR-Cas9 plant genome editing.
The Workshop gave an overview on CRISPR-Cas9 applications, reported on its current use in plant
genome editing and on ongoing developments, especial y in regard to optimization of efficiency and
specificity. A meeting report has been published by the organisers [10].
Speakers presented data of successful genome editing by CRISPR-Cas9 in a wide variety of species,
also in the crop plants maize, rice, wheat, tomato and potato. Vladimir Nekrasov (John Innes Centre,
UK) described the production of a powdery mildew resistant tomato variety (cv “Moneymaker”).
They used CRISPR-Cas to knock out the MILDEW RESISTANCE LOCUS O1 (Mlo1). Homozygous
knockout mutants were present in the first generation of transgenic plants, and transgene free plants
stably inherited the mutation.
mlo1 plants showed complete resistance against
Oidium
neolycopersici. In rice, Bing Yang (Iowa State University, USA) reported CRISPR-Cas mediated
production of two independent OsSWEET13 knock-out lines which conferred resistance to
Xanthomonas oryzae, the causal agent of rice bacterial blight.
Examples of ongoing work to further improve gene editing efficiency at various steps in the process
included a database for gRNA design now also of use for diverse plant species (Edward Perello
Desktop Genomics, UK) or explanation of various multiplexing strategies, like the use of synthetic
tRNA-gRNA polycistronic genes (Bing Yang). The issue of specificity was for example addressed by
6
Introduction
Oleg Raitskin (The Sainsbury Laboratory, UK) who screens variants of Cas9 nuclease and sgRNA
combinations in order to find increased specificity. Holger Puchta (Kit, Germany) presented an
already available strategy to decrease off-target effects by using two Cas9 nickase variants guided to
adjacent positions and so resulting in a desired double strand break only if two nickases are placed in
vicinity.
7
Introduction
1.4 Recommendations (“Handlungsempfehlungen”)
The present report on CRISPR-Cas and accelerated breeding applications in plant breeding provides
background information on the fundamentals and the application potentials of these techniques as
well as the state of development. It describes intended and unintended effects on the plant genome
in relation to other plant breeding techniques and biotechnological methods.
It is intended as an information document for policy makers and stakeholders. The discussion about
the so called new plant breeding techniques (NPBT) and their legal classification in the EU is now
nearing a decade. In the meantime, as exemplified by the existence of this report, further techniques
and breeding strategies have been developed and applied and knowledge on biotechnological
methods and its impact on plant breeding have been increasing. Al of the techniques hold great
potential for utilization in plant breeding and development of crop cultivars. On the other hand, the
legal classification of NPBTs, whether classified as falling under the GMO definition of Directive
2001/13/EC, them being exempted, or development of different regulatory procedures [11, 12], has
consequences on their use and application in plant breeding.
To date, there are solid information documents available by scientific experts on the fundamentals of
the different techniques and their potentials, furthermore, position statements from many
stakeholder groups have been put forward; overall, a huge amount of scientific, legal and economic
efforts have been carried out in regard to diverse aspects of NPBTs and related biotechnological
methods. Therefore, the next step is to be done by policy makers to decide on the handling of NPBTs
in order to ensure legal certainty to developers and plant breeders for their products.
Information of the public by public authorities in respect to plant breeding and biotechnological
methods, their development and application in plant breeding should be an active process and
guided by the current state of science and technology.
8
link to page 15 link to page 107 link to page 16
CRISPR-Cas
2 CRISPR-Cas
2.1 Introduction
CRISPR-Cas
CRISPR-Cas (Clustered regularly interspaced short palindromic repeats – CRISPR associated gene) is
an RNA-guided DNA endonuclease complex present in bacteria and archaea. In 2012 it was
recognized that it can be employed for targeted genome editing [13] and since then publication
numbers have risen to develop and apply genome editing using CRISPR-Cas in various organisms,
ranging from bacterial to human cells (see for example Table 1 in [14])
. Fig. 2.1 illustrates publication
activity for CRISPR-Cas9 in plant research.
Fig. 2.1 CRISPR-Cas9 publications in plant research 2012-2015. Publications were retrieved from
pubmed, Web of Science, Scopus and Ovid according to defined search criteria (see Appendix
7.1).
(A) Publication numbers per year, subgrouped based on experimental (research and methodical
articles), review (reviews, opinion articles and book chapters) or other publication type (meeting
abstracts, publications in languages other than english, etc). (B) Country of origin of scientific papers
reporting experimental data on CRISPR-Cas9 in plant research 2013 – 2015 (based on first author).
CRISPR-Cas is a recently understood adaptive “immune system” in prokaryotes against foreign DNA
and RNA (reviewed for example in [15]). Present in about 90% and >40% of to date known archaeal
and bacterial genomes, three main types of CRISPR-Cas systems have been identified.
Fig. 2.2
outlines CRISPR-Cas function based on type II systems [16]: (1) mediated by CRISPR associated (Cas)
genes, invading DNA is recognised and fragments (termed spacers) of foreign DNA are incorporated
into the bacterial genome at the CRISPR locus; (2) the CRISPR locus is transcribed as precursor RNA;
(3) the precursor RNA is processed into mature CRISPR RNAs (crRNAs), then hybridizes to a trans-
activating CRISPR RNA (tracrRNA) and is bound by a Cas9 protein; (4) the CRISPR-Cas9 complex is
9
CRISPR-Cas
guided to specific DNA locations specified by the spacer region of crRNA component and DNA
cleavage is mediated by the Cas9 protein. The RNA component of the CRISPR-Cas9 type II complex is
also termed dual guide RNA (crRNA hybridized to tracrRNA).
CRISPR-Cas subtypes are classified based on the Cas genes involved, and as a consequence differing
ribonucleo-protein complexes and modes of target interference. CRISPR-Cas type II has also the
ability to target and cleave RNA [16].
Fig. 2.2 Simplified model of CRISPR-Cas organisation, biogenesis and targeting exemplified by the
type II system. (1) The CRISPR-Cas genomic locus contains the Cas protein coding genes and the
CRISPR locus coding for the RNA component of the CRISPR-Cas complex. The latter is composed of
acquired spacers from invading DNA and interspersed repeat sequences. (2) The Cas genes (coding
for example for Cas9) and the CRISPR precursor are transcribed and (3) the CRISPR precursor RNA
cleaved into crRNA moieties, which hybridized to a tracrRNA, is bound by the Cas9 protein. The
crRNA and tracrRNA components together are called dual guide RNA. (4) Mature CRISPR-Cas9
complexes target DNA sequences showing complementarity to the spacer region of the crRNA and
induce DNA double strand breaks. crRNA: CRISPR RNA. tracrRNA: trans-activating CRISPR RNA.
Genome editing
The technology of random mutagenesis is used to induce genetic variability in plant breeding and
research. Upon exposure to, for example, radiation or chemical mutagens a large population of
plants has to be screened phenotypically or genotypically to select those with desired
phenotypes/genotypes. With genome editing technology it is now possible to target genomic
positions to introduce variability, i.e. to generate plants with precise modifications or to insert
foreign DNA at targeted genomic positions. Genome editing has been made feasible by development
of several systems, all based on proteins acting as site directed nucleases (SDN), i.e. enzymes
10
link to page 16 link to page 16 link to page 18
CRISPR-Cas
introducing DNA double strand breaks (DSB): zinc finger nucleases (ZFN), TAL effector nucleases
(TALEN), meganucleases (MN) and recently CRISPR-Cas9 (reviewed for example in [17]). These
technologies share the same mechanism: they are programmable for precise typesetting of DNA
double strand breaks (DSB) which are then recognised by diverse endogenous cel ular repair systems.
In some cases these are imperfect and incorporate errors, alternatively, DSB repair mechanisms can
be tricked into modifying genomic sequences or inserting extraneous DNA by providing repair
templates, all of which is exploited in genome editing.
CRISPR-Cas9 is a genome editing technique that can be used to introduce mutations at selected genomic
loci. It is based on components of a naturally occurring pathway present in bacteria and archaea: the
enzyme Cas9 that is able to introduce a double strand break into DNA; the associated RNA component can
be easily re-programmed to target Cas9 to selected loci of eukaryotic genomes.
2.1.1 CRISPR-Cas9 mediated genome editing: underlying processes
To date, type II CRISPR-Cas9 modules are mainly used for genome editing of pro- and eukaryotes
[14], including plants [18]. The Cas9 protein is mostly based on the sequence of the homolog of
Streptococcus pyogenes (
SpCas9). However, Cas9 homologs of further organisms, as well as other
CRISPR-Cas subtypes are used and/or investigated for application in genome editing, or other uses.
Below, CRISPR-Cas9 - target DNA interaction is explained in more detail to aid in understanding of
the issue of off-target effects; it relates to the type II CRISPR-Cas9 subtype, if not indicated otherwise.
Furthermore, DNA repair pathways operating in plant cells are briefly introduced.
DNA double strand break (DSB) generation by CRISPR-Cas9
To recognize target DNA sequence and execute a DNA double strand break (DSB), a natural CRISPR-
Cas9 complex consists of the DNA endonuclease Cas9 protein (executing the DSB) bound to the
crRNA:tracrRNA (termed dual guide RNA)
(Fig. 2.2). The 5´ end of crRNA harbours the spacer, i.e. the
complementary region for target recognition, the crRNA 3´end hybridizes with the tracrRNA to form
a secondary structure required for Cas9 binding
(Fig. 2.2).
It was discovered that engineering a chimeric guide RNA, called single guide RNA, that carries a
spacer sequence of choice (depending on the desired genomic target) at the 5´end fol owed by a
3´end hairpin structure (mimicking tracrRNA:crRNA secondary structure) also form functional entities
(Fig. 2.3) [13], which is exploited for use in genome editing.
Cas9 proteins possess two separately acting nuclease domains homologous to HNH and RuvC
nucleases, cutting the complementary and non-complementary DNA strand, respectively [13].
However, Cas9 is also involved in target recognition: its PAM Interacting (PI) domain scans target
11
link to page 18 link to page 18
CRISPR-Cas
DNA for protospacer adjacent motifs (PAMs)
(Fig. 2.3). PAMs are short signatures (typically 2-5
nucleotides [19]) directly downstream (type II) of protospacers (i.e. the signature sequences in the
target DNA) but not incorporated into the CRISPR loci that are crucial for target recognition; they also
dictate the location of the DSB executed by the Cas9 nuclease domains. Thus, if Cas9 loaded with
gRNA recognises PAM sequences, the gRNA-Cas9 complex interrogates DNA directly upstream to
PAMs for complementarity to the spacer sequence. In the course a RNA:DNA heteroduplex is
formed, and in case of substantial complementarity target DNA is cleaved approximately three
nucleotides upstream of the PAM at both strands (reviewed for example in [14]). Efficient target
cleavage is dictated by near complementarity of the last 8-12 nucleotides of the spacer sequence
(cal ed seed region) to the target and the presence of the PAM nucleotides in the protospacer
sequence [20]
(Fig. 2.3).
Fig. 2.3 Schematic depiction of elements involved in CRISPR-Cas9 – target DNA recognition. The PAM
interacting domain (PI) of Cas9 scans DNA for PAM sequences (typically NGG in type II system of
Streptococcus pyogenes). In case the protospacer region upstream of the PAM shows high
complementarity (special y in the seed region) to the spacer region of the sgRNA, Cas9 executes a
DNA DSB approximately 3 nucleotides upstream of the PAM in the target DNA. The 8 - 12 nucleotides
constituting the seed region proximal to the PAM are depicted in grey. DSB: double strand break.
PAM: protospacer adjacent motif; PI: PAM interacting domain; sgRNA: single guide RNA.
DNA double strand break (DSB) repair pathways exploited for genome editing
In plant cells, non-homologous end-joining (NHEJ) and homologous recombination (HR) mediated
repair pathways execute repair of occurring DNA DSBs. NHEJ, the prevalent mechanism in somatic
plant cells [21], is error-prone and often introduces smal er insertion or deletion mutations upon re-
ligation of DNA ends. DNA ends ligated together do not need to show homology. HR mediated repair
mechanisms rely on information from homologous regions. DNA ends at DSBs are processed into
single-stranded 3´overhangs by 5´-3´exonuclease activity, and bound by HR-proteins (for example
RAD51) which scan DNA for homologous regions. In somatic plant cells mainly two HR mediated
12
link to page 20 link to page 20
CRISPR-Cas
repair pathways seem to operate, single-strand annealing (SSA) and synthesis-dependent strand
annealing (SDSA). SSA results in ligation of two annealing ssDNA strands. SDSA scans for
complementary regions in duplex DNA by strand invasion and uses a detected homologous strand as
repair template by initiating DNA synthesis. Synthesis finishes, and in case the now extended strand
harbours again complementary sequence to the second resected single-stranded 3´overhang, the
DSB can be repaired (for review see [21]).
Genome editing using CRISPR-Cas9 employs two molecular modules: it uses an engineered CRISPR-Cas9
module to execute a DNA double strand break (DSB) at a chosen site in the plant genome; in a second step,
DNA DSBs, which also occur under natural conditions, are repaired by endogenous DNA DSB repair
pathways. These repair pathways are error-prone, resulting in mutations; alternatively, these may be used
to mediate site specific integration (at the DSB) of cis-, intra-, or transgene.
2.1.2 Production processes of CRISPR-Cas9 genome edited plants
For a given plant species, the production process of CRISPR-Cas9 genome edited plant lines depends
on established transformation and, if a cell culture step is included, regeneration procedures
(Fig.
2.4). They all share a step of (a) delivery of a gRNA-Cas9 module (and optionally a repair template)
into plant cells and (b) screening for genome edited lines. There are several modes of gRNA-Cas9
delivery, including different vector systems, in use. CRISPR-Cas9, and in extension genome editing
techniques involving site directed nucleases (SDN), introduce heritable changes
in trans, therefore
transgenic integration of a CRISPR-Cas9 gene cassette during the production process is not obligatory
and if present, can be segregated out in sexual y reproducing species.
Stable transformation of gRNA-Cas9 gene cassettes: gRNA-Cas9 gene cassettes including a selectable
marker gene are transformed into plant cells and have become stably integrated during a selection
step. gRNA-Cas9 is expressed from transgenic DNA. Transformation methods mainly used are
Agrobacterium-mediated gene transfer and microprojectile (particle) bombardment, or
electroporation and polyethylene-mediated transformation for plant protoplasts. In crop species
which can be propagated by sexual reproduction genome edited progeny free of the CRISPR-Cas9
cassette including the marker gene can be selected in the next generation(s). In this case, transgenic
events are present in intermediate products during the production process but are lacking in the final
established plant line (resulting organism). Production processes involving stable transformation to
date are the main published production processes in plants.
Transient transformation of gRNA-Cas9 gene cassettes: gRNA-Cas9 gene cassettes are transformed
into plant cells and CRISPR-Cas9 is expressed from these templates. Transformation methods are as
13
link to page 21 link to page 21
CRISPR-Cas
above. The production process does not include a selection step for stable genomic integration of the
gene cassette. A second strategy for transient delivery of the gRNA-Cas9 gene cassette uses viral
vectors. They may either be delivered via
Agrobacterium-mediated gene transfer, via virions or
isolated viral RNA (RNA viruses). Genome editing using DNA virus (Cabbage Leaf Curl virus (CaLCuV),
bean Yellow Dwarf virus (BeYDV)) systems [22, 23] and an RNA virus (Tobacco Rattle virus (TRV))
system [24, 25] have been shown to date. RNA viral vector systems were not yet shown to deliver a
complete CRISPR-Cas9 gene cassette, but were shown to deliver sgRNAs into plants stably expressing
the Cas9 component. However TRV virion delivery engineered to express ZFNs has been used to
generate genome edited tobacco lines [26].
Delivery of pre-assembled gRNA-Cas9 ribonucleo-protein complexes: Ribonucleo-protein complexes
are delivered into plant cells and directly exert their function [27]. PEG mediated delivery of particles
has been carried out. This method does not involve DNA delivery into plant cel s in case of SDN1
techniques (for definition of SDN1 please refer to chapte
r 2.1.3).
Fig. 2.4 Production processes of genome edited plants using CRISPR-Cas9 (modified after [28]). Grey
boxes indicate methods to deliver CRISPR-Cas9 into cells. Delivery of CRISPR-Cas9 activity may be
independent or dependent on DNA transfer into plant cel s. Grey unbroken lines: DNA transfer; grey
dotted lines: no DNA transfer; grey dashed lines: DNA transfer optional.
Minimal gene cassette requirements in case of recombinant DNA based transformation procedures
A gRNA-Cas9 minimal gene cassette consists of a Cas9 gene (to date mostly derived from
Streptococcus pyogenes) fused to a nuclear localization signal (NLS) located between a polymerase II
promoter and terminator to initiate and terminate transcription, respectively
(Fig. 2.5). The sgRNA is
driven and terminated general y by polymerase II regulatory sequences. The spacer sequence, in
14
link to page 22 link to page 22
CRISPR-Cas
plants typically 19-22 nucleotides in length, is selected based on the target of interest. For optimal
guide selection and to reduce off-target potential bio-informatic tools are available (for example [29,
30]). The two components may be placed on the same or on two separate vectors [18, 20]. In case
genome modification or insertion of cis-, intra-, or transgenic sequences is the goal, additionally a
sequence acting as repair template is included.
Fig. 2.5 Minimal gene cassette requirements for CRISPR-Cas9 mediated genome editing. The coding
sequence giving rise to Cas9 is placed between a polymerase II promoter and terminator sequence,
to initiate and stop transcription, respectively. Cas9 is fused to a nuclear localization sequence (NLS)
to ensure nuclear localization. The sgRNA sequence is generally placed between a polymerase II
promoter and terminator sequence.
Production processes of genome edited plants using CRISPR-Cas9 may involve generation of intermediate
plants stably incorporating a gRNA-Cas9 transgene. In case of sexually reproducing crops, resulting genome
edited lines without the transgene but with the intended mutation are selected.
Furthermore, genome edited lines may be established using transient transformation procedures, i.e.
plants are transformed with a gRNA-Cas9 transgene, but it is not integrated into the genome. The
generated mutation, but not the gRNA-Cas9 transgene, is passed on to the next generation.
Finally, gRNA-Cas9 complexes may be delivered to the cells without the involvement of DNA, as pre-
assembled ribonucleoprotein complexes.
2.1.3 Techniques (SDN1,2,3)
There are different types of targeted genome modifications that can be achieved by using site
directed nucleases (SDN) including CRISPR-Cas9, by placing a targeted DSB(s) and, optionally, at the
same time providing a repair template
(Fig. 2.6): (1a) generating gene knock outs by inducing site
specific random mutations due to erroneous NHEJ repair, (1b) gene deletions by placing two DSBs
leading to the loss of the genomic region within, (2) gene modification by site specific nucleotide
sequence changes mediated by a repair template with homology and (3a, 3b) gene insertion by
providing repair or donor templates. The NTWG (active under the request of competent authorities
(CA) under Directive 2001/18/EC) subcategorized ZFN and related SDN techniques in genome editing
based on their outcomes into SDN1, SDN2 and SDN3 [31] which correspond to repair pathways 1a, 2
and 3a i
n Fig. 2.6, respectively.
The strategic outcomes are recapitulated for CRISPR-Cas9 below:
15
link to page 22 link to page 22 link to page 22
CRISPR-Cas
Fig. 2.6 CRISPR-Cas9 genome editing (after [14]). Targeted DSBs induced by CRISPR-Cas9 can either
lead to random mutations at the DSB site (1a) or, in case two DSB are induced, to deletion of the
genomic region within (1b), both mediated by NHEJ. In case a repair template with regions of
homology is provided together with the CRISPR-Cas9 module, pre-defined mutations (2) or precise
insertions of DNA sequences (3a) can be implemented at the DSB by HDR. Gene insertions can also
be generated by providing donor molecules without homology which are inserted at the DSB by NHEJ
(3b). DSB: double strand break; HDR: homology dependent repair; indel: insertion/deletion mutation;
NHEJ: non-homologous end joining. SDN1, 2, 3: categories of the technique according to the
definitions used in a regulatory context (site directed nuclease).
Technique SDN1: sgRNA-Cas9 activity module is delivered into cells and introduces a targeted DSB.
DSBs repaired by NHEJ may lead to site specific random mutations, i.e. insertions, deletions,
substitutions or a combination of these. These can be exploited in cases mutations lead to, for
example, gene knock-outs by frameshift mutations when targeted to coding regions. The DSB can
also be targeted to non-coding regions, for example to impair or delete regulatory elements, thereby
inducing a change in gene expression
(Fig. 2.6 (1a)). In extension to the original definitions by the
NTWG, two DSBs can be placed by delivery of two Cas9-gRNA modules targeting different locations,
resulting in deletion of the region in-between
(Fig. 2.6 (1b)). Finally, placing of two DSBs has the
potential to induce chromosomal re-arrangements (inversion, duplication or translocation events)
which may be exploited for genome editing [32].
Technique SDN2: sgRNA-Cas9 activity together with a DNA repair template is delivered into cells. The
repair template is homologous to the targeted region with exception of site specific nucleotide
sequence changes (single nucleotide changes or smal insertions/deletions). sgRNA-Cas9 activity
induces a targeted DSB. In the course of HDR, the repair template may be used and the desired site
specific nucleotide sequence changes are implemented at the genomic locus
(Fig. 2.6(2)).
16
link to page 22 link to page 22 link to page 22
CRISPR-Cas
Technique SDN3: sgRNA-Cas9 activity together with a repair template harbouring a cis-, intra-, or
transgene is delivered into cells. The repair template consists of a DNA stretch intended for insertion
and is flanked on both sides by sequences homologous to the target region. sgRNA-Cas9 activity
induces a targeted DSB. In the course of HDR, the DNA to be inserted is precisely inserted at the
target site
(Fig. 2.6 (3a)). SDN3 thus enables insertion of cis-, intra- or transgenes at specific loci.
Furthermore, targeted gene insertion can also be achieved by using the repair pathway of NHEJ
(Fig.
2.6 (3b)). In this case, the donor DNA harbouring the cis-, intra-, or transgene to be inserted does not
need to be flanked by regions of homology to the target locus.
2.2 Application in plant breeding
In 2012 it was realized that CRISPR-Cas9 provides a valuable addition to already established systems
for genome editing [13]. In the meantime further applications other than genome editing and of
interest to plant breeding have been proposed. In these potential applications CRISPR-Cas9 is used as
a transgenic locus to confer protection of plant virus infection [33-35]. Also, there is ongoing
development of CRISPR-Cas modules for endogenous gene expression regulation [36].
2.2.1 Potential applications of SDN1
The SDN1 technique may seem of restricted use in plant breeding in comparison to transgene
technology or mutation breeding since traits can mainly be altered by elimination of gene/promoter
function. However, metabolic and developmental pathways function as networks and so elimination
of gene function can be used to affect traits in a variety of modes, depending on the nature of the
pathway and the targeted step, the eliminated gene function (positive/negative regulator) and the
overall genetic architecture of the trait (redundant gene function). The SDN1 technique shares trait
modification by elimination of gene function with the RNAi technology. Traits that have been
engineered before using RNAi technology might now, where sensible, be implemented using SDN1
technology, and further traits beyond these wil be modified using SDN1 technology.
2.2.1.1.1 Elimination of unwanted compounds
An apparent SDN1 application is elimination of unwanted compounds. Anti-nutritional compounds
can be eliminated or lowered by knocking-out genes coding for enzymes in biosynthetic pathways,
for example leading to phytate in maize [37], or to linamarin, a toxic compound in the staple food
cassava [38]. In order to engineer food grade oil in rapeseed varieties and in other
Brassica species,
including under-utilised species like
Camelina sativa [39] or in rapid domestication of wild species like
Thlaspi arvense [40], low erucic acid and glucosinolate content are breeding goals (00 varieties).
Potential SDN1 targets for that for example are
FATTY ACID ELONGASE 1 (
FAE1) and the transcription
factor
HIGH ALIPHATIC GLUCOSINOLATE 1 (
HAG1), respectively [41, 42]. Low erucic acid and
17
CRISPR-Cas
glucosinolate content are quality parameters for food and feed use in Brassicas: low erucic acid
varieties in general are used for production of edible oils, and low glucosinolate content allows use of
the seed meal for feed purposes.
Tissue specificity, conferred in RNAi technology by promoters, can be achieved by SDN1 technology
through knowledge on tissue specific gene function. In plants, paralogs, gene family members with in
some cases exchangeable gene function, are often expressed in a tissue specific manner. The
rapeseed (
Brassica napus) genome encodes three functional paralogs of
FATTY ACID DESATURASE 2,
of which
FAD2-4 is expressed in root and seed tissue only, while
FAD2-1 and
FAD2-2 are expressed
ubiquitously [43]. The protein derived from
FAD2 catalyses monounsaturated oleic acids into
polyunsaturated fatty acids (PUFA) and for some industrial applications low PUFA content is desirable
(f.e. it has higher thermal stability and a longer shelf life). Knocking out specifically
FAD2-4 might be
an approach to change the fatty acid profile of rapeseed in specific organs of interest only,
maintaining fatty acid metabolism in the remaining tissues. In other cases, tissue specificity may be
implemented by affecting transport mechanisms: targeting homologs of
ARABIDOPSIS THALIANA
GLUCOSINOLATE TRANSPORTERS 1 and
2 (
GTR1, 2) in agronomically important
Brassicas, may be
used to lower glucosinolate content specifically in seeds while maintaining glucosinolate production
and therefore biological function (protection against herbivory, for example) in source tissues, since
GTR1 and
GTR2 are required for glucosinolate transport into seeds [44]. Another problem, reducing
or eliminating allergenic epitopes causing celiac disease might be chal enging in the (near) future,
since α-gliadin alone is encoded numerous times (at least 40 times without taking into account
pseudogenes) in the wheat genome [45] and at the same time gluten is an important quality
parameter of wheat. However, eliminating allergenic epitopes of less-abundant proteins eliciting
strong response is a feasible breeding goal with SDN1 technology. In soybean, the p34 protein shows
low abundancy, but is one of the major soybean al ergens [46]. p34 is a member of the papain
superfamily of Cys proteases, with as yet no reported enzymatic activity [46], and a BLAST search
against the
Glycine max genome detects few paralogous loci (3-4 loci; assembly V1.0, at
EsemblPlants platform). In an RNAi approach p34 downregulated soybean lines were viable and
similar in growth behavior in comparison to wild-type plants [46].
2.2.1.1.2 Increasing production of desired compounds
Besides elimination of unwanted products, knock-out of genes using SDN1 technology can also be
used to change plant metabolism to enhance production of a desired metabolic product or trait. To
revisit fatty acid metabolism in
Brassicas, deletion of
FAE1 to eliminate erucic acid production at the
same time leads to elevated levels of monounsaturated oleic acid content [42]. There is another
breeding strategy to increase overall oil content in oil crops: reduction of fruit components (pericarp,
testa) not containing oil, like for example hardened ovary tissue protecting the seed in achene fruits
18
link to page 71
CRISPR-Cas
of sunflowers, or for example thick testa tissue in non-oil pumpkins; but also linseed, poppy and
rapeseed cultivars with reduced sclerenchymatic tissue exist [47]. The thin testa of Styrian oil
pumpkin cultivars is known to be based on a recessive mutation in a major gene [47, 48], and if once
mapped and based on a loss of function mutation, the locus may be an SDN1 target to generate a
high oil content trait in other cultivars while at the same time maintaining the cultivars favourable
genetic background (SDN2 technique might be used in case the recessive allele is a functionally
recessive al ele). Knock-out of inhibitors of pathways represents an additional strategy to enhance
production of traits of interest using SDN1 technology. For example, non-glandular trichome
production in
Brassicas is governed by a suite of activators (certain members of
WD40/bHLH/R2R3MYB genes forming a protein complex) and inhibitors (R3MYB) [49]. Knock-out of
the latter increases trichome production [50]. Glandular and non-glandular trichome density has
been shown to be positively correlated with protection from insect herbivory [51-53]. The genetic
basis of trichome production is at least partially conserved across plant genera [54-57] and is starting
to be discovered for glandular trichomes in
Cucumis sativus [58, 59]. Trichomes, in particular
glandular trichomes, are also the natural production site of a suite of specialised metabolites across
plant species with commercial value (pharmaceuticals (artemisinin), fragrances/flavour (Lamiaceae
plant family) or natural pesticides (involvement in pyrethrin biosynthesis) [60-62]). Therefore, a
strategy to increase production of valuable trichome derived metabolites might be to increase
trichome production by SDN1 targeting of trichome inhibitors, alternatively, SDN1 targeting of
trichome activators might be used to generate favourable glabrous vegetable varieties [58].
WD40/bHLH/R2R3MYB complexes together with R3MYB inhibitors are also involved in regulation of
the flavonoid biosynthesis pathway [55, 63-66] and production might be enhanced via targeting of
the pathway specific R3MYB inhibitor by SDN1.
2.2.1.1.3 Engineering pathogen resistance by targeting recessive resistance genes
Genome editing may be used to target so cal ed susceptibility (S) genes (or recessive resistance
genes) [67] to establish lines with biotic stress tolerance (for brief introduction to S genes refer to
chapter
4.2.1).
A specific example are the effector targets in rice
Xa13 and
Xa25/
OsSWEET13 of
Xanthomonas
oryzae pv.
oryzae (Xoo), which causes bacterial blight.
Xoo encodes effector genes (transcription
activator like effectors (TAL effectors)) that bind to effector binding sites (EBE) in promoter regions
and thereby upregulate host target genes in order to promote virulence [68]. Using CRISPR-Cas9 to
establish a knock out line for Xa25/OsSWEET13 in a japonica rice line otherwise susceptible to a Xoo
strain transformed with a TAL effector designed to target Xa25/OsSWEET13, it could be shown that
disease susceptibility was lost [68]; plants were reported to have no obvious detectable phenotype in
this study. However, recessive resistance genes are endogenous plant genes with biological functions
19
CRISPR-Cas
in plants, for some of which pleiotropic effects have been reported [69]. One strategy to minimize
the effects of engineering pathogen resistance of S genes, that are upregulated upon TAL effector
EBE binding, has been shown using another site directed nuclease system, ironically TALEN. In this
study, OsSWEET14 was not targeted by TALEN genome editing in the protein coding region to
establish a knock out line, but in the EBE site of the promoter region in order to abolish Xoo TAL
effector binding, and at the same time retain other OsSWEET14 functions [70]. It could be shown
that genome edited rice lines with induced smal deletions of 4 or 9 bp in the EBE did not induce
OsSWEET14 expression after inoculation with an Xoo strain carrying the avrXa7 TAL effector protein,
and displayed a resistance phenotype [70]. They mimic naturally occurring recessive resistance
alleles of
Xa13 and
Xa25/
OsSWEET13, since these are also not null alleles, but possess
polymorphisms in the EBE sequence of the promoter [71]. This study demonstrates that by
introducing small, targeted mutations using genome editing valuable traits of use in plant breeding
may be engineered.
SDN1 (and SDN2) genome editing techniques provide refined means to plant breeding and complement
transgenic technology, traditional and mutation breeding. Conventional mutation breeding programmes
offer the discovery of novel, artificially induced, trait variation; further, by TILLING (Targeting Induced Local
Lesions in Genomes), mutant populations can be screened for desired variation at a locus of interest. The
potential of SDN1 and SDN2 techniques in genome editing is linked to already present and increasing
knowledge derived from basic and applied research on molecular variation underlying phenotypic trait
expression as well as on gene function and metabolic pathways in general. By using SDN1 and SDN2
techniques, the breeder directly and specifically works with the understanding of molecular variation that
has been discovered to underlie phenotypes of agronomic interest. The examples given above provide an
overview on the potential use of SDN1 technology to alter traits of interest to plant breeding: removal of
unwanted compounds (phytate, glucosinolates), increasing desired compounds (oleic acid, secondary
metabolites) or engineering of pathogen resistance by altering recessive resistance genes.
Among other techniques, sequencing technology is generating an enormous amount of information
on genomic variation within and between species (150 Tomato Genome ReSequencing Project [72],
3000 Rice Genomes Project [73], 44 sorghum line genomes [74], 302 soybean accessions [75], 115
cucumber lines [76]), which can be mined for meaningful variation of trait expression in phenotypic
screens (for developments in high throughput phenotyping see for example [77, 78]; for examples of
genome-wide association studies refer to [79-83]).
20
CRISPR-Cas
2.2.2 Potential applications of SDN2
The SDN2 technique can be used similar to SDN1 to translate knowledge on meaningful molecular
trait variation into plant breeding programmes. The SDN2 technique, in addition to induce loss of
function mutations, is of particular interest to transfer favourable functional molecular variation
between cultivars or from closely related (wild) species (allele transfer) but also introducing
favourable amino-acid changes deduced from methodological genetic screens into elite cultivars. A
recent example is the generation of herbicide resistant maize lines by introduction of specific single
nucleotide substitutions in the gene ACETOLACTATE SYNTHASE 2 (ALS2) [84].
2.2.3 Potential applications of SDN3
The SDN3 technique can be used similar to conventional cis-, intra-, or transgenesis technology to
insert cis-, intra-, or transgenes into plant genomes, with the difference that the location of insertion
can be determined
a priori. To date, insertion of cis-, intra- or transgenes is largely based on insertion
at random genomic loci. Several independent lines need to be screened to select suitable candidate
lines which do not show undesired phenotypes because of compromised target sites, for example by
gene disruption, and at the same time express the inserted gene in an adequate manner. With the
SDN3 technique, insertion can be targeted at a defined locus and possibly take advantage of
knowledge about regions of permissive gene expression. With this technique also gene stacking is
possible, i.e. introduction of several genes in close proximity. This facilitates breeding programmes in
that favourable new traits are not separated in successive breeding cycles and in turn can be easily
introduced into further varieties/germplasm segregating as a single-locus trait.
The potential applications of cis- and intragenesis have been described in the study by AGES [3].
2.2.4 Applications other than genome editing
The CRISPR-Cas module provides a programmable tool to target a protein component to defined
dsDNA (type II) or ssRNA (type II ) sequences. The CRISPR-Cas module thus provides the potential to
be used for applications other than genome editing. Both groups of potential applications below
involve activity of CRISPR-Cas as a transgenic locus in plant lines. These applications have been
reported recently and it remains to be seen whether they develop the potential for use in plant
breeding.
CRISPR-Cas9 as a tool for conferring virus resistance in plants. Recently, three independent
publications have shown in proof of principle experiments that CRISPR-Cas9 can confer protection
against different types of geminiviruses in
N. tabacum and
A. thaliana [33-35]. The use of CRISPR-
Cas9 for generation of geminivirus resistant crops offers advantages over other strategies (multiplex
21
CRISPR-Cas
targeting, fast response in targeting of newly emerging viral strains), but there are currently still
questions to be addressed (off-targeting, selection pressure on virus populations) for its potential
deployment as a resistance trait in plant breeding [85].
CRISPR-Cas as a tool for targeted gene expression regulation. A nuclease de-activated “dead” Cas9
(dCas9) alone or fused to effector domains is guided to loci of interest to interfere with (CRISPRi) or
activate (CRISPRa) gene expression (reviewed in [36]). The mode of regulating expression is dictated
by the specific dCas9 fusion protein and includes steric hindrance of transcription, mediating
transcription via activation domains or epigenetic modification. In a proof of principle experiment,
dCas9 guided to a reporter locus decreased gene expression in bacteria and human cel s [86], and in
human cells the repressive effect was enhanced by fusion of dCas9 to the chromatin modifier domain
KRAB (Krüppel-associated box) [87]. The KRAB domain guided by dCas9 to HS2, a distal enhancer
element of globin genes, efficiently induced histone modifications indicative of closed
heterochromatin and at the same time reduced globin gene expression [88]. The feasibility of CRISPRi
and CRISPRa, has been demonstrated in several systems [14] and recently also in
Nicotiana
benthamiana [89]. CRISPR-Cas subtypes (for example type II -B) may also be employed for targeted
RNA interference in the future [90].
These methods offer interesting alternatives to RNAi based methods in gene expression regulation,
however, it remains to be seen whether they will also be applied for plant trait development in plant
breeding. They are not further considered in the remaining chapters.
CRISPR-Cas9 has become widely applied also because it has uncoupled the technique of genome editing
from know-how intensive protein engineering, as is required in TALEN- or ZFN-based genome editing.
Because of that, although first applied in plants in 2013, already a large number of genome edited crop
plants have been published in the scientific literature.
CRISPR-Cas9 may be applied in genome editing to introduce targeted mutations and by that engineer, for
example, plants with reduced unwanted compounds, increased desired compounds or disease resistance.
Traits so conferred via SDN1 technology to date are explored with great emphasis also in crop plants with
prospect of applications. Genome editing resulting in accurate site directed insertion of transgenes (SDN3)
is promising great strides also in basic plant research, yet still lacks ease in successful implementation. In
the future, establishment of commercial cis-, intra-, and transgenic plants may benefit from developments
in SDN3 technology.
Other potential applications use CRISPR-Cas9 as a trait conferring virus resistance or as a regulator (positive
and negative) of gene expression. These applications would entail the insertion of foreign DNA and
therefore generate transgenic plants falling under the current EU GMO legislation.
22
link to page 15
CRISPR-Cas
2.3 State of research and development in plants
In 2013, several research groups reported the first proof-of-principle experiments of CRISPR-Cas9
mediated genome editing in plants [91-100]. Primarily carried out in
O. sativa, N. tabacum and
A.
thaliana, co-expression of recombinant Cas9 and gRNAs reproducibly induced targeted indels in
endogenous genes in cultured cel s and
in planta. Using donor templates homologous to endogenous
genes or specially designed reporter constructs it was shown that targeted DSBs carried out by
CRISPR-Cas9 can result in HR-mediated repair of genomic regions. Since CRISPR-Cas9 can be easily
reprogrammed via the spacer sequence of the gRNA, simultaneous delivery of multiple sgRNAs
targeting different loci demonstrated the feasibility of multiplex genome editing as well as of
deletion of intervening sequences in these first reports. The number of scientific publications with
experimental CRISPR-Cas9 data in the plant field is increasing
(Fig. 2.1).
2.3.1 Transferability of the system to plant species
CRISPR-Cas9 technology has since been shown to be transferrable to various crop plants (mostly
using SDN1 technique), for example to soybean (
Glycine max) [101], wheat (
Triticum aestivum) [98],
maize (
Zea mays) [84], barley (
Hordeum vulgare) [102], potato [103] and tomato [104] (
Solanum
tuberosum and
S. lycopersicum), but also for example to tree species, like
Populus [105] and
Citrus [106]. By now, it has been established that genome edited sites are stably transmitted to progeny
independent of CRISPR-Cas9 presence. In tomato, in a cross of a wild type plant with a bi-allelic
genome edited individual, progeny lacking the CRISPR-Cas9 transgene was heterozygous with either
one of the two edited alleles in combination with the wild-type al ele [104]. Independent
transmission has also been analysed and shown for example in
Arabidopsis, rice, barley and
Brassica
oleracea [107-110] [102].
23
CRISPR-Cas
2.3.2 Techniques (SDN1, 2, 3)
SDN1 and, in extension, multiplexing for simultaneous editing of genes or deletion of genes are the
most frequently reported genome editing techniques in the scientific literature, since it involves
delivery of the nuclease component only. In wheat, for example, plant lines carrying mutations in
MILDEW RESISTANCE LOCUS (
MLO)
A1 have been generated [111], and in rice, knock-out of
SWEET13 has proven its function in bacterial blight susceptibility [68]. Intended targeted deletions
reported using two DSBs range from small deletions of for example ~50 bp in tomato [104] to ~245
kbp in rice. The latter resulted in deletion of a diterpenoid synthetic gene cluster of ten loci [110],
exemplifying the potential to eliminate large genomic regions. Multiplexing ability using CRISPR-Cas9
has been shown for example in rice plants targeting up to 7 [112] and 8 [113] sites simultaneously
with different gRNAs, the latter using a specially designed gRNA processing platform, or in
Arabidopsis using a gRNA with perfect complementarity to two loci [108]. Endo
et al., exploited the
off-target activity of a gRNA targeted at CYCLIN DEPENDANT KINASE 2 (CDKB2) to generated rice lines
edited at 2 further gene family members, CDKA2 and CDKB1 [114].
There are fewer reports on the SDN2 and SDN3 techniques. They, together with delivery of CRISPR-
Cas9, provide templates for HR-mediated repair to either modify a locus (SDN2) or insert a cis-, intra-
or transgene (SDN3). That both techniques are feasible using CRISPR-Cas9 technology in plants has
been shown for example in maize, soybean and rice [84, 101, 115]. In maize, endogenous
ACETOLACTATE SYNTHASE 2 (ALS2) has been converted into a sulfonylurea herbicide resistant allele
by site specific modification of a proline to a serine (P165S) via SDN2; additionally, a
PHOSPHINOTHRICIN ACETYLTRANSFERASE (PAT) gene driven by a constitutive promoter was site
specifically inserted via SDN3. Genome edited plants transmitted the modifications into the
subsequent generations. The modified ALS2 (P165S) gene conferred herbicide resistance in two
tested generations [84]. In soybean, ALS1 was modified similarly (P178S) to confer herbicide
tolerance via SDN2 and a hygromycin phosphotransferase (HPT) gene linked to an endogenous
soybean promoter was targeted for insertion to a specific locus [101]. The SDN3 technique has
further been demonstrated in
Arabidopsis [116], and in tomato by targeting a strong promoter
(CaMV 35S) in front of an anthocyanin biosynthesis gene resulting in accumulation of pigments [22].
In the study in
Arabidopsis, one targeted insertion event has been reported with perfect repair as
intended, in the study in tomato, in addition to a perfectly repaired insertion event, an event with
nucleotide substitutions was recovered.
2.3.3 Delivery methods
The main delivery method of CRISPR-Cas9 reported for production of genome edited plants involves
transformation of a CRISPR-Cas9 gene cassette integrated on vector systems with selectable marker
24
CRISPR-Cas
genes into cultured plant cells. The presence of marker genes allows regeneration of plants stably
transformed with the CRISPR-Cas9 construct and subsequent screening of a reduced number of
plants for genome edited individuals. Since the presence of the CRISPR-Cas9 transgene is not
necessary and may lead to off-target effects upon retention in plant lines, it can be segregated from
the intended genome modification in subsequent generations in sexual y propagating crop species
(for example see [107-110] [102]).
Two delivery methods of CRISPR-Cas9 into plant cells independent of DNA transfer were reported.
Analogously as shown with TALEN and meganucleases [117], pre-assembled CRISPR-Cas9 ribonucleo-
protein particles were delivered directly into plant cells [27]. In proof-of-principle experiments in
protoplasts derived from
A. thaliana, tobacco, lettuce and rice, genome editing was detected by this
delivery method. Regenerated lettuce individuals transmitted the modified al ele into the next
generation [27]. A bottleneck for general application of this strategy is the ability to regenerate
plants from protoplasts which is not a well-established procedure in different plant species. An
alternative reported strategy uses delivery by an RNA virus [24, 25].
Tobacco rattle virus (TRV) has a
bipartite positive strand RNA (TRV1 and TRV2) genome of which TRV2 can be modified to harbour
foreign genes, which is commonly exploited in different viral systems in virus induced gene silencing
(VIGS). A gRNA driven by a pea early browning virus (PEBV) promoter and targeting PDS was cloned
into TRV2 and was agro-inoculated together with TRV1 into
N. benthamiana transgenic lines stably
expressing Cas9. Cas9 expression from the
N. benthamiana genome was necessary because of
limited capacity of the viral genome to harbour foreign genes and retain functionality. Gene editing
at PDS took place in agro-infiltrated and in systemic
N. benthamiana leaves [24, 25] and the edited
PDS al ele was transmitted into the next generation [25]. Limiting factors for a broader application
using DNA free delivery by RNA viruses are the small capacity of the viral genome, the varying host
range of viruses and systems to obtain virus-free genome edited progenitor plants (REF).
2.3.4 Types of mutations generated by SDN1 technique
Datasets describing the type of mutations generated by SDN1 technique are reported mainly for
Arabidopsis, rice and soybean [107, 112, 118-125]. The mutations arise during repair by the
endogenous DNA repair pathway of DSBs, mainly NHEJ in somatic cel s [21].
In
Arabidopsis and rice, based on to date available data, the most frequently detected mutations are
insertions of a single adenosine or thymidine nucleotide, followed by small deletions of
predominantly one nucleotide and deletions of <10 nucleotides [107, 112, 119, 120, 123-125]. Other
detected mutations are nucleotide replacements and insertion of >1 nucleotides, but to a lesser
extent. Based on the data available at present from
Arabidopsis and rice, the mutation spectrum may
be generalised over experimental systems, mutations detected in protoplasted cells, transgenic lines
25
link to page 18 link to page 18 link to page 110
CRISPR-Cas
generated by floral dip transformation (
Arabidopsis) or somatic embryogenesis after agro-inoculation
(rice). In soybean, the most frequently detected mutations were deletions <10 nucleotides [118, 121,
122]. There is the indication that dependent on the sgRNA or the targeted locus the mutation
spectrum may differ in some instances; in the study of Jacobs
et al., one sgRNA induced
predominantly single nucleotide insertions, independently of the experimental system (soybean
hairy root and somatic embryogenesis) [121]. Similar observations were made in rice [124, 125]. The
location of the generated mutations predominantly occur starting three nucleotides off the PAM in
the proto-spacer sequence (for example [118-121]).
2.3.5 Off-target activity
Recognition of the target site, the so called protospacer, by a CRISPR-Cas9 complex is guided by two
different signatures, the presence of a protospacer adjacent motif (PAM) and the complementarity
of the spacer sequence in the sgRNA to the protospacer sequence [13, 126]
(Fig. 2.3). The PAM is
present at the genomic target site directly 3´ to the protospacer and is recognized by the PAM
interacting domain of Cas9 protein.
Streptococcus pyogenes Cas9 (SpCas9) recognises the PAM
sequence 5´-NGG-3´ and, with less efficiency, 5´-NAG-3´ [126]. A systematic analysis of CRISPR-Cas9
target specificity found that the 8-12 nucleotides proximal to the PAM (called seed region,
(Fig. 2.3))
are on average less tolerant to mismatches than the distal region [126]. The efficiency of perfectly
matched CRISPR-Cas9 modules in DSB induction was analysed over three datasets in a study by Xu
et
al., [127]. They find and confirm [128-130] that DSB induction efficiency is dependent on several
features, for example nucleotide composition in the spacer region (where some nucleotide positions
influence Cas9 gRNA loading) or the influence of nucleotide positions 3´downstream of the PAM (i.e.
outside of the protospacer region). These, as well as for example structural features of the gRNA
backbone [131] influence CRISPR-Cas9 efficiency and thus also contribute to off-target activity. There
is ongoing research into specificity and efficiency which will be implemented in genome editing
systems in the future, particularly in metazoan systems, since specificity and efficiency are highly
critical parameters for potential therapeutic applications of CRISPR-Cas9.
Upon application of CRISPR-Cas9 in plant genome editing, characterization of off-target activity was
also of interest in plant species.
Table 7.1 (Appendix) lists studies which report analyses of off-target
activity in plant cells. A set of 15 randomly mutant gRNAs were tested against a target locus in wheat
suspension culture cel s [98]. Similarly to the above studies in bacterial and human cells, mismatches
at the distal region (non-seed region) were rather tolerated than mismatches proximal to the PAM,
which often abolished DSB formation [98]. Two studies, in
Arabidopsis and rice, report whole
genome sequencing (WGS) data of CRISPR-Cas9 genome edited plant lines in order to survey
genome-wide possible off-target effects [107, 132]. In rice, Nipponbare plant lines each transgenic
26
CRISPR-Cas
with one of 6 different gRNAs did not reveal significantly higher SNP and indels than wild type
controls when compared to the rice reference sequence; in a Kasalath background a comparison was
hampered by the large difference of sequencing depth between the wildtype and transgenic lines
[132]. In
Arabidopsis, comparison of 3 WGS datasets of CRISPR-Cas9 transgenic plant lines
(harbouring the same gRNA) did not show an increased SNP or indel number in comparison to wild
type controls when mapped to the Col-0 reference genome [107]; Off-target activity in the remaining
studies was analyzed to different degrees and with different methods. While in some studies off-
target sequence searches were carried out systematically by BLASTN searches, in some cases
supported by available software programmes [29, 30], which search for and (CRISPR-P) rank potential
off-target sites based on an experimentally derived score, other studies chose loci based on prior
knowledge of sequence homology. Detection of off-target activity was either based on sequencing
methods, restriction enzyme/PCR methods (PCR/RE) [96, 100] or enzyme mismatch cleavage
methods [133]. In a study in soybean, 10 potential off-target sites, with varying degrees of
mismatches distributed over the protospacer region, were tested and off-target effects were not
detected [134]. While in this dataset a target with only three mismatches in the distal region was not
targeted, another gRNA exhibited off-target activity at a locus with two mismatches in the proximal
region in the same study [134]. Similar results were obtained in rice: testing 3 gRNAs on altogether
13 potential off-target sites identified one off-target locus harbouring 1 mismatch in the distal region
[132] and testing 4 different gRNAs on the highest ranked potential off-target sites by CRISPR-P, off-
target activity was detected at one site with one mismatch in the distal region [109]. Other potential
off-target sites in these two studies harboured 3-7 and 2-4 mismatches, respectively, with varying
distribution over seed and non-seed region sites [109, 132].
Based on the published data, specificity of CRISPR-Cas9 in plant cells seems to be governed by the
same parameters as in other eukaryotic systems. While the majority of detected off-targets
harboured mismatches in the distal region, there were also exceptions to this rule (see for example
[134]).
In summary, CRISPR-Cas9 mediated genome editing has been shown to be transferable to diverse plant
species. Initial proof-of-principle experiments established that all techniques (SDN1, 2, 3) are principally
feasible, the highest number of publications to date report application and development of SDN1.
Increasingly, there are research publications using CRISPR-Cas9 as an alternative to conventional methods
in reverse genetics to analyse gene function [68, 135-137], indicating general acceptance as a validated and
efficient method in plant science. Ongoing research and development is focused on establishing efficient
genome editing platforms and vector systems for diverse species, and on the development of delivery
modes (including those without DNA transfer). The ability to specifically modify target sites offers an
alternative and site-directed mode to create variability for plant breeding. Prior knowledge of a gene
27
CRISPR-Cas
function and its physiological or phenotypic effect on plant traits can thus be implemented in a specific and
efficient manner into plant breeding programmes.
2.3.6 Limiting off-target effects
There are several strategies to limit off-target effects. Hsu
et al., postulated a set of rules to guide
gRNA selection and there are several software applications to automate gRNA selection (for example
[29, 30]).
Experimental strategies include the application of paired nickases or RNA-guided FokI nucleases
(reviewed in [138]). Paired nickases are Cas9 proteins with introduced point mutations destroying
either one of their two endonuclease domains. The resulting proteins introduce single stranded DNA
breaks (nicks), and targeting two complementing paired nickases properly spaced to the same locus,
a DSB is generated. At the same time, specificity is increased since two spacer sequences are needed
for induction of a DSBs. Paired nickases have been used in proof-of-principle experiments in plants
[119]. RNA-guided FokI nucleases confer specificity by the same principle and are based on gene
fusions between dCas9 and a dimerization dependent
Flavobacterium okeanokoites (Fok1) nuclease
(reviewed in [138]).
Recently, mutant Cas9 proteins, [139] and SpCas9-HF1 [140] have been shown to confer higher
specificity to the CRISPR-Cas9 complex in human cel s by weakening non-specific interactions of Cas9
with its target; in the case of eSpCas9 interaction with the non-complementary target strand, in the
case of SpCas9-HF1 four aa substitutions were introduced to weaken non-specific interactions of
Cas9 with the target strand. Since specificity is of high importance for therapeutical applications of
CRISPR-Cas, further strategies and/or mutant versions of CRISPR-Cas may be developed which may
also be utilized then in plant applications.
2.4 Intended and unintended effects of CRISPR-Cas9 in genome editing
The intended effect using CRISPR-Cas9 in genome editing is the targeted site specific modification of
a target locus and thereby changing expression of trait(s) modulated by that locus. Intended genetic
modifications have been categorized by the NTWG (national experts nominated by the Competent
Authorities of EU Member States) as site specific random mutations (SDN1), site specific
modifications (SDN2), and site specific insertion of cis-, intra-, and transgenes (SDN3) [2].
Furthermore, in multiplexing, targeting of several loci or deletion of regions in between may be the
intended goal.
A potential unintended effect due to application of the CRISPR-Cas9 technology in genome editing is
off-target activity by placing of DSBs at loci with imperfect complementarity to the spacer sequence.
28
link to page 35
CRISPR-Cas
This might lead, depending on the SDN technique, to either (i) induction of random mutations at off-
target loci, to (I ) deletion of genomic fragments, (iii) integration of cis-, intra-, or transgenes at
unintended loci or (iv) a combination of those.
Potential unintended effects by means of using transgenic CRISPR-Cas9 intermediate lines may be (i)
retention of the transgene in resulting organisms and (i ) generation of background mutations due to
the performed transformation process, which are passed on to resulting organisms. An unintended
effect due to the use of viral vector systems is viral contamination of progeny.
2.5 Safety considerations
2.5.1 SDN1 technique in genome modification of plants
The SDN1 technique targets specific loci to introduce mutations of
a priori unknown sequence
changes. Intended changes mostly are loss of function mutations of genes or regulatory elements,
since these are most likely generated using this technique. In general, the SDN1 technique introduces
small insertions, small deletions or nucleotide replacement mutations at a site or sites near the PAM
in the protospacer sequence. However, also larger deletions or insertions may arise. When targeting
two CRISPR-Cas9 modules on the same chromosome, it is also possible to generate deletions of the
genomic region in between the two DSBs. The specificity and therefore the amount of DSBs induced
in the genome is determined by the spacer region.
Provided that the resulting plants do not carry a CRISPR-Cas9 module stably integrated in the
established plant line, the SDN1 technique can therefore be compared to conventional physical and
chemical mutagenesis techniques based on intended and unintended changes.
Genome edited, transgene-free resulting plant lines may be established, for example, (i) by selecting
nul -segregants of transgenic plants, (ii) in cases where a CRISPR-Cas9 DNA module had been
transiently transformed and (iii) in cases where ribo-nucleoprotein complexes have been directly
introduced.
2.5.1.1 Comparison of CRISPR-Cas9 and conventional mutagenesis techniques in relation
to mutational load and type of modifications
Physical (for example gamma ray, X-ray) and chemical (for example ethyl methanesulfonate (EMS))
mutagenesis is used to induce variation in plants to generate mutants for conventional plant
breeding. There are 3,220 mutant cultivars, in over 210 species [141], collected in the worldwide
Mutant Variety Database (MVD, FAO/IAEA)
2 which have been officially and/or commercially
2 https://mvd.iaea.org/
29
link to page 43 link to page 44 link to page 44 link to page 43 link to page 44 link to page 32 link to page 32 link to page 45
CRISPR-Cas
released. Tables 2.1 and 2.2 list studies reporting on induced genetic variation after chemical (EMS
and NaN3/MNU) and physical (gamma irradiation) mutagenesis. EMS is an alkylating agent
(predominantly of guanine) resulting in SNPs by changing G/C nucleotides into A/T nucleotides [142]
and gamma irradiation is suggested to induce DSBs resulting in diverse mutation categories [143,
144]. The amount of induced genomewide mutational events varies, among other things, with dose
and concentration of physical and chemical mutagens, respectively, but is also dependent for
example on the treated tissue. Common mutation densities/effects are given in
Table 2.1 and
Table
2.2 for reported chemical and physical mutagenesis experiments. In TILLING datasets typical
mutation densities are between 1 mutation /100 – 500 kbp (higher densities are typically present in
polyploid species since they are able to buffer deleterious mutations), but may also lie outside these
ranges depending on the TILLING population (see for example [142] for an overview). These mutation
densities translate into several hundreds of genomewide mutations per individual in the “smaller”
genomes of soybean and rice
(Table 2.1), and in around 340,000 mutations per individual in wheat.
Reported effects of gamma irradiation in wheat leads to estimates of around 82-110 gene deletions
per individual; in a study in rice, with a high irradiation dosis, it was estimated that 9% of the genome
was altered
(Table 2.2). For breeding purposes, there is a trade-off in mutational density, since on the
one hand the lower, the larger the population to be screened for desired genotpyic and/or
phenotypic mutants, but on the other hand a large mutational load potential y affecting many other
loci is undesired; depending on the species and propagation system, mutagenised individuals are
either directly or indirectly (as part of breeding programmes) used for establishing commercial
cultivars [145].
Whole genome sequencing of genome edited rice and
A. thaliana lines did not suggest a
genomewide elevated mutational increase when compared to control plants in the datasets of these
two studies (see chap
ter 2.3.5; [107, 132]). In contrast to chemical and physical mutagenesis, CRISPR-
Cas9 does not randomly (genomewide) induce mutagenesis events, but is restricted to the target loci
and to potential off-target loci with a certain amount of sequence complementarity (see chapter
2.3.5; see selected examples of studies reporting off-target effects for soybean, rice and barley in
Table 2.3). This is reproduced
in planta, for example in the study of Endo
et al., 2015 off-target
effects were detected at two loci ranked as most likely candidates by the software CRISPR-P,
however, no mutations were detected at loci ranked 3rd, 5th, 9th and 10th likely to be targeted
[114]. Zhang
et al., 2014 report similar results: while for two gRNAs off-target effects could not be
detected at 5 and 3 candidate loci, one gRNA lead to off-target effects in 1 out of 5 candidate off-
target loci; the effected locus showed 1 mismatch in the non-seed region in comparison to the
intended target, while the remaining potential off-target loci differed at 4, 6 or 7 positions and were
not targeted in this experiment [132].
30
CRISPR-Cas
The type of mutations generated by application of CRISPR-Cas9 based genome editing have been
summarised in Chapter 2.3.4: small insertions (<10 nucleotides), smal deletions and nucleotide
replacements have predominantly been detected at sites targeted by a given gRNA. Depending on
the particular mutagenic agent used, conventional mutagenesis generates for example
predominantly substitutions in the case of chemical mutagenesis using EMS (Table 2.1), while for
gamma irradiation substitutions, indels and copy number variations were reported for example in
rice (Table 2.2). However, also in the case of EMS mutagenesis the isolated mutation used for
breeding may be based, for example, on a deletion. Natural variation, including natural variation
found in domesticated species, is based on the same mutation categories: for example, in a study
resequencing landraces, wild progenitors and improved imbreds of
Sorghum bicolor [74], nucleotide
replacements, indels, copy number variations and larger deletions leading in some cases to gene loss
can be detected.
In comparison to conventional mutation breeding techniques, the CRISPR-Cas9 SDN1 technique induces
specific mutations at intended loci and potentially a smaller number of further off-target loci that can be
predicted to a certain extent. This also reflects the difference in intended use of these techniques in
breeding applications. Thus, the (random) unintended mutational load of CRISPR-Cas9 genome edited
plants is much smaller in comparison to conventional mutation breeding methods, based on available
datasets.
Generally, for plant breeding applications, CRISPR-Cas9 specificity is of importance, however, since during
plant breeding practices often several generations are passed with selection based on phenotype and/or
genotype and there is the possibility of backcrosses, off-target effects are tolerable and can be removed
(analogous to classical mutation breeding), in contrast to therapeutic genome editing applications.
2.5.1.2 Safety considerations in respect to CRISPR-Cas9 transgene retention, background
mutations caused by transformation procedures and the use of viral vectors
For safety considerations based on the above, please refer to chapter 3.4, since these are covered
also in the context of rapid-cycle breeding.
2.5.2 SDN2 technique in genome modification of plants
The SDN2 technique targets specific loci to introduce mutations of
a priori known sequence changes.
For that, together with the CRISPR-Cas9 module DNA repair templates are co-transformed that are
identical in sequence to the targeted locus with the exception of the intended sequence changes. In
a certain proportion of cells these are used as templates by the HR repair pathway of the CRISPR-
Cas9 induced DSB and thus the changes are implemented at the targeted locus.
31
link to page 35
CRISPR-Cas
For SDN2, the same applies in regard to unintended mutational load as for SDN1.
The repair template may be integrated as a whole at the locus with the targetd DSB for example by
the NHEJ repair pathway, as well as at sites in the genome. Analysis of genome edited plant lines for
ectopic integration of cisgenes can be done by standard methods (Southern Blot, PCR based
methods) and plant lines without ectopic integration can be selected accordingly.
2.5.3 SDN3 technique in genome modification of plants
Safety aspects of of cis- and intragenic plants have been covered in comparison to transgenic plants
in the study of AGES [3] and in a Scientific Opinion by EFSA [146]. In contrast to conventional y
generated cis-, intra-, and transgenic plants, the SDN3 technique is used to insert DNA at
a priori intended loci. Safety aspects concerning impairment of endogenous genes and creation of novel
reading frames can therefore be already addressed at the development phase of the plant line.
2.6 Detection and identification
It is to be expected that genome edited plant lines free of CRISPR-Cas9 transgenes will be
established, where feasible due to the production and/or the breeding process. In cases where a
CRISPR-Cas9 module is present in the established genetically modified plant line detection and
identification rationale follows those of conventionally transgenic plants. The CRISPR-Cas9 transgenic
sequence in combination with its genomic integration location then provides a marker for GM
detection and event-specific identification.
2.6.1 Detection and identification of SDN1 and SDN2 genome editing
CRISPR-Cas9 generates random site directed mutations, smal insertions/deletions, larger deletions
and nucleotide substitutions (SDN1), and mediates incorporation of
a priori designed mutations,
mostly small insertions/deletions and/or nucleotide substitutions (SDN2).
The quality of SDN1 and SDN2 mutations do not allow conclusions on their origin
Nucleotide changes (down to single nucleotide polymorphisms (SNPs)) are detectable by standard
PCR based, hybridization based or sequencing methods [147]. The induced genomic changes cannot
be distinguished from naturally occurring variation or from changes derived from conventional
mutagenesis (see chapter
2.5.1.1). Therefore, the presence alone of a mutation at a genomic site
cannot be causally linked to it being generated by the application of CRISPR-Cas9 technology.
Circumstantial evidence based on background markers may be used for identification of a genome
editing event. In case a particular mutation of a genome editing event is described in combination
with marker states of the background genome of the plant line in which it was generated, these in
32
link to page 108 link to page 17 link to page 21 link to page 19
CRISPR-Cas
combination may be used to indicate the probability of the origin of a mutation (and therefore
identification) in a sample. However, the use of the genome edited line in breeding programmes will
break up linkage to those background markers and therefore decrease or abolish evidence of the
origin of the mutation.
It is to be expected that genome editing wil be targeting various loci inducing different site directed
random mutations or modifications (like conventional mutagenesis). As a consequence, a general
screening strategy for the detection of mutations derived from SDN1 and SDN2 techniques would
have to include a combination of many tests, rather than few universal tests to collectively cover
several events.
In case prior knowledge of induced mutations is absent, detection and identification is technically
impracticable.
2.6.2 Detection and identification of SDN3 genome editing
Detection and identification of SDN3 genome editing follow the same principle as for conventionally
generated transgenic plants. For cis- and intragenic lines the detection step, i.e. the detection of
distinct sequences indicating cis- or intragenic status in a general screening step, is made more
labour-intensive because of sequence homology of inserted sequences to endogenous genes (as
discussed by AGES for conventionally generated cis- and intragenic plants [3]). Use of event-specific
analyses (identification) provides unambiguous evidence of cis- or intragene presence or absence.
Genome modifications generated by SDN1 and SDN2 genome editing techniques can be detected,
however, their presence does not provide evidence on how they originated: they cannot be distinguished
from naturally occurring variation or mutations derived from conventional mutagenesis.
Genome modifications generated by SDN3 carry a cis-, intra-, or transgene, therefore, detection and
identification is analogous to conventionally established cis-, intra-, or transgenic plants.
2.7 Aspects of GMO classification of CRISPR-Cas9 genome edited plants
Directive 2001/18/EC provides a definition of GMO (Annex
7.2). This report provides information on
the CRISPR-Cas9 technology and its application in genome editing in plants: (i) a description of the
origin and molecular mode of action of CRISPR-Cas9 (chapte
r 2.1.1), (ii) a description of the different
types of genetic modifications possible to generate in plants (chapte
r 2.1.3) and (iii) an overview on
production processes to obtain genome edited plants (chapter
2.1.2). By that, it covers potential y
relevant aspects to classification according to Directive 2001/18/EC, which are summarised briefly for
each technique below.
33
link to page 22
CRISPR-Cas
A functional CRISPR-Cas9 entity is a ribonucleo-protein complex that can be programmed to target
certain genomic locations where it induces a DSB in the targeted DNA sequence. A DSB is repaired by
endogenous cellular repair mechanisms and gives rise to an
a priori unknown mutation. There are
nucleotide changes that occur preferential y, driven by the mode of action of the endogenous repair
mechanism active, and possibly dependent on the plant species and the nature of cel s in question.
Targeting of the CRISPR-Cas9 module is mediated by the RNA component and may lead to
unintended off-target effects at genomic locations with sequence complementarity to the so cal ed
spacer sequence. If, in addition to the CRISPR-Cas9 module, DNA molecules are transformed into
cells, they can be employed to achieve further modes of genome editing (SDN2, SDN3).
Genetic modifications possible to generate using ZFN technology and in extension other SDN
technologies have been categorized into three classes (SDN1, SDN2 and SDN3;
Fig. 2.6) by the New
Techniques Working Group (NTWG) [2]. SDN1 generates site directed random mutations, SDN2 site
directed intended (
a priori) mutations and SDN3 inserts cis-, intra-, or transgenes at the targeted
genomic locus. In addition, due to ease of multiplexing ability, CRISPR-Cas9 can also be used to
generate small or large deletions at targeted genomic locations (subsumed under SDN1 technique
based on similarity of the production process (see also Study on behalf of BAFU (Federal Office for
the Environment, Switzerland) [148]). In contrast to other SDN technologies (ZFN, TALEN, MN), a
functional CRISPR-Cas9 module consists of a protein and an RNA component.
SDN1, the targeted mutation of a locus with
a priori unknown sequence change, is a form of
mutagenesis using an organic particle as mutagenic substance. During the process, DNA coding for a
CRISPR-Cas9 module (or, in case of multiplexing, two or more modules) may be employed for
delivery of a CRISPR-Cas9 module into cel s, and depending on the production process, it may be
integrated as a transgenic locus in the genome. The intended heritable genetic modification is
independent of the CRISPR-Cas9 module, therefore, mutant plant lines devoid of the transgenic locus
can be selected in subsequent generations in sexual y propagated species. The CRISPR-Cas9
transgene has thus been present in individuals (intermediate organism) during the production
process, but is not present in the final established genome edited plant line (resulting organism).
Alternatively, RNA coding for the sgRNA and Cas9 or ribonucleo-protein complexes can be delivered
into cells as such.
SDN2, the targeted mutation of a locus with a priori intended sequence change, requires additional
delivery of a DNA fragment into cells, which is used by endogenous repair pathways as a template,
and by that incorporates the intended genomic modification at the targeted locus. The nucleotide
sequence of the repair template is identical to the target locus with exception of a single or a small
number of nucleotide sequence changes or small deletions or insertions of a few nucleotides. As for
SDN1, the intended mutation is independent of the presence of the CRSIPR-Cas9 module and the
34
CRISPR-Cas
same delivery methods as for SDN1 apply for the CRISPR-Cas9 module. Similar to SDN1, it is a
mutagenesis technique involving a CRISPR-Cas9 module and a repair template leading to
a priori intended mutations.
SDN3 intends to insert a cis-, intra-, or transgene at a targeted genomic locus and requires co-
delivery of the DNA sequence to be inserted. In contrast to SDN1 and SDN2 it does not aim at
modification of an endogenous genomic locus, but at precise integration of extra-genomic
sequences. Similar to SDN1 and SDN2, the intended sequence insertion is independent of the
presence of the CRSIPR-Cas9 module and therefore, plant lines carrying the insertion but lacking a
CRISPR-Cas9 module can be generated. SDN3 generated plant lines are similar to cis-, intra-, or
transgenic plant lines, however the extra-genomic sequence has been inserted at a targeted genomic
locus mediated by a DSB introduced by a CRISPR-Cas9 module.
2.7.1 Evaluation of ZFN and related genome editing techniques by the German
expert commission ZKBS
The position statement [7] of the ZKBS (Zentrale Kommission für die Biologische Sicherheit) includes
the assessment of ZFNs, and as noted in their position statement, their assessment can be
extrapolated to other DSB-producing site directed endonucleases (SDN). In their statement they
provide conclusions on their interpretation of the term GMO in Directive 2001/18/EC in relation to
SDNs:
In relation to delivery of ZFNs, in their opinion, type B intermediate organisms (i.e. organisms with
transiently present recombinant DNA which has not been chromosomal y integrated) do not fal
under the definition of Directive 2001/18/EC. Further, if ZFNs are delivered by isolated mRNA or as
isolated proteins, in their opinion they are not covered by the GMO definition of Directive
2001/18/EC since there was no heritable genetic material involved in the production process.
In relation to resulting organisms, plants derived by ZFN1 and ZFN2 techniques are assessed by the
ZKBS as not falling under the 2001/18/EC GMO definition. They remark for ZFN1 that the resulting
organisms carry mutations generated with involvement of the endogenous mechanism of NHEJ and
that the same mutations may be generated by natural processes as well as by conventional
mutagenesis breeding.
For ZFN2, in their opinion organisms altered by the size of 20 or less nucleotides do not fal under the
definition. This is based on the notion that the genetic difference between the co-delivered repair
template and the endogenous to be edited gene in that case does not represent a “recombinant
nucleic acid”.
35
CRISPR-Cas
Organisms resulting from ZFN3 are falling, in their opinion, under the definition of GMO given in
Directive 2001/18/EC.
While SDN3 techniques generate cis-, intra-, or transgenic plants falling under the EU GMO definition
(Directive 2001/18/EC), there is legal uncertainty whether genome modified plants resulting from SDN1
and SDN2 techniques do so as wel .
SDN1 and SDN2 technique lead to plants with targeted introduced mutations. In the process of establishing
SDN1 and SDN2 genome edited plants intermediate plants may be generated that stably integrate a
CRISPR-Cas9 transgene. In sexual y propagated crops, the transgene and the intended genome modification
can be separated resulting in progenitors with the genome modification but not possessing any transgene.
Furthermore, techniques delivering CRISPR-Cas9 into cells without transfer of heritable, genetic material
are being developed.
A national expert group in Germany (ZKBS) published a position statement, in which they conclude that in
their opinion resulting genome edited plants without a stably integrated transgene do not fal under the EU
GMO definition (based on ZFN mediated genome editing).
Directive 2001/18/EC explicitly excludes plants generated by conventional mutagenesis breeding and plants
generated by cell or protoplast fusion, as well as does not consider plants generated by polyploidy
induction; plants generated by these techniques are exempted from the risk assessment and regulatory
procedure established by Directive 2001/18/EC that – based on the precautionary principle – has the
objective to protect human health and environment.
Directive 2001/18/EC therefore implicitly states that the risks associated with intended and unintended
mutations by the exempted techniques (mutagenesis breeding, cel culture methods and bringing together
related genomes or multiplication of genomes), are considered to be manageable outside the regulatory
procedure of Directive 2001/18/EC, that is by the breeding practices implemented by breeders. This is
based on the considerations that the Directive should not apply to techniques of genetic modification
which have conventionally been used and have a long safety record (recital 18 of the Directive). From a
scientific view, the mutations – intended and unintended - generated by SDNs in (cis-, intra-, and transgene
free) genome edited plants are not qualitatively different from plants arising from natural mutation events
or generated by breeding practises not falling under Directive 2001/18/EC. With respect to the quantity of
mutations, genome editing induces a minimal number of mutation events, i.e. far less than induced by
e.g. chemical mutagenesis breeding (typically 100s to 1000s).
36
CRISPR-Cas
2.8 Tables
Table 2.1 Reported chemical mutagenesis effects on plant genomes in TILLING projects
Species
Mutagen
Nr of loci
Size of M2
Mutation density
Predominant type
Reference
Derived average
(concentration) analysed
population
(1 mut/kbp)
of mutation
genomewide nr of
mutations per
individual*
Glycine max
EMS
(Forrest)
(40 mM)
7
529
1/140
G/C>A/T
Cooper
et al., 2008
[149]
~ 7900
Glycine max
EMS
(Wil iams 82)
(40 mM)
7
768
1/550
G/C>A/T
Cooper
et al., 2008
[149]
~ 2000
Oryza sativa (japonica
EMS
Nipponbare)
(1,5%)**
10
768
1/294
G/C>A/T
Til
et al., 2007
[150]
~ 1300
Oryza sativa
NaN3/MNU
(japonica Nipponbare) (1mM/15mM)
10
768
1/265
G/C>A/T
Til
et al., 2007
[150]
~ 1400
~ 340,000 (Chen
et
Triticum aestivum
EMS
al
(0,8%)***
3
512
1/47
G/C>A/T
Chen
et al., 2012
[151]
., 2012) [151] per
individual
*based on haploid genome size of ~1115 Mb [152, 153] and ~389 Mb [154] of
Glycine max (Williams 82) and
Oryza sativa (
japonica Nipponbare), respectively.
** ~145 mM
*** ~77 mM
EMS: ethyl methanesulfonate; NaN3-MNU: sodium azide-methyl nitrosourea
37
CRISPR-Cas
Table 2.2 Reported gamma irradiation effects on plant genomes
Mutation type
Species
Dosis
Subject of
Mutation
(Gray)
study
SNPs
Indels
Copy number
Presence/absence
detection
Reference
(1-5 bp)
variations
variation
Oryza sativa
Red-1
9.19 % of
Solexa whole
Cheng
et al.,
(9311)
300
(M6 inbred genome
381,403* 50,116
1,279
10,026
genome
line)
altered
sequencing
2014 [155]
Synthetic
1,510
2 % marker
wheat SW58 350/450 DGRH1
loss in D
nd
nd
nd
nd
35 SSR marker
Kumar
et al.,
individuals genome
2012 [156]
4 confirmed gene
deletions
(homozygous) of
derived:
TaPFT1-D across
Hybridisation
on average
Triticum
M2 individuals
based qPCR
Fitzgerald
et 110/82 gene
aestivum
50
4500 M2
specific for
al., 2010
deletions
(Chara)
individuals nd
nd
nd
nd
homeolog
3 confirmed gene
deletion
[157]
(homozygous)
deletions
detection
per M2
(homozygous) of
individual **
TaPFT1-A across
M2 individuals
*validation of SNPs by Sanger sequencing: 60/63 true
**based on the assumption of 124,000 genes (A,B,D) in
Triticum aestivum [158]
DGRH1: D genome radiation hybrid panel; SSR: simple sequence repeat.
38
CRISPR-Cas
Table 2.3 Excerpt of Appendix Table 5.3 : off-target identification of CRISPR-Cas9. Studies/experiments with detected off-target effects are coloured green.
Target
Off-target candidate
Nr. of mismatches
locus
locus identification
distribution
Method of detection
Off-target activity detected
Experimental system
Reference
G. max
BLASTn (e value
2-6 mm
Amplicon sequencing
none detected
soybean hairy root system
[134]
07g14530 threshold 5)
Distributed
(n=10 biological replicates)
10 candidate loci
DDM1
BLASTn (e value
4 mm
Amplicon sequencing
none detected
soybean hairy root system
[134]
gRNA1
threshold 5)
Distributed
(n=10 biological replicates)
1 candidate loci
DDM1
BLASTn (e value
2 mm
Amplicon sequencing
Yes, in al experimental
soybean hairy root system
[134]
gRNA2
threshold 5)
seed region
(n=10 biological replicates)
repeats
1 candidate locus
BLASTn (e value
3 mm
Amplicon sequencing
none detected
soybean hairy root system
[134]
Met1
threshold 5)
Distributed
(n=5 biological replicates)
1 candidate locus
BLASTn (e value
6 and 2 mm
Amplicon sequencing
yes, gRNA with 2 mm in
soybean hairy root system
[134]
miR1514
threshold 5)
Non-seed region
(n=4 biological replicates)
non-seed region in al
2 candidate loci
experimental repeats
H. vulgare
2 candidates based on
1 mm in seed region
Sequencing in 93/95 T1
Yes, gRNA with mm (further stable transformation
[102]
HvPM19-1 homology
each
individuals of two independent
away from PAM than 2nd
T0 lines
off-target) in seed region,
3/93 individuals
HvPM19-3 2 candidates based on
1mm in seed r.
Sequencing in 76 T1 individuals None detected
[102]
homology
3 mm distributed
of one T0 line
O. sativa
Selected based on
3-5 mm
Sequencing at target locus in 20 none detected
stable transformation
[132]
DERF1
homology
2 only in non-seed
GE lines (T0 and T1, al
5 candidates
region
independent lines)
Selected based on
3-5 mm
Sequencing at target locus in 20 none detected
stable transformation
[132]
MYB1
homology
2 only non-seed region
GE lines (T0 and T1, al
3 candidates
(5 mm)
independent lines)
Selected based on
1-7 mm
Sequencing at target locus in
Yes, at 1 candidate locus 7
stable transformation
[132]
YSA1
homology
2 only non-seed region
~70 Cas9 positive lines
plants with off-target
5 candidates
(1 and 7 mm)
(independent T0 lines)
activity: locus with 1 mm in
non-seed region
39
CRISPR-Cas
Target
Off-target candidate
Nr. of mismatches
locus
locus identification
distribution
Method of detection
Off-target activity detected
Experimental system
Reference
CRISPR-P
3, 4 mm
Sequencing of target locus
none detected (50 plants of
stable transformation
[109]
AOX1a
Selected 2 highest
distributed
T0 and T1)
ranked
CRISPR-P
3, 4 mm
Sequencing of target locus
none detected (49 plants of
stable transformation
[109]
AOX1b
Selected 2 highest
distributed
T0 and T1)
ranked
CRISPR-P
2, 3 mm
Sequencing of target locus
none detected (60 plants of
stable transformation
[109]
AOX1c
Selected 2 highest
distributed
T0 and T1)
ranked
CRISPR-P
1 mm non seed r.
Sequencing of target locus
Yes, activity detected in 2
stable transformation
[109]
BEL
Selected 2 highest
3 mm distributed
plants at locus with 1 mm
ranked
(89 plants of T0 and T1)
3 candidates selected
1 mm non seed r.
CAPS marker, sequencing
Yes, activity detected (6/13
stable transformation
[114]
based on homology,
regenerated plants)
confirmed by CRISPR-P
Yes, activity detected (10/13
as among possible
2 mm seed/non-seed r.
regenerated plants)
targets (rank 1, 2, 10)
none detected (0/13): mm
nearest to PAM
3 mm seed/non seed r.
(al regenerated plants from
1 transformation event
CDKB2
(cal us); conclusion
repeatable in 3 further
transformation events
(cal i))
Further 3 candidates
CAPS marker
none detected
ranked 3, 5, 9 by CRISPR-
P
40
link to page 47
Rapid cycle breeding
3 Accelerated breeding – rapid cycle breeding
3.1 Introduction
Accelerated breeding, also termed rapid cycle breeding, is a technique to shorten the duration of breeding
programmes. Specifically of interest in species with long generation times, as in perennial, woody plants
(shrubs, trees), it is achieved by establishing plant lines carrying transgenes that confer a dominant
precocious flowering phenotype. These lines are used as crossing partners to shorten the individual
breeding cycles. At the end of the breeding process, individuals carrying the desired trait/trait or
trait/genomic background combinations but lacking the early flowering transgene are selected for further
propagation
(Fig. 3.1) [159].
Fig. 3.1 Conventional versus rapid cycle breeding timeframes (after [159]). Conventional breeding
cycles in apple may take 6-12 years. In rapid cycle breeding, first an early flowering transgenic
cultivar is established and may be used for different breeding objectives, here the introgression of a
desired trait from a wild relative. The transgenic line dominantly inducing early flowering is crossed
with the wild relative and backcrosses of selected individuals can be carried out after shortened
cycles. At a cycle where individuals carrying the trait of interest in a domestic apple background are
present, individuals lacking the early flowering transgene are selected for further propagation
(arrow). BC: backcross; EFT: early flowering transgene; F1: hybrid.
41
link to page 48
Rapid cycle breeding
In addition to the above, currently it is also explored to cause precocious flowering transiently in
each generation by viral induced gene expression/silencing or by grafting on transgenic rootstock
[160, 161]. Naturally occurring genetic diversity (or induced by conventional mutagenesis) may be
used for the same purpose, however, as of yet, there is a lack of suitable precocious flowering
mutants in perennial species [159].
The juvenile phase, per definition the vegetative phase in which plants are not competent to flower
independent of otherwise favourable environmental conditions, can for example, last up to 6 - 12
years for apple and pear in field conditions (see Table 1 in [162]) and so is a major determinant of
generation time. The timing of flowering in plants is coordinated by an extensive gene network: it is
depending on environmental and autonomous signals and is altogether suppressed in plants going
through juvenile phases [163]. An increasing number of flowering time regulators are uncovered,
several of which were tested for their potential in rapid cycle breeding in diverse species
(Fig. 3.2).
The key to successful application of rapid cycle breeding in the context of a given plant
species/cultivar lies in identification of suitable candidate genes that shorten the juvenile phase and
at the same time retain proper floral organ development and fertility;
Arabidopsis thaliana LEAFY (
AtLFY), for example, induces early flowering in a citrus hybrid (
Citrus sinensis ×
Poncirus trifoliata)
[164] but not in an apple cultivar (
Malus ×
domestica cv ‘Pinova’) [165].
BpMADS4, a
FRUITFUL (
FUL)
homolog from birch, is used for accelerated breeding programmes in apples [166] and poplar
FLOWERING LOCUS T1 (
PtFT1) in plums [167].
Fig. 3.2 Fraction of the gene network regulating juvenile to reproductive phase transition in
Arabidopsis thaliana (extracted from [163]). Homologs of for example
FT and
FUL are used in rapid
cycle breeding programmes [168, 169]. Arrow: activation; broken arrow: indirect activation; bar
head: repression; line: interaction with unknown direction.
AP1:
APETALA1;
FT:
FLOWERING LOCUS T;
FUL:
FRUITFUL;
LFY:
LEAFY;
TFL1:
TERMINAL FLOWER1.
Synonyms used are high-speed breeding, fast breeding, FasTrack (fast track) breeding and rapid cycle
breeding; it was agreed upon using the term rapid cycle breeding in the future [166].
42
Rapid cycle breeding
3.2 Potential applications in plant breeding
Plant breeding in species with long juvenile phases, such as in, for example, shrubs and trees, is a
time consuming process. Juvenile phases of apple and plum cultivars (or wild relatives) last between
5-12 [162, 170] and 3-7 years [167], respectively; during this time flower formation is suppressed. A
central role in plant breeding play controlled crosses between varieties within a species or to related
species, and depending on the breeding goal may involve several cycles of successive crosses. Rapid
cycle breeding, in establishing transgenic lines with reduced juvenile phases, has the potential to
reduce breeding programmes in that the crossing cycles are shortened in time.
In apple, disease resistance germplasm is also present in wild relatives [171] (fire blight:
Malus fusca
[168],
Malus robusta [172]; apple scab:
Malus floribunda [173],
Malus sieversii [174]) and markers
tagging resistance genes are being developed [172, 174]. If these disease resistance gene resources
from wild apple relatives are to be used by introgression into
M. domestica, several successive cycles
of pseudo-backcrosses need to be done to re-establish the
M. domestica background genome (on
average, 5 backcrosses lead to < 2% of the related species in the background genome [175]). Many of
today´s scab resistant cultivars rely on Rvi6/Vf mediated resistance which was derived from the wild
relative
Malus floribunda, with initial hybridization crosses tracing back to 1914 [173, 176] and it
taking several decades to establish elite cultivars carrying Vf resistance genes [177]. In apple, rapid
cycle breeding programmes based on a transgenic early flowering line have been established. One
breeding goal is to introgress the apple scab resistance from
Malus fusca; generation cycles reported
lasted ~ 12 months [166, 176].
Successive crosses are also needed when pyramiding genes of interest in cultivars: it is known for
disease resistance that when based on a monogenic trait in combination with widespread use the
possibility of resistance breakdown increases. For example, there are sporadic observations that
Rvi6/Vf resistance has been overcome by a
Venturia inaequalis strain (causative organism of apple
scab), but the virulence gene has not spread through the
V. inaequalis population due to pathogen
management [173]. Therefore, breeding goals are to pyramid multiple resistance loci in a cultivar or
breed for quantitative resistance, i.e. several genes underlying the resistance trait, as well as cultivars
carrying resistance genes against diverse pathogens, by carrying out crosses with appropriate
breeding partners. In recent years, in addition to rate breeding offspring phenotypically, also for
perennial species marker assisted selection (MAS) has become feasible by establishing an increasing
number of markers tagging major QTLs underlying traits of interest for breeders (markers established
in apple can for example be found in [178]) and MAS applications are further being developed
(RosBREED programme, USA [179], FruitBreedomics project (EU FP7 funded [180])). At the same time
genome databases have begun adding genome sequences and assemblies also of perennial species,
43
link to page 60 link to page 47
Rapid cycle breeding
like fruit trees (apple, pear, peach, orange [181-184]), which in the future brings the potential to
integrate a large array of markers into breeding processes (for reviews and opinions please refer to
[185-187]). Rapid cycle breeding, together with MAS, has the potential to support and initiate
breeding programmes in perennials by reduction of time and cost of infrastructure [185-187] for
current breeding goals such as disease resistance breeding, low allergenicity apples and quality traits
underlying processing and fresh-cut market requirements [188].
Particularly in perennial species there is often low genetic diversity present in commercially used
cultivars, because of time and costs associated with breeding; breeding is then often based on
crosses between a few successful cultivars, as well as mutation breeding and spontaneous mutations
(‘sports’) often contributing to cultivar development [186]. As a consequence for example in apple,
although there is a high number of germplasm accessions and a large genepool in related species,
only a smal number of genotypes have been used for commercial development in the last century
[189, 190]. Rapid cycle breeding may thus also contribute to increase genetic diversity in commercial
species by making crossing cycles manageable in a reasonable timeframe.
Rapid cycle breeding, by shortening breeding programmes in species with long generation times, may
contribute to resistance breeding in for example fruit trees, and in general may increase the number of
breeding programmes. By that, it may increase genetic diversity in available germline used for breeding and
establishing commercial cultivars in species with otherwise often narrow genetic breeding material.
3.3 State of development
3.3.1 Species of interest and genes tested for precocious flower induction
One of the earliest reports on precocious flower initiation induced by transgenesis is based on the
A.
thaliana gene
LEAFY (
LFY) and its constitutive expression in its own genomic background as wel as in
hybrid aspen (
Populus tremulus × tremuloides) in 1995 [191]. To date, homologs of at least 5 genes
involved in juvenile – reproductive phase transition and/or floral meristem initiation have been
shown to be able to induce precocious flowering in certain woody species when overexpressed,
AP1 (
APETALA1),
FT (
FLOWERING LOCUS T),
LFY (
LEAFY) and
MADS4 (a FRUITFUL homolog), or
downregulated (via RNAi),
TFL1 (
TERMINAL FLOWER1), depending on the regulatory function in the
genetic network (se
e Table 3.1 and references therein;
Fig. 3.1). The research focus for applications
in breeding is in woody, perennial species; the most scientific publications can be found for apple and
poplar, followed by citrus. Single studies can also be found for birch, eucalyptus, pear and plum in
the scientific literature. However, for plum there has been set up a rapid cycle breeding program in a
44
link to page 60 link to page 60 link to page 60
Rapid cycle breeding
collaboration of the USDA/ARS and several US based Universities (see below). Furthermore, there is
one study published in soybean (
Glycine max), an annual plant with a range of maturity groups
describing the duration of the vegetative phase [192]. In this report mid to late maturity types were
induced to flower after 35 – 45 days post inoculation, half the time than control plants, independent
of photoperiod conditions which was discussed to be of potential interest also for soybean breeding
[192].
Early flowering in most cases was induced in a certain proportion of independent transgenic lines in a
given study, possibly depending, among other reasons, on the locus of transgene insertion.
Furthermore, the ability of a certain transgene to induce early flowering in a species may be
dependent on the respective genetic backgrounds (compatibility of origin of transgene (species,
cultivar) and target genetic background (species, cultivar), as for example
Arabidopsis thaliana LEAFY (
AtLFY), induced early flowering in a citrus hybrid (
Citrus sinensis ×
Poncirus trifoliata) [164] and in
hybrid aspen [191] but not in transgenic apple lines (
Malus ×
domestica cv ‘Pinova’) [165]. Where
reported and where early flowering was successfully induced, flowers were fertile (Table x.1). Several
studies report a certain extent of morphological/developmental deviations of floral organs in
comparison to wild type flowers (for example in plum [193], poplar [194], apple [195]), the extent
depending on the species and the transgenic strategy used to induce early flowering. In some cases
this leads to reduced fertility (for example reported in relation to breeding program in apple [166]).
Crosses performed (see
Table 3.1 and references therein) lead to viable offspring in citrus, apple,
plum, pear and poplar. Data from, for example, plum [193] and apple [195] show expected
segregation pattern of progeny for the early flowering transgene
(Table 3.1).
3.3.2 Experimental systems to induce precocious flower induction
In most studies stably transformed transgenic lines were generated to test the potential of
transgenes to induce an early flowering phenotype and which potential y might be used as a
breeding partner. The expression of the transgene conferring early flowering was mostly driven by a
constitutive CaMV 35S promote
r (Table 3.1). There are several studies using heat inducible
promoters (heat shock promoter (HSP) from
Glycine max) for expression of the early flowering
transgenic traits in order to minimize the effect of the transgenes during plant development at times
where transgenic activity is not needed [194, 196-201]. For use of the inducible system, regimens of
heat treatment had to be established, in order to induce gene expression but at the same time
maintain plant habitus and meristem viability. Use of the inducible system to induce early flowering
was successful for example in poplar [194] and apple [200] but not in the study of Weigl
et al., 2015
[197] due to the negative effect of heat treatment on flower formation. Transformation was mostly
carried out by
Agrobacterium mediated transformation, the tissue transformed and the cell culture
45
link to page 60 link to page 65
Rapid cycle breeding
procedures of generating stable transgenic lines differ depending on established protocols in each
species.
In apple, three related studies report apple latent spherical virus (ALSV) vector system to drive gene
expression for precocious flower induction in apple [160, 161, 170] and one study in pear [202]. ALSV
is a member of the genus Cheravirus which are bipartite (+) ssRNA viruses [203]. A vector system for
ALSV is established [204]. ALSV inoculation of plants can be carried out without DNA transfer,
however, ALSV has been shown to be seed transmissible [170]. In the study in pear a virus
elimination procedure based on growth at high temperature has been established [202]. ALSV driven
precocious flowering was also used in the study with soybean where there was a certain percentage
of seed transmission [192].
Several studies (mostly in apple, one study in poplar;
Table 3.1) tested whether the induced early
flowering phenotype is graft transmissible, i.e. whether wild type plants (acting as scions) also are
induced to flower precociously after grafting onto rootstock of transgenic early flowering lines [168,
194, 195, 200, 205]. Although in a study of transgenic poplar mRNA of AtFT could be detected in the
scion [194], graft transmissibility of the phenotype, precocious flowering, based on FT, TFL1-RNAi or
MADS4 could not be shown in any of the studies. FT is a compelling candidate for testing graft
transmissibility since it has been shown to be part of the systemic flower inducing “florigen” signal;
movement of both, FT mRNA and protein, has been implicated in florigen activity [206].
3.3.3 Current rapid-cycle breeding programmes
Based on the work cited above breeding programmes in apple and plum have been established
(Table 3.2), both with involvement of Federal Research Agencies.
Example apple
In the published rapid-cycle breeding programmes in apple the goal is to pyramide disease resistance
genes, both from wild apple species and domestic apple cultivars, into domestic apple to generate
commercial cultivars [166, 168, 207, 208]. Federal Agencies involved are the Julius Kühn-Institut,
Germany, and Agroscope, Switzerland. The breeding is built on a transgenic early flowering line (due
to transformation with BpMADS4) of the cultivar ‘Pinova’ (T1190) which was crossed to
Malus fusca to introgress fire blight resistance. Since markers are not established for the
Malus fusca fire blight
resistance, F1 individuals were screened phenotypically for resistance. Resistant individuals carrying
the transgenic precocious flowering locus were then crossed to lines with (i) known scab and fire
blight resistance loci (
Rvi2,
Rvi4, FB-F7; cv ‘Regia’) or (ii) powdery mildew resistance loci (
Pl1,
Pl2;
germplasm 98/6-10) fol owed by a pseudo-backcross to ‘Golden Delicious’ to continue introgression
46
link to page 53
Rapid cycle breeding
of the resistance loci into commercially used background germplasm (refer to breeding scheme in
Fig. 5 in [168]). Individual breeding cycles were realized within a year [168].
Based on the line T1190 F1 crosses were carried out also with the ornamental apple ‘Evereste’ coding
for a strong fire blight resistance locus (Fb_E), followed by pseudo-backrosses to various commercial
M. domestica cultivars [207, 208]. Some BC2 individuals carried already less than 15% of background
genome of the ‘Evereste’ while maintaining the Fb_E resistance locus [208].
The breeding programmes are made difficult by the small number of offspring as a by-product to the
precocious flowering phenotype, and growth conditions were being adapted, as wel as suitable
crossing partners (age of wild type crossing partner) chosen [166, 168, 207, 208]. T1190 line was
chosen for the breeding program because its precocious flowering phenotype is based on a single
transgene insertion which was mapped to linkage group 4 (LG4) [168]. Further transgenic
M.
domestica early flowering transgenic lines to be used for breeding programmes were subsequently
established which each carry the transgenic construct on different LGs in several different
commercial cultivar backgrounds [166]. This ensures the presence of a diverse set of crossing
partners for breeding programmes which often aim for introgression of loci present on different LGs
and pyramiding of traits of interest in plant lines.
The published data show that the combination of (i) rapid cycle breeding and (i ) marker assisted
selection (to optimize choosing of offspring for subsequent crosses in relation to desired trait and
background genome) is a feasible breeding strategy that greatly reduces breeding time in woody
species.
Example plum
A “FasTrack” breeding programme in plum is carried out in a collaboration of the United States
Department of Agriculture (USDA) Agricultural Research Service (ARS) with University of California
Davis, Clemson University and Pennsylvania State University, US [209-211]
3. It is based on
continual y flowering transgenic plums that have been generated by stably introducing
FT1 from
P.
trichocarpa driven by the CaMV 35S promoter into diverse genomic backgrounds. A patent has been
granted in plums for this system in the United States [212]. A continually flowering plum line of the
cv ‘Blubyrd’ has been published [167]. Supported by the California Dried Plum Board (State of
California), a breeding goal is to breed plum varieties suitable for dried plum production in California
[210]. For that, a panel of different cultivars/germplasms have been selected for transformation with
PtFT1 in order to generate FasTrack crossing partners with a range of desirable traits (for example
differing harvest times, sugar content, good dried appearance and flavour). Specifically, one short
3 http://ucanr.edu/sites/fastrack/Approach/
47
link to page 54 link to page 54
Rapid cycle breeding
term goal is the introgression of the transgenic plum pox virus (PPV) resistance trait of ‘Honeysweet’
(a fresh market plum) into the genetic background of the dried plum cultivar ‘Improved French’,
which is the main planted dried plum cultivar in California. The transgenic PPV resistance trait of
‘Honeysweet’ (event C5) is approved for cultivation and food use in the US by APHIS, FDA and EPA
[211] (the application to APHIS (petition 04-264-01p)
4 contains data from experimental field tests
collected in three European countries [211]). Once the PPV transgenic trait is introgressed after
several cycles of backcrossing into the genome of ‘Improved French’, null segregants for the early
flowering trait PtFT1 will be selected for potential commercial cultivation. Null segregants derived
from FasTrack Breeding are not regulated by the USDA
5. In 2013, BC1 individuals were reported to be
germinated for a further cycle of backcrossing [210]. A long term goal in the breeding program is to
understand, using molecular markers, high fruit sugar level, which is based on complex genetic
architecture. Established markers will then be used to breed elite dried plum cultivars using the
FasTrack system1.
3.3.4 Establishing infrastructure for rapid-cycle breeding programmes
To optimize rapid-cycle breeding programmes for a given species, it is of interest to generate a panel
of independent precocious flowering lines with mapped and characterized transgene locations, each
carrying a single transgene on a different linkage group. Furthermore, established lines ideally
maintain high fertility and exhibit a plant habitus supporting fruit growth [166]. Known insert
location facilitates breeding processes because the breeder can choose suitable breeding partners
depending on the breeding goal. For example, in apple if the breeding goal was to introgress a locus
of interest with known linkage group location from a wild relative into domestic apple, it is of
advantage to choose a breeding partner which carries the early flowering transgene on a non-
homologous chromosome or as far apart as possible on the homologous chromosome [168]. If they
are located on the same homologous chromosome BC1 progenies inherit both traits only in case of
crossing over taking place. The closer the loci are located to each other, the smal er the number of
individuals in the progeny carrying both traits. The same applies at completing the breeding process,
since the early flowering transgene needs to be segregated away from the introgressed locus to
generate resulting organisms which are null-segregants for the transgene.
Therefore, Weigl
et al., 2015 [166] established several transgenic early flowering lines with transgene
insertion sites at different genomic locations and in different cultivars. Initially, transformed
individuals were screened for lines carrying single T-DNA insertions by Southern blotting. Insertion
sites were identified by genome walking and verified by PCR assays [166]. Similarly, in the dried plum
4 https://www.aphis.usda.gov/biotechnology/petitions_table_pending.shtml
5 USDA/APHIS response to Letter of Enquiry by USDA/ARS
48
link to page 47 link to page 55
Rapid cycle breeding
breeding program it was planned to establish several independent flowering lines in various
germplasms chosen based on presence of traits of interest
6 in order to ensure a set of breeding
partners.
The basis for application of rapid cycle breeding, establishing transgenic lines with precocious flowering
behaviour has been achieved in several perennial species, for example apple, poplar, citrus, pear or plum.
Optimal precocious flowering lines for breeding programmes selected need to retain fertility, and need to
be characterised for insert number and genome location. Breeding programmes have been established in
apple and plum. In apple, a major breeding goal is to combine disease resistance loci in a commercial
cultivar background.
Furthermore, it is explored to induce precocious flowering using transgenic rootstock as well as transient
induction using viral vectors.
3.4 Intended and unintended effects
The intended effect of using a transgenic, precocious flowering breeding partner is to shorten
breeding program durations by decreasing the time between successive crosses may be carried out.
To date, precocious flower initiation is induced in breeding programmes by using breeding partners
carrying dominant transgenes, furthermore, applied research explores precocious flower induction
by (i) grafting scions onto transgenic (i.e. harbouring precocious flower induction locus) rootstocks,
and (ii) transiently expressing transgenes (conferring precocious flower induction) using viral vector
systems. At the end of the breeding process, resulting individuals carrying the desired trait/trait or
trait/genomic background combinations but lacking the early flowering transgene (in case of using
transgenic breeding lines) are selected for further propagation
(Fig. 3.1).
Unintended effects by means of using transgenic plant lines as breeding partners may be (i) retention
of the transgene in resulting organisms and (i ) background mutations in the transgenic precocious
flowering lines due to the performed transformation process, which are passed on to resulting
organisms. An unintended effect due to application of rapid-cycle breeding in the case of using viral
vector systems is viral contamination of progeny by seed transmission.
Unintended effects caused by the novel combination of different genomic backgrounds due to the
breeding process are not unique to or caused by application of rapid-cycle breeding and may occur as
in conventional breeding programmes.
6 http://ucanr.edu/sites/fastrack/Approach/Obj1System/
49
link to page 54 link to page 51
Rapid cycle breeding
3.5 Safety considerations
Retention of transgene in resulting organism
Resulting organisms in a rapid cycle breeding programme are selected based on the desired
trait/trait or trait/genomic background combination analogously to conventional breeding
programmes, additionally, resulting organisms lacking the precocious flowering phenotype
conferring transgene are selected. Transgenic lines generated for rapid cycle breeding are evaluated
for transgenic state (insert number, location) since it is integral to an efficient breeding programme
to use well characterised transgenic lines. Presence/absence of the transgene is monitored during
the rapid cycle breeding process to ensure the use of appropriate breeding partners (refer to chapter
3.3.4). Standard PCR techniques are used to map transgene integration sites and used to confirm
presence/absence of the transgene; Southern blotting is routinely used to analyse transgene copy
number.
For confirmation of transgene absence in resulting organisms, PCR techniques, Southern Blotting
and/or genome sequencing using next generation technologies [213] may be used.
Background mutations caused by the transformation procedure elsewhere in genome
Experimental procedures during establishment of transgenic lines may lead to mutations elsewhere
in the genome. In relation to partial/additional transgene copies the above considerations apply.
Background mutations may be silent as wel as non-silent in regard to changes in the expression of
the genome. In the latter case, mutations may have beneficial or adverse effects, or may be neutral.
Unintended, unknown mutations similarly arise in conventional and mutation breeding. The
transgenic line is used as an initial breeding partner to introduce the precocious early flowering
transgene, and breeding programmes often involve several successive cycles of crosses. Therefore,
background mutations arisen from the transformation procedure are diminished at each cycle (on
average by half, with exception of mutations linked to the transgene) in the case transgenic lines are
not used in successive cycles.
Viral contamination
Precocious flower formation may be induced using viral vectors (see chapte
r 3.3.2). Viruses may be
passed on through seeds with a certain degree of transmissibility [214]. There have been several
strategies of viral elimination established (heat treatment or chemical treatment, passage through
tissue culture; [215]). For example, in the framework of establishing induction of precocious flower
formation using the ALSV vector system in apple and pear, it has been shown that heat treatment
might be an effective strategy to obtain viral free plants [202]. To control for viral contamination,
50
link to page 108 link to page 47 link to page 52
Rapid cycle breeding
therefore, elimination procedures exist and/or may be developed for the specific virus/plant species
combination and viral absence in resulting organisms can be confirmed using standard DNA or
protein based methods [214].
3.6 Identification and detection
Rapid-cycle breeding uses intermediate plants with precocious flower formation to shorten the
crossing cycles within breeding programmes. Intermediate plants with the precocious flower
formation phenotype may be transgenic plants. In that case, the transgenic locus in combination with
its genomic integration location provides a marker for GM detection and event-specific identification.
Individuals meeting the breeding goal, achieved by the conventional breeding process of crossing
selected breeding partners, and at the same time being nul -segregants for the precocious flowering
transgene are selected for further propagation. Therefore, the resulting organism does not carry a
transgene and cannot be detected or identified as being generated by rapid-cycle breeding by means
of DNA marker based methods.
Similarly, in the case (i) transgenic rootstock is used to induce precocious flowering in the scion or (ii)
organisms transiently expressing information of precious flower formation (VIGE or VIGS vectors that
are not seed transmissible, other transient expression systems) is used to induce precocious flower
formation, the resulting organism does not carry a transgene and cannot be detected or identified as
being generated by that process by means of DNA marker based methods.
Intermediate plants may carry a cis-, intra-, or transgene therefore, detection and identification is
analogous to conventionally established cis-, intra-, or transgenic plants.
Resulting organisms which do not carry a cis-, intra-, or transgene are not distinguishable to organisms
resulting from conventional breeding programmes.
3.7 Aspects of GMO classification
Directive 2001/18/EC contains a definition of organisms falling under the authorization procedure
(refer to Ann
ex 7.2 for definition). This report provides information on rapid cycle breeding: (i) a
description of the underlying principle (chapter
3) and (ii) a description of an ongoing breeding
program in
Malus domestica (chapter
3.3.3). By that, it covers potentially relevant aspects to
classification according to Directive 2001/18/EC.
Rapid cycle breeding uses transgenic organisms during breeding programmes (intermediate
organisms). The transgenic locus induces precocious flower formation and thereby shortens crossing
51
Rapid cycle breeding
cycles within breeding programmes. Individuals of each generation segregate for the transgene in
combination with genomic marker states of interest. Individuals are evaluated in terms of genotype
and phenotype in each generation. Those meeting the selection criteria for desired marker states
and/or phenotypes, but lacking the transgene, may be generated at certain cycles of the breeding
program. The resulting organisms represent non-transgenic (nul -segregant in relation to the
transgene) individuals which have been passed through a breeding program using transgenic crossing
partners.
Alternatively, precocious flowering may be induced by (i) grafting scion onto transgenic rootstock
(which has not yet been successful y shown to induce early flowering in the scion) and (ii) by viral
induced gene expression/repression. Grafting using transgenic rootstock in plant breeding in general
has been covered in a report of AGES to the BMG [4]. Grafting and viral induced gene
expression/repression, in case an RNA virus is used as vector, induce precocious flowering transiently
in the scion and the transfected plant, respectively.
3.7.1 Evaluation of a related breeding practise by the German expert commission
ZKBS
The method of rapid cycle breeding has not been analysed by the ZKBS [7].
From the aspect of the use and the state in respect to the transgene of the resulting individuals, the
use of a transgene in rapid cycle breeding may be compared to that in reverse breeding. The
transgene, i.e. the transgenic line, in both breeding approaches is used as a tool, not as a trait or as a
breeding goal. Transgenic lines are crossing partners to shorten the individual crossing cycles (by
conferring the trait of precocious flowering) in a breeding program, which follows conventional
breeding goals to generate novel recombined genomic states by crossing of selected breeding
partners. When achieving the breeding goal, null segregant individuals for the transgene conferring
precocious flowering are selected from the breeding population.
The position statement of the ZKBS concludes on steps in reverse breeding that may be used
analogously for evaluation of rapid-cycle breeding (and possibly other techniques using transgenic
lines as breeding partner intermediates in the future). In rapid cycle breeding intermediate
organisms are used with a precocious flower initiation phenotype that may be generated via
different strategies. To date, mostly lines carrying a transgene conferring precocious flower
production are used. Intermediate organisms with a stably integrated transgene are assessed by
ZKBS as falling under the GMO definition of Directive 2001/18 EC by the ZKBS (here in relation to
transgenic intermediates generated for suppression of meiotic recombination in reverse breeding).
Furthermore, intermediate organisms exhibiting precocious flower production may be created by
52
Rapid cycle breeding
transient transgene expression, i.e. no stable integration of transgenes into the genome, for example
by viral induced gene expression. In the case viral vectors that are not seed transmissible are used, in
an analogous situation for reverse breeding (recombinant DNA is present only transiently in the
intermediary organism and is not passed to its progeny), the ZKBS assesses these intermediary
organisms as not falling under the GMO definition of Directive 2001/18 EC, however they may
contain a GMO (recombinant virus). Precocious flowering may also be conferred by grafting a scion
onto transgenic rootstock. The ZKBS assesses progeny of these chimeras as not falling under the
GMO definition of Directive 2001/18 EC.
In respect to resulting organisms, in all three (non-exhaustive) breeding strategies (precocious flower
formation by transgenesis, grafting on GM rootstock and virus induced gene expression of non seed-
transmissible virus) progeny is generated that does not carry recombinant DNA, i.e. the trait of
precocious flowering information. In case of using transgenic lines to confer precocious flower
formation, null-segregants are selected among the progeny. An analogous situation in reverse
breeding is assessed as not falling under the GMO definition of Directive 2001/18 EC by the ZKBS.
Rapid cycle breeding uses transgenic intermediate plants to shorten the individual crossing cycles. At a
generation yielding plants with the desired breeding goal, individuals harbouring the desired genotypes but
lacking the transgene are selected.
While intermediate transgenic plants fall under the EU GMO definition (Directive 2001/18/EC), there is
legal uncertainty whether plants resulting from rapid cycle breeding and lacking a transgene do so as well.
A national expert group in Germany (ZKBS) published a position statement, in which they conclude on an
analogous case, transgene free plants resulting from reverse breeding, that in their opinion these do not
fall under the EU GMO definition.
As covered in the chapter of CRISPR-Cas9, Directive 2001/18/EC implicitly states that the risks associated
arising from intended and unintended mutations by exempted techniques of mutagenesis breeding, cel
culture methods and bringing together related genomes or multiplication of genomes, are considered to be
manageable outside the regulatory procedure of Directive 2001/18/EC, that is by the breeding practices
implemented by breeders.
From a scientific aspect, the mutations – intended and unintended – generated or introduced in (cis-, intra-,
and transgene free) plants resulting from rapid cycle breeding are not qualitatively different than to
resulting plants generated by breeding practises not falling under Directive 2001/18/EC.
53
Rapid cycle breeding
3.8 Tables
Table 3.1 Studies reporting on genetic engineering for precocious flowering in woody species
Species/cultivar
Precocious
Trait donor
Reference
Precocious flowering detected **
Fertility
flower
induction
transgene*
Betula pendula
Juvenile phase under natural conditions:
(birch)
10-15 years (Elo
et al., 2007) [216]
Betula pendula
BpMADS4
Betula
Elo
et al., 2007
Yes (11 days versus 85 days non
not reported
‘BPM2’ (early
pendula
[216]
transgenic control)
flowering clone)
‘JR1/4’, ‘K1898’
Yes (86 days after rooting)
Citrus
Juvenile phase under natural conditions:
6-20 years (Pena
et al., 2001) [164]
Citrange
AtLFY
A. thaliana
Pena
et al., 2001
Yes (6 T0 lines (out of 22) between 2 and fertile, F1 progeny with early
Citrus sinensis ×
[164]
20 months)
flowering phenotype
Poncirus trifoliata
Citrange
AtAP1
A. thaliana
Pena
et al., 2001
Yes (2 T0 lines (out of 12) after 13 and
fertile, F1 progeny with early
Citrus sinensis ×
[164]
15 months)
flowering phenotype
Poncirus trifoliata
Cervera
et al.,
Sweet orange
2009
C. sinensis
[217]
Poncirus trifoliata
CiFT
Citrus unshiu
Endo
et al., 2005 Yes (T0: 12 weeks – 8 months after
fertile, F1 progeny with segregating
[218]
transfer to greenhouse
early flowering phenotype
F1: 2 weeks)
Eucalyptus
Juvenile phase under natural conditions:
1-7 years (Klocko
et al., 2015) [198]
Eucalyptus grandis × AtFT
A. thaliana
Klocko
et al.,
Yes (1-5 months after transplanting to
fertile, viable F1 generation
urophylla
HSP::PtFT1
P.trichocarpa
2015 [198]
glasshouse)
Malus domestica
Juvenile phase under natural conditions:
(apple)
5-12 years (Yamagashi
et al., 2014); 6-12
years (Weigl
et al., 2014); 4-8 years
(Kotoda
et al., 2010) [166, 170, 219]
Malus × domestica
MdTFL1
M. domestica
Kotoda
et al.,
Yes (8 months)
fertile, seed production
54
Rapid cycle breeding
Species/cultivar
Precocious
Trait donor
Reference
Precocious flowering detected **
Fertility
flower
induction
transgene*
‘Orin’
RNAi
2006 [220]
Malus × domestica
BpMADS4
B. pendula
Flachowsky
et al., Yes (3-4 months)
fertile, seed production
‘Pinova’
2007 [221]
Malus × domestica
MdTFL1
M. domestica
Szankowski
et al., Yes (6 months)
not reported
‘Holsteiner
RNAi
2009 [222]
Cox’,’Gala’
Malus × domestica
MdFT1
M. domestica
Kotoda
et al.,
Yes (2-6 months after regeneration)
not reported
‘JM2’
2010 [219]
Malus × domestica
MdFT2
M. domestica
Traenkner
et al.,
Yes (already during
in vitro cultivation)
not reported
‘Pinova’
2010, 2011 [205,
signal not graft transmissible
223]
Malus × domestica
AtLFY
A. thaliana
Flachowsky
et al., early flowering phenotype not detected
early flowering phenotype not
‘Pinova’
2010 [168]
(7 transgenic lines)
detected (7 transgenic lines)
Malus × domestica
MdTFL1
M. domestica
Sasaki
et al., 2011 yes (1.5 – 2 months after virus-
fertile, viable seed production
RNAi
[160]
inoculation of seedlings)
viral
expression
system #
Malus × domestica
AtFT
A. thaliana
Yamagashi
et al.,
yes (1.5 – 2 months after virus
fertile, viable seed production
‘Fuji’, ‘Orin’, ‘Golden
2011 [161]
inoculation of seedlings)
Delicious’
F1 generation virus free
MdFT1
M. domestica
not detected
viral
expression
system #
Malus × domestica
AtFT & RNAi
A. thaliana
Yamagashi
et al.,
yes (1.5-3 months after virus inoculation fertile, viable seed production
MdTFL1-1 or M. domestica
2014 [170]
of seedlings)
MdTFL2
55
Rapid cycle breeding
Species/cultivar
Precocious
Trait donor
Reference
Precocious flowering detected **
Fertility
flower
induction
transgene*
(combined)
MdFT1 or
M. domestica
not detected
MdFT2 &
MdTFL1
viral
virus seed transmissible (detected
expression
in part of F1 lines), possibly cultivar
system #
dependent
Malus × domestica
MdTFL1
M. domestica
Flachowsky
et al., Yes (6 months;
fertile, F1 progeny with segregating
‘Holsteiner Cox’,
RNAi
2012 [195]
preliminary data: signal not graft-
early flowering phenotype
‘Gala’, ‘Galaxy’,
transmissible)
‘Pinova’
Malus × domestica
HSP::PtFT1
P. trichocarpa Wenzel
et al.,
Yes (6 days after 28 day heat treatment; fertile, seed production
‘Pinova’
HSP::PtFT2
2013 [200, 201]
Signal not graft-transmissible (although
PtFT RNA could be detected in scion in
one case))
Malus × domestica
HSP::MdTFL
M. domestica
Weigl
et al., 2015 heat treatment abolished floral organ
heat treatment abolished floral
‘Pinova’, ‘Gala’
1-1,2 RNAi
[199]
formation
organ formation
(same
construct as
in
Flachowsky
et al., 2012)
Populus
Juvenile phase under natural conditions:
(poplar)
P. tremula 7-10 years (Hoenicka
et al.,
2012) [197]
P. tremula × alba (f)
AtLFY
A. thaliana
Weigel
et al.,
Yes (T0: 5 months)
not reported
P. tremula ×
1995 [191]
tremuloides (m)
P. tremula × alba
PtLFY
P. trichocarpa Rottmann
et al.,
Yes (but only 1 line)
not reported
female
2000 [224]
P. tremula ×
56
Rapid cycle breeding
Species/cultivar
Precocious
Trait donor
Reference
Precocious flowering detected **
Fertility
flower
induction
transgene*
tremuloides male
P. tremula female
PtLFY
P. trichocarpa Boehlenius
et al.,
Yes(within 4 weeks on transformed stem not reported
P. tremula ×
2006 [225]
segments)
tremuloides male
P. tremula
BpMADS4
Betula
Hoenicka
et al.,
no
early flowering phenotype not
pendula
2008 [226]
detected
Populus tremula
MdFT2
M. domestica
Traenkner
et al.,
Yes (6-10 months)
not reported
2010 [205]
P. tremula × alba
HSP::AtFT
A. thaliana
Zhang
et al.,
Yes
fertile, seed production
female
HSP::PtFT1,
P. trichocarpa 2010 [194]
Signal not graft transmissible
2
P. tremula ×
tremuloides male
P. tremula ×
HSP::AtLFY
A. thaliana
Hoenicka
et al.,
not detected (heat treatment disturbed
not reported
tremuloides (male)
2012 [197]
plant growth)
P. tremula (male)
35S::AtLFY
A. thaliana
yes, early flowering (time not indicated)
not reported
35S::PtFT
P. trichocarpa
yes, early flowering ((time not indicated) not reported
HSP::AtFT
A. thaliana
Hoenicka
et al.,
yes, early flowering
fertile, viable F1 seedlings
2014 [196]
Prunus domestica
Juvenile phase under natural conditions:
(plum)
3-7 years (Srinivasan
et al., 2012) [167]
Prunus domestica
PtFT1
P. trichocarpa Srinivasan
et al.,
Yes (1-10 months)
fertile, F1 progeny with segregating
‘Blubyrd’
2012 [167]
early flowering phenotype
Graham
et al.,
2015 [227]
57
Rapid cycle breeding
Species/cultivar
Precocious
Trait donor
Reference
Precocious flowering detected **
Fertility
flower
induction
transgene*
Pyrus communis
Juvenile phase under natural conditions:
(pear)
9-14 years (Freiman
et al., 2012) [228]
Pyrus communis
PcTFL1-1,
Pyrus
Freiman
et al.,
Yes (already under tissue culture
fertile, F1 progeny with early
‘Spadona’
PcTFL1-2
communis
2012 [228]
conditions, rooted plants 1-8 months)
flowering phenotype
RNAi
Pyrus communis
AtFT & RNAi
Arabidopsis
Yamagishi
et al.,
Yes (1-3 months after inoculation of
Normal flower morphology,
PcTFL1-1 or
thaliana,
2016 [202]
cotyledons)
developing fruits
(combined)
Pyrus
communis,
AtFT & RNAi
Arabidopsis
MdTFL1-1 or
thaliana,
(combined)
Malus
domestica
viral
expression
system #
AP1: APETALA1; CiFT: Citrus unshiu FLOWERING LOCUS T; HSP: heat shock promoter; LFY: LEAFY; TFL1: TERMINAL FLOWER1. VIGS: virus induced gene silencing.
Green: Studies detecting no early flowering phenotype.
*if not indicated otherwise, EFTs are transgenes which are overexpressed in the target plant under the constitutive Cauliflower Mosaic virus (CaMV) 35S promoter.
RNAi denotes constructs using RNA interference to knock down genes with inhibitory effect on flower formation/juvenile phase progression
**time may vary between independent lines; earliest observed time listed
# apple latent spherical virus (ALSV)
58
Rapid cycle breeding
Table 3.2 Accelerated breeding programmes in apple and plum
Species/cultivar
Overexpresse
Breeding goals
Trait donor
Breeding cycle
Status
Reference
d transgene
duration
(year)
Malus domestica
(apple)
Malus × domestica
BpMADS4
Fire blight resistance
Malus fusca
BC1 (2011)
Flachowsky
et al., 2011
‘Pinova’ (T1190)
refer to Fig.5 of [168] [168]
Rvi2, 4 scab resistance
cv ‘Regia’
for breeding scheme
FB-F7 fire blight resistance
Pl-1, 2 powdery mildew
clone 98/6-10
resistance
Malus × domestica
BpMADS4
Fire blight resistance locus
Ornamental apple
BC2 (2012)
Le Roux
et al., 2012,
‘Pinova’ (T1190)
Fb_E
cultivar ‘Evereste’
~1 year
Refer to Table 1 in
2014 [207, 208]
[207] for
specification of
crosses
Malus × domestica
BpMADS4
Integration of early
/
/
Weigl
et al., 2015 [166]
‘Pinova’, ‘Gala’,
flowering transgene on
‘Mitchgla Gala’,
various linkage groups in
‘Santana’
different cultivars for
breeding as in Flachowsky
et
al., 2011 [168]
Prunus domestica
(plum)
Prunus domestica
PtFT1
Plum pox virus resistance
P. domestica
BC1 individuals
Scorza
et al., 2013
(transgenic trait) from fresh ‚Honeysweet‘
(2013)
[210]
market plums into dried
Srinivasan
et al., 2011
plum cultivars (f.e.
[212]
‘Improved French’)
http://ucanr.edu/sites/
fastrack
BC: backcross.
59
Small RNA-directed techniques
4 Small RNA-directed techniques
4.1 Introduction
Small RNA directed techniques use the cel ular machinery of RNA silencing pathways to
downregulate gene expression of target genes. For applications in plant breeding, targets may be
endogenous genes of the plant, but also of plant pathogens after interaction with the plant (feeding,
viral entry, …).
In plants, RNA silencing or RNA interference (RNAi) acts through several pathways to suppress or
decrease RNA abundance of, for example, endogenous genes, transposons or viral RNA, and so is
involved in regulating plant development and physiology, in maintenance of genome integrity and is
used by plants to battle viral attacks [229].
Fig. 4.1 Generalised overview of “the” RNAi pathway. dsRNA molecules are cleaved by DCL proteins
into small RNAs. These are incorporated (as single stranded molecules) into the so called RISC
complex, which based on sequence complementarity to the incorporated smal RNA silences target
RNAs by, depending on the pathway, target cleavage or translation inhibition, or, in the case of
transcriptional gene silencing, by
de novo DNA methylation at the target locus. AGO: ARGONAUT;
DCL: DICER-LIKE; dsRNA: double stranded RNA; PTGS: post transcriptional gene silencing; RdDM: RNA
directed DNA methylation; RdRP: RNA dependent RNA polymerase; RISC: RNA induced silencing
complex; TGS: transcriptional gene silencing.
RNAi is a mechanism found in diverse eukaryotes, sharing common core components and exhibiting
distinct features. Generally (see for recent reviews in plants [229-231]), central to triggering RNAi are
60
link to page 66
Small RNA-directed techniques
double stranded RNA (dsRNA) molecules of diverse sources
(Fig. 4.1). They are recognised and
processed by members of the Dicer family of endonucleases (DCL) into small RNA (sRNA) fragments,
in plants typically ~ 21 – 25 nucleotides in length. sRNAs are loaded (as single stranded molecules)
into complexes termed RISC (RNA induced silencing complex) containing at least a member of the
ARGONAUTE (AGO) family of proteins. AGO proteins are the main silencing effectors and possess an
RNase-H-like fold that exhibits endonuclease (“slicer”) activity. Within RISC, AGO selects the sRNA
guide strand, ejects the passenger strand and mediates sRNA – target RNA recognition. Depending
on the particular RNAi pathway, sRNA – target recognition results in post transcriptional gene
silencing (PTGS) or transcriptional gene silencing (TGS). In the former, RNA targets are cleaved or
translationally repressed/destabilized, in the latter epigenetic modification is induced, RNA-directed
DNA methylation (RdDM). In plants, RNAi pathways may also include the action of RNA dependent
RNA polymerases (RdRP), for signal amplification or on single stranded RNAs recognized as foreign or
aberrant [229-231].
RNAi pathways are further grouped based on origin and biogenesis of sRNAs and engaged members
of DCL and AGO proteins into microRNA (miRNA) and small inhibitory RNA (siRNA) pathways [229-
231]. sRNAs may act local or systemic; generally, in plants miRNAs act cell-autonomous or move cel -
to-cell over short distances, whereas siRNAs have the potential for systemic movement [232, 233].
RNAi based methods exploit the naturally occurring cellular RNAi machinery in order to downregulate
expression of target RNAs. In plants for example, a biological role of RNAi is protection from viral attacks.
Double stranded (ds) RNA molecules are recognised and processed into smal RNAs (sRNAs) approximately
20 nucleotides in length by Dicer proteins. They are loaded into a complex termed RISC. RISCs are targeted
based on complementarity to the sRNA to to target RNAs, which are cleaved by the RISC component AGO
and thereby inactivated.
4.1.1 miRNAs
miRNAs in plants have been shown to be involved in regulation of plant developmental processes
and in biotic and abiotic stress responses [234]. They are encoded at
MIR loci, non-protein coding
nuclear genes, and many belong to evolutionary conserved gene families [230].
MIR loci
preferentially encode a single miRNA
in vivo [235] and most plants code for ≥ 100 loci [236].
MIR genes are transcribed by DNA polymerase II, their products may be spliced and give rise to
imperfect self-complementary foldback precursor structures, the pri-miRNA. pri-miRNAs carry a
stabilizing 5´cap structure and 3´polyadenylated tail and are processed by different progressions
depending on their family affiliation. DCL1 is the main dicer activity on pri-miRNAs and finally
61
link to page 66 link to page 68
Small RNA-directed techniques
processes them into miRNA/miRNA* (guide/passenger strand) duplexes predominantly 21
nucleotides in length. They assemble in RISCs predominantly containing AGO1; the sorting
determinant being a 5´uridine [236]. The thermodynamic stability of the miRNA/miRNA* duplex
plays a role in guide strand determination and passenger strand elimination within RISC. Target sites
of the miRNA in plants are frequently located in open reading frames (ORF) of mRNAs [237]. Target
recognition is sequence complementarity based but perfect complementarity is not needed.
Comprehensive studies identified key features in respect to thermostability, consensus sites and
sequence homology important for biogenesis, strand selection and target recognition and thus
effective gene silencing (summarised amongst others for plants in [238, 239]).
Target recognition of miRNAs in the RISC complex may preferentially lead to direct target cleavage
(slicing) or translational inhibition/destabilization [234].
Fig. 4.2 Minimal gene cassette requirements for induction of RNAi using amiRNA constructs. The
precursor amiRNA is placed between a promoter and terminator sequence, to initiate and stop
transcription, respectively. The transcript gives rise to a stem-loop miRNA precursor transcript,
processed primarily by DCL1 into amiRNA/amiRNA* (guide/passenger strand (see chapte
r 4.1))
duplexes. The guide strands are incorporated into RISC complexes and trigger downregulation of
target RNAs. amiRNA: artificial miRNA. DCL1: Dicer like 1.
Methodology
Gene cassettes for induction of RNAi using miRNAs contain an artificial miRNA (amiRNA) precursor
between polymerase I regulatory modules for transcription initiation (promoter) and termination
(terminator) of choice
(Fig. 4.2) [239]. amiRNAs carry the miRNA sequence designed to target the GOI
in the context of a miRNA backbone [240]. The backbone used may be selected from a
MIR gene
from the same as well as a from a different plant species [239]. amiRNA design is guided by
knowledge on binding specificity parameters, thermostability and consensus sites. Web MicroRNA
Designer [239] or Plant Small RNA Maker Suite (P-SAMS) [241] are examples of programmes that
integrate this knowledge and calculate and rank potential amiRNAs by sensitivity and specificity for a
given target and plant species. Further, functional screens may be used to test the most efficient
candidates among predicted amiRNAs [242].
62
Small RNA-directed techniques
Transformation methods in use to stably introduce amiRNA constructs in plants mainly are
Agrobacterium-mediated gene transfer and microprojectile (particle) bombardment [243].
4.1.2 siRNAs
In plants, small inhibitory RNAs (siRNAs) arise mainly by DCL2, 3 and 4 activity on dsRNA derived from
diverse sources, for example viral origin, transcription of natural antisense transcripts (nat-siRNAs),
trans-acting siRNA (TAS) genes and transposon sequences. siRNAs derived from transposons and
repeat sequences depend on plant specific DNA-dependent RNA polymerases IV and V (thus termed
p4/p5-siRNAs) and ultimately mediate RNA-directed DNA methylation (RdDM). The remaining
pathways function through slicing activity on target RNAs [229-231].
Biogenesis of siRNAs differs between pathways. Common to all, and as a distinctive feature to miRNA
biogenesis, siRNA pathways do not depend on single siRNAs but usual y dsRNA is diced into several
entities. siRNA pathways in plants further may involve signal amplification steps carried out by RdRPs
[229-231] which additionally to signal amplification may lead to transitive signals, i.e. secondary
siRNAs different in sequence to the primary siRNAs [244].
Fig. 4.3 Minimal gene cassette requirements for induction of siRNA mediated RNAi using for example
(A) hairpin/inverted repeat constructs or (B) antisense constructs. The dsRNA generating constructs
are placed between a promoter and terminator sequence to initiate and stop transcription,
respectively. The transcript gives rise to a stem-loop structure, which is processed by members of the
DCL family of endonucleases into siRNA duplexes. The guide strands are incorporated into RISC
complexes and trigger downregulation of target RNAs. DCL: Dicer like.
63
link to page 69 link to page 84 link to page 87 link to page 87
Small RNA-directed techniques
Methodology
Gene cassettes for induction of RNAi using siRNAs usual y contain a hairpin construct between
polymerase II regulatory modules for transcription initiation (promoter) and termination (terminator)
of choice
(Fig. 4.3) [239]. A hairpin construct consists of inverted repeats complementary to the
target region and separated by a spacer. Transcribed hairpin RNA folds into dsRNA and acts as RNAi
trigger. Common repeat lengths are between 100 and 1000 nucleotides [245]. Alternatively,
antisense and sense constructs may be expressed which trigger RNAi by base pairing to the native
sense RNA and by a mechanism cal ed co-suppression, respectively [238, 245]. Co-suppression occurs
in situations where overexpression of sense transgenes leads to reduction of expression of both, the
transgene and the homologous endogenous gene [238].
Transformation methods used to stably introduce RNAi constructs in plants mainly are
Agrobacterium-mediated gene transfer and microprojectile (particle) bombardment [243].
RNAi pathways are distinguished based on origin and biogenesis of sRNAs and engaged members of DCL
and AGO proteins into microRNA (miRNA) and small inhibitory RNA (siRNA) pathways. Both pathways are
exploited to alter targeted traits in RNAi-based plants.
miRNAs are encoded at MIR loci which give rise to defined predominantly 21 nucleotide in length miRNAs.
They have been shown to be involved in regulation of plant developmental processes and in biotic and
abiotic stress responses.
siRNAs are processed from diverse double stranded RNA sources, for example viral RNA, natural antisense
transcripts or transposon sequences. Common to all, and as a distinctive feature to miRNA biogenesis,
siRNA pathways do not depend on single siRNAs but usual y lead to a pool of differing siRNAs.
4.2 Application of RNAi approaches in plant breeding
RNAi techniques are used to study gene function by downregulation of target gene expression and
have been adopted in applied plant research and development.
Table 4.1 lists RNAi-based transgenic
crop plants present in the scientific literature; entries are selected to exemplify potential areas of
application in plant breeding (or, in case of VIRCA project, which are in development phase).
Table
4.2 lists examples of RNAi-based transgenic crops which have been developed for the market and
have already been evaluated by regulatory agencies; some of these are or have been placed on the
market.
The RNAi-based transgene may target plant endogenous genes, and thereby affect quality or
agronomical traits as well as for example affect traits involved in abiotic and biotic stress tolerance,
64
link to page 84 link to page 84
Small RNA-directed techniques
furthermore, it may be designed to target genes expressed in plant pathogens. The latter can be used
to establish plants resistant to viral diseases, or, col ectively termed host induced gene silencing
(HIGS), protect against insects , nematodes (feeding on plants), fungal and bacterial diseases. In the
US plants expressing transgenes (RNAi-based and proteinaceous pesticidal substances) acting against
plant pests are termed plant incorporated protectants (PIPs).
4.2.1 Applications based on targeting plant endogenous genes
Most examples in the scientific literature of RNAi-based transgenic crop plants illustrating application
in plant breeding are altered in respect to quality traits or in respect to abiotic stress tolerance.
Furthermore, RNAi approaches targeting so called susceptibility (S) genes (or recessive resistance
genes) [67] may be exploited to establish lines with biotic stress tolerance. These are plant genes
that when downregulated or present as loss of function alleles (in a homozygous state) confer (often
broad-spectrum) resistance to pathogens, in turn, effectors are produced by pathogens to upregulate
those genes creating a favourable cellular environment [67]. MLO (MILDEW RESISTANCE LOCUS)
genes are a typical example; naturally occurring and induced MLO loss of function genotypes are
used as durable resistance loci for example in plant breeding in barley [246]. The principal feasibility
of using RNAi mediated downregulation of recessive resistance genes to mediate biotic stress
tolerance has been shown in rice (downregulation of Os-11N3 mediates tolerance to certain
Xanthomonas oryzae strains [247]) and in a transient expression experiment in wheat
(downregulation of TAS3 mediates tolerance to
Blumeria graminis [248]
) (Table 4.1). Whether RNAi-
based approaches (versus genome editing) in engineering resistance via S genes will be the method
of choice remains to be seen, since the chal enge wil be to alter targets in respect to its response as
susceptibility gene but at the same time retain function in its other cellular contexts.
Examples of how to develop abiotic stress tolerance traits are published in respect to drought
tolerance, in canola, corn and potato
(Table 4.1; [249-252]). In canola, an inverted repeat construct
designed to downregulate farnesyl-transferase (FTA) leads to a reduced transpiration rate by
enhanced stomatal closure [251, 252]. FTA is a negative regulator of abscisic acid (ABA) signaling and
downregulation also leads to delayed growth and to developmental defects. To bypass these
undesired effects, the inverted repeat construct targeting FTA is driven by a drought inducible, shoot-
specific
Arabidopsis promoter [252]. Under limited irrigation conditions in two field trials, seed yield
was significantly higher in two transgenic lines compared to the parental line (between 10 – 20%
yield increase), and, crucially, the transgenic lines did not perform worse under optimal irrigation
conditions. In potato, transpiration rate was reduced by using an amiRNA construct to downregulate
Abscisic Acid Hypersensitive 1 (ABH1; also known as cap binding protein 80 (CBP80)) [250]. In corn,
65
link to page 84 link to page 84 link to page 87
Small RNA-directed techniques
an enzyme involved in ethylene biosynthesis, ACC synthase 6 (ACS6) was downregulated [249]; the
plant hormone ethylene is involved in diverse pathways, but it was tested as a means to engineer
drought tolerance based on the finding that kernel abortion at the ear tip of corn is correlated with
ethylene concentration. In several field tests over two years, two transgenic lines were detected that
showed consistently a moderate but significant yield increase under drought conditions while
maintaining performance under low stress environment conditions [249]. The increased yield in
these lines correlated with a decreased anthesis-silking interval (ASI) under drought stress compared
to wild type plants, which ensures efficient pollination of ovaries [249].
Among published crop plants with altered quality traits, there are examples with increased content
of desired substances, like amylopectin (potato; [253]), amylose (wheat; [254]) or secondary
metabolites (rapeseed, tomato; [255, 256]
) (Table 4.1). Furthermore it is possible to reduce the
amount of unwanted compounds, like phytate (shown in rice [257]) or of immunogenic epitopes.
Immunogenic epitopes were shown to be reduced in transgenic apple (Mal d 1 downregulation;
[258, 259]) and carrot (Dau c 1.01/ 1.02 downregulation; [260]) lines in skin prick and oral challenge
tests, respectively, in humans and several wheat lines with downregulated α- and/or ω-gliadins
showed impaired stimulatory capacity of gliadin reactive T-Cell clones isolated from celiac disease
(CD) patients [261-263]
(Table 4.1). Transgenic rice lines with reduced phytic acid content were
generated based on downregulation of IPK1 (Inositol 1,3,4,5,6-pentakisphosphate 2-kinase), an
enzyme involved in late stages of phytic acid biosynthesis, using a seed specific promoter [257]:
Transgenic lines maintained a similar level of total phosphorus content in seeds in comparison to wild
type plants, the decrease in phytate content was compensated by an increase in inorganic phosphate
content. Despite these physiological changes, transgenic lines displayed normal phenotype especially
assayed for agronomic parameters (grains/panicle, 1000 seeds dry weight, number of effective
tillers,…), for germination behaviour, myo-inositol content and amino acid profiles of storage
proteins [257]. This is in contrast to many low phytic acid (lpa) mutants which are negatively affected
in seed performance and yield [264]. Plants use phytate to store minerals in seeds and a high
percentage of total phosphorus in crop seeds (> 65%) is present in the form of phytic acid, however,
phytic acid phosphorus and minerals complexed to phytic acid cannot be efficiently utilized by non-
ruminants, and by that also contribute to waste management problems [264, 265]. Therefore,
targeting IPK1 orthologs in a tissue specific manner may be of use to implement low phytic acid
content in other crops important for food use of non-ruminants.
There are several examples of RNAi based transgenic crop plants with altered quality traits that have
passed regulatory approv
al (Table 4.2). In the EU there are two soybean lines authorized under
Regulation (EC) 1829/2003 on genetically modified food and feed (GMO register) altered for
66
link to page 87 link to page 87 link to page 87 link to page 87
Small RNA-directed techniques
increased oleic acid content. One of the first transgenic plants authorized for growth (1992) and food
use (1994) was the FlavrSavrTM tomato in the US engineered for longer shelf life and with changed
viscosity behavior of processed fruits (see
Table 4.2 and regulatory agency reference therein).
Further, recently authorized transgenic plants in the US are an alfalfa line with reduced lignin
content, as well as a potato and an apple line both downregulating polyphenol oxidase genes to
withstand oxidative browning after slicing or bruising
(Table 4.2). The potato line additionally is
engineered for purposes of processing involving heat treatment; it does not form high acrylamide
content when for example fried, based on it having lower levels of reducing sugars and asparagine by
downregulating enzymes involved in their synthesis
(Table 4.2).
4.2.2 Applications by targeting RNA expressed by plant pathogens
Viral disease resistance
RNA silencing is used naturally by plants as a strategy of antiviral defense. Double-stranded viral RNA
– either of structured genomic regions or replication intermediates of RNA viruses, or of structured
transcripts of DNA viruses – present in plant cel s is thought to be recognised by some members of
the Dicer-like (DCL) protein family to initiate silencing and viral immunity [266]. Genetically
engineered virus resistance via RNAi uses transgenes that are designed to induce siRNA formation
(f.e. inverted repeat constructs) or amiRNAs aimed at viral sequences. There are several examples in
the scientific literature for RNAi mediated viral resistance in crop plants, for example in barley [267],
tomato [268], and wheat [269] against barley yellow dwarf virus (BYDV), cucumber mosaic virus
(CMV) and wheat streak mosaic virus (WSMV), respectively (Table x.1). In cassava, an important
staple crop for example in East African countries, the Virus resistant Cassava for Africa (VIRCA)
project has been initiated to engineer resistance against two viral diseases [270]. In the example of
wheat, a polycistronic amiRNA precursor construct giving rise to five different amiRNAs targeting
WSMV genomic positions was designed using a naturally occurring miRNA precursor from rice, in
order to counteract resistance breaking by rapidly evolving viruses [269].
There are at least two cases of crop plants engineered for virus resistance using RNAi with regulatory
approv
al (Table 4.2). Plum resistant against plum pox virus (PPV) was developed by the US
Agricultural research Service (ARS; [169]) and gained approval in the US around 2010. The PPV
resistance trait has been shown to be stable over 15 years of field testing by natural aphid
transmission and by graft inoculations; the latter showed that the virus does not spread far into the
grafted wood but remains close to the graft site (reviewed in [169]). The second transgenic plant
passing regulatory approval in Brazil (2011) is a bean golden mosaic virus (BGMV) resistant common
bean [271, 272].
67
link to page 84 link to page 87
Small RNA-directed techniques
Host-induced gene silencing (HIGS) of fungi, insects and nematodes
Analogous to RNAi applications in viral protection, using HIGS RNAi constructs are expressed in plants
but target RNAs in pathogenic fungi, and bacteria, insects and nematodes [273]. Targeting plant
endogenous recessive resistance genes for engineering biotic resistance by RNAi needs knowledge
on potential target genes and plants with recessive genotypes need to retain agronomical
performance under low stress conditions, thus, engineering suitable candidate loci by RNAi is
chal enging. HIGS does not interfere with endogenous plant genes but requires knowledge on
candidate genes in respective plant pathogens whose downregulation can be induced and which are
central to the pathogen life cycle or survival. Published examples of HIGS in crop plants are listed in
Table 4.1.
A recent review including the concept of
in planta delivery of RNAi in nematode crop protection can
be found in Lilley
et al., 2012 [274]; one of the first studies in a major crop plant (soybean) using HIGS
was published in 2006. Both, economical y important root knot and cyst nematodes feeding on
transgenic RNAi plants were shown to be amenable to HIGS Table x.1; [275-277]). In insect control,
HIGS offers the potential to transgenically control also phloem feeders, such as aphids, which cannot
as efficiently be controlled as chewing type insects with
Bacil us thuringiensis (Bt) toxins [278]. In
wheat, HIGS of the carboxylesterase CbE E4 of the aphid
Sitobion avenae reduced progeny
production [279]. Additionally,
in vitro data suggest it may render
S. avenae more sensitive to
organophosphate, carbamate, and pyrethroid pesticides, since the orthologue of CbE E4 in another
aphid species has been shown to mediate pesticide resistance [279]. Proof of principle studies in crop
plants targeting insects started to be published around 2007, describing an engineered maize line
showing resistance against the western corn root worm [280]. Recently, it has been shown that HIGS
can also be exploited for fungal protection [281-283]. A specialized cell, the haustorium, formed by
biotrophic fungal pathogens is used for signal exchange and nutrient uptake, and is believed to also
mediate HIGS [283]. Novara
et al, generated a barley line targeting the
Blumeria graminis effector
protein avra10 which lead to reduction in fungal development [283]. Further examples used HIGS to
generate barley and wheat lines with resistance against
Fusarium graminearum [281-283].
Recently, US-EPA issued a registration note concerning a maize line (MON-87411-9) engineered via
RNAi to target an essential gene of the western corn root worm
(Table 4.2). The registration is valid
for 2 years for the purposes of agronomic evaluation, seed increase and production in breeding
nurseries (not for commercial planting).
Traits of RNAi-based plants are modified by targeted downregulation of desired genes. Examples of RNAi-
based crop plants in regard to altered quality traits (enhanced secondary metabolites, reduced allergen
68
link to page 84 link to page 87 link to page 84 link to page 87 link to page 69 link to page 87
Small RNA-directed techniques
potential) or abiotic stress (drought) and biotic stress tolerance (refer to
Table 4.1) have been published.
Furthermore, several RNAi-based GM plants have undergone successful regulatory approv
al (Table 4.2). Recent interest in RNAi-based GM plants has come up in regard to engineering biotic stress resistance,
however, the sRNA expressed from the transgene is targeted at viral RNA or RNAs expressed by plant
pathogens coming into contact with plants; the term host induced gene silencing (HIGS) is used for this
phenomenon. Proof of principle in engineering such traits has been shown for example by targeting
Fusarium in barley, the aphid
Sitobion avenae in wheat or nematodes in soy
(Table 4.1). A transgenic maize
line targeting the western corn root worm is at the moment analysed in field trials in the US, a plum and a
common bean line both engineered for resistance against viral diseases have passed regulatory approval in
the US and Brazil, respectively
(Table 4.2).
4.3 State of development
RNAi approaches have been used in research in order to deduce the function of downregulated
genes by observing the resultant phenotypes of plants. In plants, it has been the first method to
interfere in a targeted manner with genes of interest in species amenable to transformation.
To date, the main strategies to engineer transgenic plants using RNAi in plant research are the use of
artificial miRNAs (amiRNAs) and siRNA mediated RNAi (i.e. using constructs designed to result in
longer stretches of dsRNA molecules;
Fig. 4.3) to trigger silencing of target genes. Virus induced gene
silencing (VIGS) is a further alternative for transient downregulation of a GOI using viral vectors for
delivery incorporating fragments with complementarity to the target gene to be silenced [284, 285];
it is not covered further in this report. In the mid 1990ies reviews report on use of sense and
antisense suppression techniques in plant research and designate these accepted techniques for
gene expression manipulation [286-288]; in parallel and still ongoing is the functional
characterisation of the diverse RNAi pathways in plants. One of the first commercial plant lines used
RNAi technology, the FlavrSavrTM tomato in the US
(Table 4.2; [287]). The use of an intentionally
designed inverted repeat construct (also called hairpin construct) to induce silencing was reported in
1998, and targeted a GUS transgene in rice [289]. The wider use of amiRNAs in plants came after
publication of the Web MicroRNA Designer (WMD) in 2006 [240], and was first applied in a
monocotyledonous species, rice, in 2008 [235].
Determinants of effectiveness of RNAi approaches
The strength of target gene downregulation (expressivity) may range between partial to falling below
detection limit and is determined by a combination of the properties of the RNAi construct as well as
its functioning as transgene in the genomic context (e.g. location of integration) for a given
69
link to page 84 link to page 87
Small RNA-directed techniques
established plant line. The phenotypes of independent lines targeting a GOI may therefore form a
series of hypomorphic to loss of function phenotype individuals of which suitable candidate lines can
be chosen. This may be of advantage for research purposes, but potential y also for applied purposes
in case of exploiting genes with severe complete loss-of-function genotypes. A further potential
advantage of RNAi-based approaches in balancing negative effects of downregulation of endogenous
plant genes is the use of tissue specific promoters, which allow elimination of gene function in target
tissues, while gene function in remaining plant organs stays unaffected, or the potential of primarily
targeting splicing isoforms (in case of the use of amiRNAs [290]).
Inverted repeat (hairpin) constructs are used now widely as RNAi-based transgenes (see also
examples in
Table 4.1 and
Table 4.2). Early studies comparing different dsRNA constructs eliciting
siRNA mediated RNAi showed that inverted repeat constructs showed a high percentage of
independently transformed lines with gene silencing effects, whereas sense or antisense constructs,
as well as constructs concomitantly expressing a sense and an antisense RNA from two promoters
showed less penetrance [245, 291]. Inverted repeat constructs containing an intron as spacer
between the inverted repeat sequences seem to be especial y effective in eliciting RNAi [245].
For amiRNA design in plants, effectiveness to date is optimized by the pre-miRNA backbone chosen
for a given species, as well as on consideration of empirically determined parameters in relation to
for example sequence requirements and thermodynamic behaviour of miRNAs effecting their
processing, their incorporation into RISC and target recognition [238].
Determinants of specificity of RNAi approaches
The sequence of the RNA component functions as a guide to target RISC complexes to its targets.
However, even though plant miRNAs exhibit relatively high sequence complementarity to their
targets [292], perfect complementarity is not obligatory. Other factors contribute to ensure proper
functioning of RNAi pathways in the cellular context, of which, to date, there is too less information
to be included into algorithms for optimization of design of RNAi constructs. Optimization of
specificity, i.e. predicting and avoiding of off-targets, to date depends on the availability of
transcriptome sequence information, as well as on the available understanding of requirements on
specific miRNA/siRNA-target interaction [290].
For plant miRNAs sequence requirements have been studied. Information on experimental y
identified miRNA-target interactions, including experiments investigating miRNA-target from non-
target interactions [293], derived general patterns of miRNA-target interaction: for example, the 5´
region (~ position 2-12) of the miRNA tends to be mismatch sensitive, while the 3´region has more
relaxed constraints; more than two mismatches next to each other and mismatches at the position
flanking the cleavage site (10, 11) seem to be uncommon in the dataset of Schwab
et al., [293].
70
link to page 69
Small RNA-directed techniques
Similar observations have been obtained by evaluation of experimentally proven miRNA-target
duplexes for the distribution of mismatches, single-nucleotide bulges and G:U base pairs [294]. Such
patterns are used in the development of scoring matrices for prediction of miRNA targets and, in
turn, are also used to predict potential off-target activity of amiRNAs designed to target a gene of
interest (for example, Plant Small RNA Maker Site (P-SAMS; [295]), Web MicroRNA Designer (WMD;
[240])). WMD also incorporates hybridization energy in target recognition/off-target avoidance
calculations [240].
Although similar factors are thought to be guiding specificity of siRNA mediated RNAi, most of the
knowledge in plants is derived from studies of miRNA-target interaction (and/or transferred from
metazoan studies). In contrast to amiRNA mediated RNAi, siRNA mediated RNAi leads to formation of
a pool of distinct siRNAs
(Fig. 4.3), each of which potentially can trigger off-target effects and
production of secondary siRNAs. Furthermore, potentially, DCL proteins may cut at any site in the
dsRNA to produce siRNAs, giving rise to different pools of siRNAs from different copies of the dsRNA.
It has been shown, that perfect complementarity is not needed for siRNA mediated downregulation
in
N. benthamiana using a virus induced gene silencing approach [296]. In a transgenic
A. thaliana line expressing an antisense construct covering the coding sequence of an endogenous gene, off-
target effects were shown on its paralog, as well as on two genes sharing a 23 nucleotide stretch of
complete homology (however, remaining similarity of the genes to the target is not reported) [297].
Downregulation of candidates with a 21 nucleotide stretch of complete homology was not detected
(again, remaining similarity of the genes to the target is not reported), as well in genes with 21 or 22
nucleotide continuous identity but one mismatch (22 candidates) [297].
In practice, sequence based considerations are integrated into the design of siRNA mediated RNAi
constructs, and potential off-target candidates showing sequence similarity can be included in
experimental characterisation of established transgenic RNAi-based lines. These considerations can
be supported by programmes which incorporate stringency criteria derived from plant and/or
metazoan studies, however, due to the high number of potential y diced siRNAs stemming from a
particular dsRNA this may be challenging. A recent RNAi technique might facilitate this
computational problem in the future. MIGS, miRNA-induced gene silencing, exploits the pathway of
trans-acting small interfering RNAs (tasiRNAs) [298]. Certain plant miRNAs, for example miR173,
target so cal ed TAS transcripts that are converted into dsRNA by RdRP activity on the 3´ fragment
and processed into a phased tasiRNA pool [298, 299]. The tasiRNA pool thus may be predictable to a
certain extent.
Recent reviews detailing current considerations in design of siRNA mediated RNAi and
implementation of specificity can be found in [300, 301].
71
link to page 78
Small RNA-directed techniques
RNAi based techniques are firmly established as a basic research tool. Recent interest in RNAi-based GM
plants has come up with the realization of engineering plants with resistance against biotic stress by
targeting gene expression in the plant pathogen (host induced stress resistance). Furthermore, already on
the market are for example soy plants with modified oleic acid content or cultivars resistant against viral
disease. Regulators have been increasingly contacted with respect to specific questions concerning RNAi-
based GM plants. One central topic is the characterization of off-target effects of RNAi pathways, since
sRNAs may also lead to downregulation of non-target RNA showing partial complementarity. Several
parameters have been specified that determine specificity for plant miRNAs; at the EU level EFSA is
currently collecting scientific advice to inform on potential adaptations of risk assessment of RNAi-based
GM plants in the framework of Directive 2001/18/EC.
4.4 Intended and unintended effects
RNAi mediated downregulation of target genes is used in development of RNAi-based GM plants to
either (i) effect plant endogenous genes or (i ) effect gene expression/RNA molecules of plant pests.
The former may be used to engineer traits in respect to, among others, altered physiology,
nutritional content, agronomical traits, biotic and abiotic stress tolerance, whereas the latter is used
to confer biotic stress resistance to plants by targeting gene expression in plant pests or viral RNA
genomes. The latter is also subsumed under the term plant incorporated protectants (PIP) in the US
risk assessment framework.
A potential unintended effect which is discussed specifically in regard to RNAi-based GM plants is the
potential off-target effect of the miRNAs/siRNAs, which may lead to unintended downregulation of
endogenous plant genes, as wel as in the case of acting as a PIP, to unintended effects in non-target
organisms.
At the moment, specifics in regard to risk assessment of RNAi-based GM plants are discussed [302],
at the EU level by EFSA. Chapter
4.5 covers ongoing work at EFSA.
4.5 Safety considerations
The European Food Safety Authority (EFSA) developed guidelines for risk assessment (RA) of GM
plants, among other documents pertaining to food and feed use [303], to non-food/non-feed use
[304], to environmental risk assessment [5] as well as supporting guiding documents for example in
assessment of potential impacts on non-target organisms [305]. These documents provide guidance
on the specific provisions for submission dossiers for authorization of GM plants under Regulation
(EC) No. 1829/2003 on GM food and feed or under Directive 2001/18/EC on the deliberate release
72
link to page 79 link to page 79
Small RNA-directed techniques
into the environment. The majority of authorized GM plants internationally and in the EU are based
on transgenic plants expressing one or more novel proteins, however, commercial development of
RNAi-based GM plants is expected to increase due to its potential for example in engineering pest
resistance or altering crop composition [306].
To date, there is an ongoing process to evaluate and refine the RA framework for the specifics of
RNAi-based GM plants. The US Environmental Protection Agency (EPA) organized a Scientific
Advisory Panel Meeting in 2014 on “RNAi Technology as a Pesticide: Problem Formulation for Human
Health and Ecological Risk Assessment”.
7 In the same year, EFSA organized the scientific workshop
“Risk assessment considerations for RNAi-based GM plants” [307, 308] in order to formulate and
discuss specific features of RNAi-based GM plants. Building on that, in 2015, EFSA published a call for
a “Literature review of baseline information to support the risk assessment of RNAi-based GM
plants” (OC/EFSA/GMO/2015/01; OC/EFSA/GMO/2015/02) “… to obtain a comprehensive literature
overview on several of the risk assessment related issues identified during the EFSA´s workshop.”
Scientific baseline data present in the scientific literature in areas relevant to the molecular
characterization, the food and feed risk assessment and environmental risk assessment will be
collected and assessed. It will inform on potential future areas of research to close knowledge gaps
of importance to RA of RNAi-based GM plants and/or on potential adaptations to the current
framework of risk assessment of GM plants in regard to specifics of RNAi-based GM plants which may
be implemented into guidance documents in the future.
Below selected topics discussed during the EFSA workshop are described (a commentary has been
published by EFSA [308], the workshop documents can be found online
8 [307]), fol owed by the
specific tasks of information retrieval identified by EFSA and subject to the call for the literature
review on support for RA on RNAi-based GM plants (OC/EFSA/GMO/2015/01;
OC/EFSA/GMO/2015/02).
4.5.1 EFSA workshop on risk assessment considerations for RNAi-based GM plants
During the EFSA workshop breakout sessions, the fol owing key topics have been discussed [307,
308]:
Off-target activity in RNAi-based GM plants
RNAi-based GM plants carry either an amiRNA construct or a dsRNA construct (leading to formation
of siRNAs) to downregulate a target sequence and thereby modifying the desired trait. Unintended
off-target effects may arise (i) due to sufficient sequence homology to non-target genes of
7 Meeting minutes can be found at
http://www.epa.gov/sap/fifra-scientific-advisory-panel-meetings
8 http://www.efsa.europa.eu/de/events/event/140604
73
link to page 75
Small RNA-directed techniques
amiRNAs/siRNAs as well as, especially in the case of dsRNA expressing plants, (ii) due to uncertainty
of the generated pool of siRNAs, which may include secondary siRNAs.
The former problem may be addressed by bio-informatic approaches to identify possible off-target
genes. The applicability and the benefit to the overall risk assessment of this approach with available
knowledge to date has been discussed: (i) depending on the stringency of off-target prediction
criteria applied (see chapte
r 4.3 for general information), specificity and sensitivity estimates vary, (ii)
target prediction is also dependent on the presence and quality of genome/transcriptome sequence
information of the transformed plant cultivar, which may differ to reference genomes due to natural
genetic variation and/or breeding history. Taking into account that sRNA-mRNA interaction is based
on a short sequence length, bio-informatic approaches to date may lead to a large variation in off-
target gene candidates depending on criteria and genome sequence used, thereby may have limited
additional value to the RNAi-based GM risk assessment to date. However, progress in making more
reliable bio-informatic predictions of sRNA/mRNA recognition as well as the presence of suitable
genome (transcriptome) sequences, in the future may provide added benefit in guiding wel -
informed case specific endpoint analyses, in addition to generic comparative analyses in risk
assessment of GM plants.
Next generation sequencing methods may be used to characterize the sequences present in a siRNA
pool in a given RNAi-based GM plant versus its comparator; a question raised was the accuracy of the
methods in regard to answer questions to problems formulated during risk assessment of RNAi-base
GM plants.
Food/Feed risk assessment of RNAi based GM plants
The comparative approach used to verify the intended and identify unintended effects of the
established GM plant in regard to compositional, phenotypic and agronomic traits was considered to
be the appropriate approach also for RNAi-based GM plants. For compositional and nutritional
analyses, OECD consensus documents [309, 310] guide in selection and measurement of appropriate
key compounds for a given crop species for food/feed use. Case specific analyses are guided by the
intended effect of the introduced RNAi construct. As mentioned above, in the future, case specific
additional analyses in risk assessment in respect to compositional, phenotypic and agronomic traits
may be guided by information based on reliable bio-informatic predictions on potential off-target
candidates.
The study of Zhang
et al., [311] was debated at the EFSA workshop: the authors detected evidence of
plant miRNAs in pooled sera of humans with a predominant plant based diet; in a feeding study in
mice they established plant MIR168a presence in sera of mice fed a rice-based diet but not in mice
fed a control diet; finally, in a feeding study in mice they report biological activity of rice MIR168a:
74
Small RNA-directed techniques
decrease of low-density lipoprotein receptor adapter protein 1 (LDLRAP1) protein levels in mouse
plasma. A study, undertaken in collaboration with Monsanto researchers, replicating the
experiments with a special emphasis on the feeding regime could not find evidence for biological
activity on LDLRAP1 by dietary miRNAs [312] and postulated that compositional differences in the
feeding regime between control and MIR168a administered groups may explain the differences in
containing LDLRAP1 protein levels in the study by Zhang
et al. A study examining plant dietary sRNAs
in published 83 animal sRNA datasets [313] found presence of plant miRNAs in 63 datasets. The
highest plant miRNA level detected was 10 times lower than that of Zhang
et al., and datasets
showed high variation (including in experimental repetitions). The authors of this study, as well as
Tosar
et al., [314] - based on analyses of publicly available human sRNA datasets and datasets from
Zhang
et al., previous to their initial finding of dietary plant miRNAs-, argue that plant sRNAs present
in animal sRNA datasets may partly be due to methodological artefacts. A current review [315]
summarises that the majority of work spurred by the publication of Zhang
et al., [311] could not
corroborate their finding: although there is evidence of plant miRNAs in animal tissues in some
studies, levels, if detected, are low, calling into question a potential biological role. However, to find
scientific consensus on the topic of dietary plant miRNAs in the framework of RNAi-based GM plant
risk assessment, this topic is also reflected in the EFSA call for baseline data (see below).
Testing of RNAi molecules
per se in oral toxicity studies was not considered relevant at the EFSA
workshop [307, 308], based on (i) history of safe consumption of RNAi molecules naturally occurring
in plants and (ii) information from pharmaceutical studies on bioavailability, metabolism and
excretion.
Environmental risk assessment (ERA) of RNAi-based GM plants
A central topic discussed in breakout sessions were RNAi-based GM plants engineered to control
insect pests (by host-induced gene silencing (HIGS)), in the US subsumed under plants expressing
plant incorporated protectants (PIPs). An “area of concern” in the ERA is the “interaction of GM
plants with non-target organisms (NTO), including criteria for selection of appropriate species and
relevant functional groups” (Scientific Opinion on guidance for risk assessment of GM plants, EFSA,
[5]). There has been issued a supporting guidance document on this particular topic by EFSA [305]. In
this context, topics specific to RNAi-based GM risk assessment were discussed:
Exposure characterisation is an integral part of risk assessment which together with hazard
characterization leads to risk characterization. Barriers to exposure (including, for example,
degradation behaviour in soil, cellular uptake mechanisms in diverse species, sensitivity of diverse
species to ingested dsRNA) were discussed since they are valuable in facilitating and refining risk
assessment. It was concluded, that at present there is insufficient understanding on parameters of
75
Small RNA-directed techniques
specific barriers to make generalisations across taxa and to refine exposure estimates, and therefore,
at the moment, most reliable conclusions are derived from non-target organism toxicity studies.
Adverse effects are tested in a tiered manner (controlled laboratory studies progressing to more
realistic field conditions); for lower tier studies (laboratory conditions) there was a discussion on the
appropriate composition of test diets (dsRNA, sRNAs, plant material). Not all potential non-target
organisms can be tested, therefore, criteria for selection of appropriate test species have been
formulated (based on for example considerations of functional groups, ecological relevance). In the
future, in the presence of reliable sequence information on transcriptomes, bio-informatic analyses
may be used to support the selection of NTO for adverse effect testing, by concentrating on those
with genes sharing homology to the gene in the target species.
4.5.2 EFSA call on literature review to support risk assessment of RNAi-based GM
plants
As mentioned above, the EFSA workshop [305] helped identify key areas to be addressed to inform
on topics specific to RNAi-based GM plant risk assessment. To continue the process, a call on a
literature review collecting and assessing these key areas has been issued (OC/EFSA/GMO/2015/01;
OC/EFSA/GMO/2015/02).
Specifically, areas to collect and assess baseline information in the literature review to support the
molecular characterization of RNAi-based GM plants identified by EFSA are: (i) characterization and
distinctive features of mode-of-action of dsRNA and miRNA pathways in selected species/taxa, (ii)
current knowledge on off-target effects of siRNAs and miRNAs and assessment of bio-informatic
programmes available to predict off-target effects, and (iii) overview on current methodology to
determine siRNA pools in plants and summary on experimental information in the scientific literature
on descriptions of siRNA pools.
Areas to gather and assess data in respect to support the food/feed risk assessment of RNAi-based
GM plants and derived products are: (i) data on the pharmaco-kinetics profile of RNAi molecules in
humans and animals (primarily based on research and development data of RNAi molecules
developed for therapeutic use and for oral administration), (ii) effects of RNAi molecules on
gastrointestinal tract and annex glands on human and animals, (iii) information on barriers to
absorption of RNAi molecules in gastrointestinal tract and placenta of humans and animals, and (iv)
assessment of plausibility of effects of RNAi molecules on the immune system of humans and
animals.
Finally, areas to be analysed to support the environmental risk assessment are the following: (i) a
systematic literature search on the use of host-delivered RNAi molecules in arthropods, nematodes,
76
Small RNA-directed techniques
annelids and molluscs (reporting defined parameters and silencing effects) in order to assess if and
under which conditions siRNA and miRNAs delivered through feeding trigger RNAi in these
organisms, (ii) a review on mechanisms of dsRNA (siRNA/miRNA if relevant) uptake in arthropods,
nematodes, annelids and molluscs, (iii) a review plausible routes of exposure of the biotic and abiotic
environment to dsRNA (siRNA/miRNA if relevant) expressed in RNAi-based GM plants, its
environmental fate and barriers of exposure, (iv) based on collected information before, a summary
on information on which factors largely influence dsRNA (siRNA/miRNA if relevant) uptake in
arthropods, nematodes, annelids and mol uscs delivered by feeding, (v) assess plausibility and
mechanisms of unintended adverse effects on arthropods, nematodes, annelids and mol uscs by
cultivation of RNAi-based GM plants, and (vi) an overview on species belonging to arthropods,
nematodes, annelids and mol uscs for which complete or partial genome data are available.
The European Food Safety Authority (EFSA) developed guidelines for risk assessment (RA) of GM plants. To
date, there is an ongoing process to evaluate and refine the RA framework for the specifics of RNAi-based
GM plants according to the framework given by Directive 2001/13/EC and EFSA is soliciting scientific advice.
EFSA organized a scientific workshop in 2015, followed by a call for a “Literature review of baseline
information to support the risk assessment of RNAi-based GM plants” in 2015. Scientific baseline data
present in the scientific literature in areas relevant to the molecular characterization, the food and feed risk
assessment and environmental risk assessment will be collected and assessed. It will inform on potential
future areas of research to close knowledge gaps of importance to RA of RNAi-based GM plants and/or on
potential adaptations to the current framework of risk assessment of GM plants in regard to specifics of
RNAi-based GM plants which may be implemented into guidance documents in the future.
4.6 Detection and identification
Genomes of RNAi based GM plants contain a stably integrated transgene that in combination with its
genomic integration location can be used to develop an event-specific detection method for
identification. In case the transgenic construct contains elements often used in development of
GMOs these can be used for screening assays for detection purposes. Examples provide the event
specific identification methods for RNAi based GM plants soybean MON 87705 and soybean DP-
305423-1 listed in the GMOMETHODS database [316, 317].
4.7 Aspects of GMO classification
RNAi-based GM plants fall under the legal definition of GMO given in EU Directive 2001/18/EC.
77
Small RNA-directed techniques
4.8 Table
Table 4.1 Examples present in the scientific literature (or in development) of RNAi based transgenic crop plants with traits of interest for potential application in
plant breeding. Selected and extended from tables in Ricroch
et al., 2015 [318], Koch
et al., 2014 [273], Kamthan
et al., 2015 [319], Saurabh
et al., 2014 [320] and
Tiwari
et al., 2014 [239].
Crop
Conferred trait*
RNAi construct
References
Quality/nutritional traits
Potato
Enhanced amylopectin content
Antisense construct containing fragment of granule
The EFSA Journal (2006) 324, 1-20
bound starch synthase (GBSS)
BASF [253]
Rapeseed
Enhanced ß-carotene, zeaxanthin, violaxanthin
Inverted repeat construct containing fragment of
and lutein content in seeds
lycopene synthase
Yu
et al., 2008 [256]
Tomato
Enhanced carotenoid and flavonoid content
Inverted repeat constructs containing a partial sequence
tomato DE-ETIOLATED 1 (TDET1; regulatory protein)
Davuluri
et al. (2005) [255]
Wheat
Enhanced amylose content
Inverted repeat constructs containing fragments of
Starch branching enzyme IIa and IIb (SBE IIa, IIb)
Regina
et al. (2006) [254]
Rice
Reduced phytic acid content
Inverted repeat construct containing IPK1 (Inositol
1,3,4,5,6-pentakisphosphate 2-kinase)
Ali
et al., 2013 [257]
Apple
Reduced al ergenic potential (skin prick test,
Inverted repeat construct containing fragment of apple
Gilissen
et al., 2005
oral chal enge test)
al ergen Mal d 1
Dubois
et al., 2015 [258, 259]
Carrot
Reduced al ergenic potential (skin prick test)
Inverted repeat construct containing fragments of
carrot al ergens Dau c 1.01 and 1.02
Peters
et al., 2011 [260]
Inverted repeat construct containing fragments from α-, Gil-Humanes
et al. 2010,
Wheat
Reduced gliadin content, retained potential for
good bread baking quality
γ-, and ω-gladins
Gil-Humanes
et al. 2014
Barro
et al., 2016 [261-263]
Abiotic stress tolerance
Inverted repeat construct with partial sequence of
Wang
et al., 2009
Canola
Drought tolerance in field trials
(reduced transpiration rate)
farnesyl-transferase (negative regulator of abscisic acid
Performance Plants, Inc. Canada
(ABA) signaling)
Waltz
et al., 2014 [251, 252]
Drought tolerance in field trials
Habben
et al. 2014
Corn
(improved kernel set at dry conditions by
Inverted repeat construct with partial sequence of ACC
Dupont/Pioneer, USA
reduction of anthesis-silking interval (ASI))
synthase 6 (ACS6; involved in ethylene biosynthesis)
Waltz
et al., 2014 [249, 251]
Potato
Drought tolerance in greenhouse conditions
amiRNA (
Arabidopsis miR319a backbone) construct
(reduced transpiration rate)
targeting cap-binding protein 80 (CBP80; negative
Pieczynski
et al., 2013 [250]
78
Small RNA-directed techniques
Crop
Conferred trait*
RNAi construct
References
regulator of abscisic acid (ABA) signaling)
Biotic stress resistance: viral pathogens
Barley
Barley yel ow dwarf virus (BYDV) resistance
Inverted repeat construct containing sequence of BYDV-
polymerase
Wang
et al., 2000 [267]
Project: Virus-Resistant Cassava for Africa
Inverted repeat constructs targeting coat protein (CP)
Cassava
(VIRCA)
Cassava brown streak disease (CBSD) resistance region of CBSD virus strains and AC1,2 genes (involved
Taylor
et al., 2012 [270]
Cassava mosaic disease (CMD) resistance
in viral genome replication) in case of CMD virus strains
amiRNA construct (
Arabidopsis miR159a backbone)
Tomato
Cucumber mosaic virus (CMV) resistance
targeting viral RdRP 2a/2b transcripts or conserved
Zhang
et al., 2011 [268]
3´UTR region of virus
Wheat
Wheat streak mosaic virus (WSMV) resistance
amiRNA construct (rice multiplex miR395 backbone)
targeting 5 viral genome locations
Fahim
et al., 2012 [269]
Biotic stress resistance: fungal pathogens
Blumeria graminis resistance
Barley
(reduced fungal development in the absence of HIGS of avra10 (putative
Bg effector proteins) by
the matching barley resistance gene
Mla10)
inverted repeat construct
Nowara
et al., 2010 [283]
HIGS of
Fg CYP51A, CYP51B and CYP51 by sense and
Barley
Fusarium graminearum resistance
antisense driven transcription of chimeric fragment
Koch
et al., 2013 [282]
harbouring partial sequences of al three genes
Wheat
Fusarium graminearum resistance
HIGS of
Fg chitin synthase (Chs) 3b by inverted repeat
construct
Cheng
et al., 2015 [281]
Downregulation of endogenous recessive resistance
Blumeria graminis f. sp. tritici
Wheat
resistance
gene
TaS3 (
Triticum aestivum susceptibility 3) using a
(transient expression experiment)
partial fragment against
TaS3 in an inverted repeat
Li
et al., 2013 [248]
construct
Biotic stress resistance: bacterial pathogens
Xanthomonas oryzae pv oryzae
Downregulation of endogenous recessive resistance
Rice
resistance
(bacterial blight)
gene Os-11N3 using a partial fragment against Os-11N3
Antony
et al., 2010 [247]
in an inverted repeat construct
Biotic stress resistance: insects
Diabrotica virgifera
Maize
resistance
HIGS of V-ATPase A using construct containing gene
(reduction in root damage)
fragment in an inverted repeat construct
Baum
et al., 2007 [280]
Sitobion avenae resistance
HIGS of CbE E4 (carboxylesterase) using construct
Wheat
(reduced progeny production and reduced
containing gene fragment in an inverted repeat
Xu
et al., 2014 [279]
resistance to phoxim insecticide)
construct
79
Small RNA-directed techniques
Crop
Conferred trait*
RNAi construct
References
Biotic stress resistance: nematodes
Heterodera glycines resistance
Soy
(development of soybean cyst nematode (SCN)
HIGS of MSP (major sperm protein) using inverted
Steeves
et al., 2006 [276]
females and number of eggs per cyst were
repeat construct
reduced)
Heterodera glycines resistance
Soy
(decrease in the number of mature SCN
HIGS of HgALD (aldolase) using inverted repeat
females)
construct; hairy root system
Youssef
et al. 2013 [277]
Meloidogyne incognita resistance
HIGS of Mi-Rpn7 (essential for the integrity of 26S
Soy
(reduced number of egg mass and egg number; proteasome) using inverted repeat construct; hairy root Niu
et al., 2012 [275]
no complete resistance)
system
*Conferred traits were described in more detail in some listed examples; where not further defined, conferred traits may be quantitative in nature (f.e. resistance)
and for the exact trait expression please refer to the cited literature. HIGS: host induced gene silencing.
80
Small RNA-directed techniques
Table 4.2 RNAi based transgenic crops which have been evaluated by regulatory agencies and have been approved for commercial purposes or +agronomic
evaluation
Species
Trait
Transgene
Agency
Developer
biotic stress resistance
traits
Plum
Plum pox virus resistance
Inverted repeat sequence of PPV coat protein
USA:
US Department of Agriculture
(Event C5;
(PPV)
driven by 35S promoter
Determination of Non-regulated
(USDA) Agricultural Research
`Honeysweet´)
(Scorza
et al., 2013)
status by APHIS, USA 2007**
Service (ARS) in cooperation with
US-FDA completed review 2009*
Research Institutes in Europe
US-EPA registration 2010 §
Common
Bean golden mosaic virus
Inverted repeat sequence of fragment of rep
Brazil:
Embrapa, Brazilian Agricultural
Bean
(BGMV) resistance
gene (AC1) of BGMV, driven by CaMV35S
Regulatory approval for food, feed
Research Corporation
(EMBRAPA
promoter
and cultivation 2011*,§§
5.1)
(Aragao
et al., 2013)
(Aragao
et al., 2009)
Maize +
Diabrotica virgifera
Inverted repeat sequence of fragment of the
USA:
Monsanto
MON-87411-9
virgifera (Western corn
WCR Snf7 gene, driven by 35S promoter
Determination of Non-regulated
rootworm (WCR))
status by APHIS, USA 2015**
resistance
US-FDA completed review 2014*
US-EPA registration 2015 for
agronomic evaluation (not
authorised for commercial purposes)
+
quality traits
Potato
impaired black spot bruise Chimeric construct consisting of 3’-untranslated
USA:
J.R. Simplot Company, USA
development
sequence of the polyphenol oxidase-5 gene
Determination of Non-regulated
InnateTM
(Ppo5) and a fragment of the asparagine
status by APHIS, USA, 2014**
potatoes 1st
impaired asparagine
synthetase-1 (Asn1) gene
generation
(Asn1) and reducing
US-FDA completed review 2015***
sugar formation (pPhL,
Chimeric construct consisting consisting of
for events in bold
Events E12,
pR1 ) which leads to low
fragment of promoter for the potato
E24, F10, F37, acrylamide content upon
phosphorylase-L (pPhL) gene and a fragment of
J3, J55, J78,
heat treatment (frying,
promoter for the potato R1 gene (pR1)
G11, H37, H50 baking, cooking)
Both designed as inverted repeat genes, each
driven by two convergent S. tuberosum
endogenous promoters special y active in tubers
81
Small RNA-directed techniques
Species
Trait
Transgene
Agency
Developer
Apple
impaired enzymatic
Suppression of four polyphenol oxidase genes
USA:
Okanagan Specialty Fruits Inc,
browning of apple flesh
PPO2, GPO3, APO5, pSR7
Determination of Non-regulated
Canada
ArcticTM Apple after slicing or bruising
status by APHIS, USA, 2015**
Events GD743,
Partial sequences, expressed together in sense
US-FDA completed review 2015***
GS784
orientation (chimeric sense-silencing RNA) by 35S Canada:
promoter
Health Canada: approved product for
sale and growth as GM Food 2015 *,
#
Alfalfa
Reduced lignin content
Partial sequence of caffeoyl CoA
USA:
Monsanto;
KK179
which al ows greater
3-
O-methyltransferase (CCOMT) designed as
Determination of Non-regulated
Forage Genetics International,
flexibility in harvest
inverted repeat, driven by
status by APHIS, USA, 2014**
USA
timing; high lignin content
US-FDA completed review for use in
affects quality negatively
animal feed 2013***
Soybean
increased oleic acid and
Partial sequences of fatty acid desaturase (fad2-
EU:
Monsanto
MON 87705
reduced linoleic acid
1A) and palmitoyl acyl carrier protein
Authorisation for use as/in Food and
(Vistive
content, which confers
thioesterase (FATB1-A) genes; designed after
Feed 2015 ###
GoldTM)
higher oxidative stability
genomic integration as chimeric inverted repeat
USA:
of the oil
construct, driven by a seed specific promoter
Determination of Non-regulated
from soybean
status by APHIS, USA, 2011**
US-FDA completed review 2011***
Soybean
increased oleic acid and
Partial sequence of endogenous fatty acid
EU:
DuPont Pioneer
DP 305423
reduced linoleic acid
desaturase (fad2-1), designed to silence the
Authorisation for use as/in Food and
(Plenish Soy)
content, which confers
expression of the endogenous fad2-1gene, driven Feed 2015 ###
higher oxidative stability
by an endogenous soybean promoter
USA:
of the oil
preferential y active in seed tissue
Determination of Non-regulated
status by APHIS, USA, 2010**
US-FDA completed review 2009***
Tomato
Decreased cel wal
Endogenous polygalacturonase gene driven by
USA:
Calgene, USA
FlavrSavrTM
breakdown which confers the 35SCaMV promoter in reverse orientation
Determination of Non-regulated
longer shelf life;
status by APHIS, USA, 1992**
processed tomatoes with
US-FDA completed review 1994*
higher serum viscosity
82
Small RNA-directed techniques
Listed RNAi plant lines may contain further transgenes to confer additional traits (for example herbicide resistance of MON87705), described are only traits based
on an RNAi transgene. Listed RNAi plant lines may have gone through regulatory approval in further countries.
*
Center for Environmental Risk Assessment (CERA)
(http://www.cera-gmc.org)
**
Petitions for Determination of Nonregulated Status Database, US Department of Agriculture (USDA) Animal and Plant Health Inspection Service (APHIS):
https://www.aphis.usda.gov/biotechnology/petitions_table_pending.shtml
***
US-FDA Inventory on Biotechnology Consultations on Food from GE Plant Varieties:
http://www.accessdata.fda.gov/scripts/fdcc/?set=Biocon
#
Health Canada, Novel Food Decisions:
http://www.hc-sc.gc.ca/fn-an/gmf-agm/appro/index-eng.php
##
FSANZ Food Standards Code – Standard 1.5.2 – Food produced using Gene Technology
https://www.comlaw.gov.au/Series/F2008B00628/Compilations
### EU Register of authorised GMOs
http://ec.europa.eu/food/dyna/gm_register/index_en.cfm
§
US Environmental Protection Agency (EPA) Plant Incorporated Protectant (PIP) registrations:
http://www.epa.gov/regulation-biotechnology-under-tsca-
and-fifra/overview-plant-incorporated-protectants
§§
ISAAA, International Service for the Acquisition of Agri-Biotech Applications, GM Approval Database:
http://www.isaaa.org/gmapprovaldatabase/default.asp
83
5 Abbreviations
ABA abscisic acid
ALS acetolactate synthase
ALSV Apple Latent Spherical Virus
amiRNA artificial microRNA
APHIS Animal and Plant Health Inspection Service (USA)
CaMV 35S promoter Cauliflower Mosaic Virus 35S promoter
Cas CRISPR associated
CRISPR Clustered regularly interspaced short palindromic repeats
crRNA CRISPR RNA
ds double stranded
DSB double strand break
EFSA European Food Safety Authority
EMS ethyl methanesulfonate
EPA Environmental Protection Agency (USA)
ERA environmental risk assessment
FDA Food and Drug Administration (USA)
GM genetically modified
GMO genetically modified organism
GOI gene of interest
gRNA guide RNA
HDR homology directed repair
HIGS host induced gene silencing
HSP heat shock promoter
indel insertion-deletion mutation
IR inverted repeat
LG linkage group
MAS marker assisted selection
miRNA micro RNA
MN meganuclease
NHEJ non-homologous end joining
nt nucleotide
NTO non-target organisms
84
NTWG New Techniques Working Group
PCR polymerase chain reaction
PIP plant incorporated protectants
PTGS post-transcriptional gene silencing
PPV Plum Pox Virus
QTL quantitative trait locus
SDN site directed nuclease
RA risk assessment
RISC RNA induced silencing complex
RNAi RNA interference
sgRNA single guide RNA
siRNA small inhibitory RNA
SNP single nucleotide polymorphism
sRNA smal RNA
ss single strand
TALEN transcription activator-like effector nuclease
TGS transcriptional gene silencing
TILLING Targeting Induced Local Lesions in Genomes
tracrRNA
trans-encoded crRNA
TRV tobacco rattle virus
USDA United States Department of Agriculture
USEPA United States Environmental Protection Agency
VIGE Viral induced gene expression
VIGS Viral induced gene silencing
ZFN zinc finger nuclease
ZKBS Zentrale Kommission für Biologische Sicherheit/Central Commission for biological Safety
85
6 References
All links in the document were last accessed on 12.08.2016.
[1] Becker, H.,
Pflanzenzüchtung, Ulmer, Stuttgart 2011.
[2] Lusser M, Parisi C, Plan D, E, R.-C.,
New plant breeding techniques, Joint Research Centre — Institute for
Prospective Technological Studies, Luxemburg 2011.
[3] AGES,
Cisgenesis - A report on the practical consequences of the application of novel techniques in plant
breeding, Bundesministerium für Gesundheit, Wien 2012.
[4] AGES,
New plant breeding techniques. RNA-dependent methylation, Reverse breeding, Grafting,
Bundesministerium für Gesundheit, Wien 2013.
[5] EFSA, Scientific Opinion on Guidance on the environmental risk assessment of genetically modified
plants.
EFSA Journal 2010,
8.
[6] Fladung, M., Cibus' herbicide-resistant canola in European limbo.
Nat Biotech 2016,
34, 473-474.
[7] ZKBS
, Position statement of the ZKBS on new plant breeding techniques, 2012.
[8] Broeders, S. R., De Keersmaecker, S. C., Roosens, N. H., How to deal with the upcoming chal enges in
GMO detection in food and feed.
Journal of biomedicine & biotechnology 2012,
2012, 402418.
[9] Angers-Loustau, A., Petrillo, M., Bonfini, L., Gatto, F.
, et al., JRC GMO-Matrix: a web application to
support Genetical y Modified Organisms detection strategies.
BMC bioinformatics 2014,
15, 417.
[10] Parry, G., Patron, N., Bastow, R., Matthewman, C., Meeting report: GARNet/OpenPlant CRISPR-Cas
workshop.
Plant methods 2016,
12, 6.
[11] Huang, S., Weigel, D., Beachy, R. N., Li, J., A proposed regulatory framework for genome-edited crops.
Nature genetics 2016,
48, 109-111.
[12] Whelan, A. I., Lema, M. A., Regulatory Framework for Gene Editing and other New Breeding
Techniques (NBTs) in Argentina.
GM crops & food 2015, 0.
[13] Jinek, M., Chylinski, K., Fonfara, I., Hauer, M.
, et al., A programmable dual-RNA-guided DNA
endonuclease in adaptive bacterial immunity.
Science (New York, N.Y.) 2012,
337, 816-821.
[14] Sander, J. D., Joung, J. K., CRISPR-Cas systems for editing, regulating and targeting genomes.
Nature
biotechnology 2014,
32, 347-355.
[15] Rath, D., Amlinger, L., Rath, A., Lundgren, M., The CRISPR-Cas immune system: Biology, mechanisms
and applications.
Biochimie 2015,
117, 119-128.
[16] Hsu, P. D., Lander, E. S., Zhang, F., Development and applications of CRISPR-Cas9 for genome
engineering.
Cell 2014,
157, 1262-1278.
[17] Osakabe, Y., Osakabe, K., Genome editing with engineered nucleases in plants.
Plant & cel physiology
2015,
56, 389-400.
[18] Belhaj, K., Chaparro-Garcia, A., Kamoun, S., Nekrasov, V., Plant genome editing made easy: targeted
mutagenesis in model and crop plants using the CRISPR/Cas system.
Plant methods 2013,
9, 39.
[19] Sorek, R., Lawrence, C. M., Wiedenheft, B., CRISPR-mediated adaptive immune systems in bacteria and
archaea.
Annual review of biochemistry 2013,
82, 237-266.
[20] Belhaj, K., Chaparro-Garcia, A., Kamoun, S., Patron, N. J., Nekrasov, V., Editing plant genomes with
CRISPR/Cas9.
Current opinion in biotechnology 2015,
32, 76-84.
[21] Knoll, A., Fauser, F., Puchta, H., DNA recombination in somatic plant cells: mechanisms and
evolutionary consequences.
Chromosome research : an international journal on the molecular,
supramolecular and evolutionary aspects of chromosome biology 2014,
22, 191-201.
[22] Cermak, T., Baltes, N. J., Cegan, R., Zhang, Y., Voytas, D. F., High-frequency, precise modification of the
tomato genome.
Genome biology 2015,
16, 232.
[23] Yin, K., Han, T., Liu, G., Chen, T.
, et al., A geminivirus-based guide RNA delivery system for CRISPR/Cas9
mediated plant genome editing.
Scientific reports 2015,
5, 14926.
[24] Ali, Z., Abul-faraj, A., Li, L., Ghosh, N.
, et al., Efficient Virus-Mediated Genome Editing in Plants Using
the CRISPR/Cas9 System.
Molecular Plant 2015,
8, 1288-1291.
86
[25] Ali, Z., Abul-Faraj, A., Piatek, M., Mahfouz, M. M., Activity and specificity of TRV-mediated gene editing
in plants.
Plant signaling & behavior 2015,
10, e1044191.
[26] Marton, I., Zuker, A., Shklarman, E., Zeevi, V.
, et al., Nontransgenic genome modification in plant cells.
Plant physiology 2010,
154, 1079-1087.
[27] Woo, J. W., Kim, J., Kwon, S. I., Corvalan, C.
, et al., DNA-free genome editing in plants with
preassembled CRISPR-Cas9 ribonucleoproteins.
Nature biotechnology 2015.
[28] Mahfouz, M. M., Li, L., TALE nucleases and next generation GM crops.
GM crops 2011,
2, 99-103.
[29] Bae, S., Park, J., Kim, J. S., Cas-OFFinder: a fast and versatile algorithm that searches for potential off-
target sites of Cas9 RNA-guided endonucleases.
Bioinformatics (Oxford, England) 2014,
30, 1473-1475.
[30] Lei, Y., Lu, L., Liu, H. Y., Li, S.
, et al., CRISPR-P: a web tool for synthetic single-guide RNA design of
CRISPR-system in plants.
Mol Plant 2014,
7, 1494-1496.
[31] Lusser, M., Parisi, C., Plan, D., Rodriguez-Cerezo, E., Deployment of new biotechnologies in plant
breeding.
Nature biotechnology 2012,
30, 231-239.
[32] Pacher, M., Schmidt-Puchta, W., Puchta, H., Two unlinked double-strand breaks can induce reciprocal
exchanges in plant genomes via homologous recombination and nonhomologous end joining.
Genetics
2007,
175, 21-29.
[33] Ali, Z., Abulfaraj, A., Idris, A., Ali, S.
, et al., CRISPR/Cas9-mediated viral interference in plants.
Genome
biology 2015,
16, 238.
[34] Baltes, N. J., Hummel, A. W., Konecna, E., Cegan, R.
, et al., Conferring resistance to geminiviruses with
the CRISPR–Cas prokaryotic immune system.
Nature Plants 2015,
1, 15145.
[35] Ji, X., Zhang, H., Zhang, Y., Wang, Y., Gao, C., Establishing a CRISPR–Cas-like immune system conferring
DNA virus resistance in plants.
Nature Plants 2015,
1, 15144.
[36] La Russa, M. F., Qi, L. S., The New State of the Art: Cas9 for Gene Activation and Repression.
Molecular
and cel ular biology 2015,
35, 3800-3809.
[37] Shukla, V. K., Doyon, Y., Mil er, J. C., DeKelver, R. C.
, et al., Precise genome modification in the crop
species Zea mays using zinc-finger nucleases.
Nature 2009,
459, 437-441.
[38] Siritunga, D., Sayre, R., Generation of cyanogen-free transgenic cassava.
Planta 2003,
217, 367-373.
[39] Vollmann, J., Eynck, C., Camelina as a sustainable oilseed crop: Contributions of plant breeding and
genetic engineering.
Biotechnology Journal 2015,
10, 525-535.
[40] Sedbrook, J. C., Phippen, W. B., Marks, M. D., New approaches to facilitate rapid domestication of a
wild plant to an oilseed crop: Example pennycress (Thlaspi arvense L.).
Plant Science 2014,
227, 122-132.
[41] Harper, A. L., Trick, M., Higgins, J., Fraser, F.
, et al., Associative transcriptomics of traits in the polyploid
crop species Brassica napus.
Nature biotechnology 2012,
30, 798-802.
[42] Tian, B., Wei, F., Shu, H., Zhang, Q.
, et al., Decreasing erucic acid level by RNAi-mediated silencing of
fatty acid elongase 1 (BnFAE1.1) in rapeseeds (Brassica napus L.).
African Journal of Biotechnology 2011,
10,
13194-13201.
[43] Lee, K.-R., In Sohn, S., Jung, J. H., Kim, S. H.
, et al., Functional analysis and tissue-differential expression
of four FAD2 genes in amphidiploid Brassica napus derived from Brassica rapa and Brassica oleracea.
Gene
2013,
531, 253-262.
[44] Nour-Eldin, H. H., Andersen, T. G., Burow, M., Madsen, S. R.
, et al., NRT/PTR transporters are essential
for translocation of glucosinolate defence compounds to seeds.
Nature 2012,
488, 531-534.
[45] Ozuna, C. V., Iehisa, J. C., Gimenez, M. J., Alvarez, J. B.
, et al., Diversification of the celiac disease alpha-
gliadin complex in wheat: a 33-mer peptide with six overlapping epitopes, evolved following
polyploidization.
The Plant journal : for cel and molecular biology 2015,
82, 794-805.
[46] Herman, E. M., Helm, R. M., Jung, R., Kinney, A. J., Genetic modification removes an immunodominant
allergen from soybean.
Plant physiology 2003,
132, 36-43.
[47] Vol mann, J., Rajcan, I., Oil Crop Breeding and Genetics, in: Vol mann, J., Rajcan, I. (Eds.),
Oil Crops,
Springer-Verlag New York 2010, p. 548.
[48] Zraidi, A., Pachner, M., Lel ey, T., Obermayer, R.,
Cucurbit Genetics Cooperative Report 2003, pp. 57-61.
[49] Balkunde, R., Pesch, M., Hulskamp, M., Trichome patterning in
Arabidopsis thaliana from genetic to
molecular models.
Current topics in developmental biology 2010,
91, 299-321.
87
[50] Wester, K., Digiuni, S., Geier, F., Timmer, J.
, et al., Functional diversity of R3 single-repeat genes in
trichome development.
Development (Cambridge, England) 2009,
136, 1487-1496.
[51] Furstenberg-Hagg, J., Zagrobelny, M., Bak, S., Plant defense against insect herbivores.
International
journal of molecular sciences 2013,
14, 10242-10297.
[52] Kivimaki, M., Karkkainen, K., Gaudeul, M., Loe, G., Agren, J., Gene, phenotype and function:
GLABROUS1 and resistance to herbivory in natural populations of
Arabidopsis lyrata.
Molecular ecology
2007,
16, 453-462.
[53] Riddick, E. W., Simmons, A. M., Do plant trichomes cause more harm than good to predatory insects?
Pest management science 2014,
70, 1655-1665.
[54] Liu, B., Zhu, Y., Zhang, T., The R3-MYB gene GhCPC negatively regulates cotton fiber elongation.
PloS
one 2015,
10, e0116272.
[55] Nukumizu, Y., Wada, T., Tominaga-Wada, R., Tomato (Solanum lycopersicum) homologs of
TRIPTYCHON (SlTRY) and GLABRA3 (SlGL3) are involved in anthocyanin accumulation.
Plant signaling &
behavior 2013,
8, e24575.
[56] Vendramin, E., Pea, G., Dondini, L., Pacheco, I.
, et al., A unique mutation in a MYB gene cosegregates
with the nectarine phenotype in peach.
PloS one 2014,
9, e90574.
[57] Zhou, L., Zheng, K., Wang, X., Tian, H.
, et al., Control of trichome formation in
Arabidopsis by poplar
single-repeat R3 MYB transcription factors.
Frontiers in plant science 2014,
5, 262.
[58] Li, Q., Cao, C., Zhang, C., Zheng, S.
, et al., The identification of Cucumis sativus Glabrous 1 (CsGL1)
required for the formation of trichomes uncovers a novel function for the homeodomain-leucine zipper I
gene.
Journal of experimental botany 2015,
66, 2515-2526.
[59] Pan, Y., Bo, K., Cheng, Z., Weng, Y., The loss-of-function GLABROUS 3 mutation in cucumber is due to
LTR-retrotransposon insertion in a class IV HD-ZIP transcription factor gene CsGL3 that is epistatic over
CsGL1.
BMC plant biology 2015,
15, 302.
[60] Schilmiller, A. L., Last, R. L., Pichersky, E., Harnessing plant trichome biochemistry for the production of
useful compounds.
The Plant journal : for cel and molecular biology 2008,
54, 702-711.
[61] Dayan, F. E., Duke, S. O., Trichomes and root hairs: natural pesticide factories.
Pesticide Outlook 2003,
14, 175-178.
[62] Duke, S. O., Canel, C., Rimando, A. M., Tel ez, M. R.
, et al., Current and potential exploitation of plant
glandular trichome productivity.
Advances in Botanical Research Incorporating Advances in Plant Pathology,
Vol 31 2000 2000,
31, 121-151.
[63] Dubos, C., Le Gourrierec, J., Baudry, A., Huep, G.
, et al., MYBL2 is a new regulator of flavonoid
biosynthesis in
Arabidopsis thaliana.
The Plant journal : for cel and molecular biology 2008,
55, 940-953.
[64] Jun, J. H., Liu, C., Xiao, X., Dixon, R. A., The Transcriptional Repressor MYB2 Regulates Both Spatial and
Temporal Patterns of Proanthocyandin and Anthocyanin Pigmentation in Medicago truncatula.
The Plant
cell 2015,
27, 2860-2879.
[65] Matsui, K., Umemura, Y., Ohme-Takagi, M., AtMYBL2, a protein with a single MYB domain, acts as a
negative regulator of anthocyanin biosynthesis in
Arabidopsis.
The Plant journal : for cel and molecular
biology 2008,
55, 954-967.
[66] Shi, M. Z., Xie, D. Y., Biosynthesis and metabolic engineering of anthocyanins in
Arabidopsis thaliana.
Recent patents on biotechnology 2014,
8, 47-60.
[67] Pavan, S., Jacobsen, E., Visser, R. G. F., Bai, Y., Loss of susceptibility as a novel breeding strategy for
durable and broad-spectrum resistance.
Molecular Breeding 2010,
25, 1-12.
[68] Zhou, J., Peng, Z., Long, J., Sosso, D.
, et al., Gene targeting by the TAL effector PthXo2 reveals cryptic
resistance gene for bacterial blight of rice.
The Plant journal : for cel and molecular biology 2015,
82, 632-
643.
[69] van Schie, C. C., Takken, F. L., Susceptibility genes 101: how to be a good host.
Annual review of
phytopathology 2014,
52, 551-581.
[70] Li, T., Liu, B., Spalding, M. H., Weeks, D. P., Yang, B., High-efficiency TALEN-based gene editing
produces disease-resistant rice.
Nature biotechnology 2012,
30, 390-392.
[71] Hutin, M., Perez-Quintero, A. L., Lopez, C., Szurek, B., MorTAL Kombat: the story of defense against TAL
effectors through loss-of-susceptibility.
Frontiers in plant science 2015,
6, 535.
88
[72] Aflitos, S., Schijlen, E., de Jong, H., de Ridder, D.
, et al., Exploring genetic variation in the tomato
(Solanum section Lycopersicon) clade by whole-genome sequencing.
The Plant journal : for cel and
molecular biology 2014,
80, 136-148.
[73] Li, J. Y., Wang, J., Zeigler, R. S., The 3,000 rice genomes project: new opportunities and chal enges for
future rice research.
GigaScience 2014,
3, 8.
[74] Mace, E. S., Tai, S., Gilding, E. K., Li, Y.
, et al., Whole-genome sequencing reveals untapped genetic
potential in Africa's indigenous cereal crop sorghum.
Nature communications 2013,
4, 2320.
[75] Zhou, Z., Jiang, Y., Wang, Z., Gou, Z.
, et al., Resequencing 302 wild and cultivated accessions identifies
genes related to domestication and improvement in soybean.
Nature biotechnology 2015,
33, 408-414.
[76] Qi, J., Liu, X., Shen, D., Miao, H.
, et al., A genomic variation map provides insights into the genetic basis
of cucumber domestication and diversity.
Nature genetics 2013,
45, 1510-1515.
[77] Araus, J. L., Cairns, J. E., Field high-throughput phenotyping: the new crop breeding frontier.
Trends in
plant science 2014,
19, 52-61.
[78] Fiorani, F., Schurr, U., Future scenarios for plant phenotyping.
Annual review of plant biology 2013,
64,
267-291.
[79] Rhodes, D. H., Hoffmann, L., Jr., Rooney, W. L., Ramu, P.
, et al., Genome-wide association study of
grain polyphenol concentrations in global sorghum [Sorghum bicolor (L.) Moench] germplasm.
Journal of
agricultural and food chemistry 2014,
62, 10916-10927.
[80] Ruggieri, V., Francese, G., Sacco, A., D'Alessandro, A.
, et al., An association mapping approach to
identify favourable alleles for tomato fruit quality breeding.
BMC plant biology 2014,
14, 337.
[81] Sacco, A., Ruggieri, V., Parisi, M., Festa, G.
, et al., Exploring a Tomato Landraces Collection for Fruit-
Related Traits by the Aid of a High-Throughput Genomic Platform.
PloS one 2015,
10, e0137139.
[82] Yang, W., Guo, Z., Huang, C., Duan, L.
, et al., Combining high-throughput phenotyping and genome-
wide association studies to reveal natural genetic variation in rice.
Nature communications 2014,
5, 5087.
[83] Yong, H. Y., Wang, C., Bancroft, I., Li, F.
, et al., Identification of a gene controlling variation in the salt
tolerance of rapeseed (Brassica napus L.).
Planta 2015,
242, 313-326.
[84] Svitashev, S., Young, J. K., Schwartz, C., Gao, H.
, et al., Targeted Mutagenesis, Precise Gene Editing, and
Site-Specific Gene Insertion in Maize Using Cas9 and Guide RNA.
Plant physiology 2015,
169, 931-945.
[85] Chaparro-Garcia, A., Kamoun, S., Nekrasov, V., Boosting plant immunity with CRISPR/Cas.
Genome
biology 2015,
16, 254.
[86] Qi, L. S., Larson, M. H., Gilbert, L. A., Doudna, J. A.
, et al., Repurposing CRISPR as an RNA-guided
platform for sequence-specific control of gene expression.
Cell 2013,
152, 1173-1183.
[87] Gilbert, L. A., Larson, M. H., Morsut, L., Liu, Z.
, et al., CRISPR-mediated modular RNA-guided regulation
of transcription in eukaryotes.
Cell 2013,
154, 442-451.
[88] Thakore, P. I., D'Ippolito, A. M., Song, L., Safi, A.
, et al., Highly specific epigenome editing by CRISPR-
Cas9 repressors for silencing of distal regulatory elements.
Nat Meth 2015,
advance online publication.
[89] Piatek, A., Ali, Z., Baazim, H., Li, L.
, et al., RNA-guided transcriptional regulation
in planta via synthetic
dCas9-based transcription factors.
Plant biotechnology journal 2015,
13, 578-589.
[90] Bortesi, L., Fischer, R., The CRISPR/Cas9 system for plant genome editing and beyond.
Biotechnology
Advances 2015,
33, 41-52.
[91] Feng, Z., Zhang, B., Ding, W., Liu, X.
, et al., Efficient genome editing in plants using a CRISPR/Cas
system.
Cel Research 2013,
23, 1229-1232.
[92] Jiang, W., Zhou, H., Bi, H., Fromm, M.
, et al., Demonstration of CRISPR/Cas9/sgRNA-mediated targeted
gene modification in
Arabidopsis, tobacco, sorghum and rice.
Nucleic Acids Research 2013,
41.
[93] Li, J.-F., Norvil e, J. E., Aach, J., McCormack, M.
, et al., Multiplex and homologous recombination-
mediated genome editing in
Arabidopsis and Nicotiana benthamiana using guide RNA and Cas9.
Nat
Biotech 2013,
31, 688-691.
[94] Mao, Y., Zhang, H., Xu, N., Zhang, B.
, et al., Application of the CRISPR-Cas system for efficient genome
engineering in plants.
Molecular Plant 2013,
6, 2008-2011.
[95] Miao, J., Guo, D., Zhang, J., Huang, Q.
, et al., Targeted mutagenesis in rice using CRISPR-Cas system.
Cel Research 2013,
23, 1233-1236.
89
[96] Nekrasov, V., Staskawicz, B., Weigel, D., Jones, J. D. G., Kamoun, S., Targeted mutagenesis in the model
plant Nicotiana benthamiana using Cas9 RNA-guided endonuclease.
Nature biotechnology 2013,
31, 691-
693.
[97] Shan, Q., Wang, Y., Li, J., Zhang, Y.
, et al., Targeted genome modification of crop plants using a CRISPR-
Cas system.
Nature biotechnology 2013,
31, 686-688.
[98] Upadhyay, S. K., Kumar, J., Alok, A., Tuli, R., RNA-Guided Genome Editing for Target Gene Mutations in
Wheat.
G3-Genes Genomes Genetics 2013,
3, 2233-2238.
[99] Xie, K. B., Yang, Y. N., RNA-guided genome editing in plants using a CRISPR-Cas system.
Molecular Plant
2013,
6, 1975-1983.
[100] Liang, Z., Zhang, K., Chen, K., Gao, C., Targeted Mutagenesis in Zea mays Using TALENs and the
CRISPR/Cas System.
Journal of Genetics and Genomics 2014,
41, 63-68.
[101] Li, Z., Liu, Z.-B., Xing, A., Moon, B. P.
, et al., Cas9-Guide RNA Directed Genome Editing in Soybean.
Plant physiology 2015,
169, 960-970.
[102] Lawrenson, T., Shorinola, O., Stacey, N., Li, C.
, et al., Induction of targeted, heritable mutations in
barley and Brassica oleracea using RNA-guided Cas9 nuclease.
Genome biology 2015,
16, 258.
[103] Wang, S., Zhang, S., Wang, W., Xiong, X.
, et al., Efficient targeted mutagenesis in potato by the
CRISPR/Cas9 system.
Plant Cell Reports 2015,
34, 1473-1476.
[104] Brooks, C., Nekrasov, V., Lippman, Z. B., Van Eck, J., Efficient gene editing in tomato in the first
generation using the clustered regularly interspaced short palindromic repeats/CRISPR-associated9 system.
Plant physiology 2014,
166, 1292-1297.
[105] Fan, D., Liu, T., Li, C., Jiao, B.
, et al., Efficient CRISPR/Cas9-mediated Targeted Mutagenesis in Populus
in the First Generation.
Scientific reports 2015,
5.
[106] Jia, H., Wang, N., Xcc-facilitated agroinfiltration of citrus leaves: a tool for rapid functional analysis of
transgenes in citrus leaves.
Plant Cel Reports 2014,
33, 1993-2001.
[107] Feng, Z., Mao, Y., Xu, N., Zhang, B.
, et al., Multigeneration analysis reveals the inheritance, specificity,
and patterns of CRISPR/Cas-induced gene modifications in
Arabidopsis.
Proceedings of the National
Academy of Sciences of the United States of America 2014,
111, 4632-4637.
[108] Wang, Z.-P., Xing, H.-L., Dong, L., Zhang, H.-Y.
, et al., Egg cel -specific promoter-control ed
CRISPR/Cas9 efficiently generates homozygous mutants for multiple target genes in
Arabidopsis in a single
generation.
Genome biology 2015,
16.
[109] Xu, R.-F., Li, H., Qin, R.-Y., Li, J.
, et al., Generation of inheritable and "transgene clean" targeted
genome-modified rice in later generations using the CRISPR/Cas9 system.
Scientific reports 2015,
5.
[110] Zhou, H., Liu, B., Weeks, D. P., Spalding, M. H., Yang, B., Large chromosomal deletions and heritable
small genetic changes induced by CRISPR/Cas9 in rice.
Nucleic Acids Res 2014,
42, 10903-10914.
[111] Wang, Y., Cheng, X., Shan, Q., Zhang, Y.
, et al., Simultaneous editing of three homoeoalleles in
hexaploid bread wheat confers heritable resistance to powdery mildew.
Nat Biotech 2014,
32, 947-951.
[112] Ma, X., Zhang, Q., Zhu, Q., Liu, W.
, et al., A Robust CRISPR/Cas9 System for Convenient, High-
Efficiency Multiplex Genome Editing in Monocot and Dicot Plants.
Molecular Plant 2015,
8, 1274-1284.
[113] Xie, K., Minkenberg, B., Yang, Y., Boosting CRISPR/Cas9 multiplex editing capability with the
endogenous tRNA-processing system.
Proceedings of the National Academy of Sciences of the United States
of America 2015,
112, 3570-3575.
[114] Endo, M., Mikami, M., Toki, S., Multigene Knockout Utilizing Off-Target Mutations of the CRISPR/Cas9
System in Rice.
Plant and Cel Physiology 2015,
56, 41-47.
[115] Sun, Y., Zhang, X., Wu, C., He, Y.
, et al., Engineering Herbicide Resistant Rice Plants through
CRISPR/Cas9-mediated Homologous Recombination of the Acetolactate Synthase.
Mol Plant 2016.
[116] Schiml, S., Fauser, F., Puchta, H., The CRISPR/Cas system can be used as nuclease for
in planta gene
targeting and as paired nickases for directed mutagenesis in
Arabidopsis resulting in heritable progeny.
Plant Journal 2014,
80, 1139-1150.
[117] Luo, S., Li, J., Stoddard, T. J., Baltes, N. J.
, et al., Non-transgenic Plant Genome Editing Using Purified
Sequence-Specific Nucleases.
Mol Plant 2015,
8, 1425-1427.
[118] Cai, Y., Chen, L., Liu, X., Sun, S.
, et al., CRISPR/Cas9-Mediated Genome Editing in Soybean Hairy Roots.
PloS one 2015,
10.
90
[119] Fauser, F., Schiml, S., Puchta, H., Both CRISPR/Cas-based nucleases and nickases can be used
efficiently for genome engineering in
Arabidopsis thaliana.
Plant Journal 2014,
79, 348-359.
[120] Hyun, Y., Kim, J., Cho, S. W., Choi, Y.
, et al., Site-directed mutagenesis in
Arabidopsis thaliana using
dividing tissue-targeted RGEN of the CRISPR/Cas system to generate heritable null alleles.
Planta 2015,
241,
271-284.
[121] Jacobs, T. B., LaFayette, P. R., Schmitz, R. J., Parrott, W. A., Targeted genome modifications in soybean
with CRISPR/Cas9.
Bmc Biotechnology 2015,
15.
[122] Sun, X., Hu, Z., Chen, R., Jiang, Q.
, et al., Targeted mutagenesis in soybean using the CRISPR-Cas9
system.
Scientific reports 2015,
5.
[123] Xu, R., Li, H., Qin, R., Wang, L.
, et al., Gene targeting using the Agrobacterium tumefaciens-mediated
CRISPR-Cas system in rice.
Rice 2014,
7.
[124] Zhang, H., Zhang, J., Wei, P., Zhang, B.
, et al., The CRISPR/Cas9 system produces specific and
homozygous targeted gene editing in rice in one generation.
Plant biotechnology journal 2014,
12, 797-807.
[125] Zhou, H. B., Liu, B., Weeks, D. P., Spalding, M. H., Yang, B., Large chromosomal deletions and heritable
small genetic changes induced by CRISPR/Cas9 in rice.
Nucleic Acids Research 2014,
42, 10903-10914.
[126] Hsu, P. D., Scott, D. A., Weinstein, J. A., Ran, F. A.
, et al., DNA targeting specificity of RNA-guided Cas9
nucleases.
Nature biotechnology 2013,
31, 827-832.
[127] Xu, H., Xiao, T., Chen, C. H., Li, W.
, et al., Sequence determinants of improved CRISPR sgRNA design.
Genome research 2015,
25, 1147-1157.
[128] Doench, J. G., Hartenian, E., Graham, D. B., Tothova, Z.
, et al., Rational design of highly active sgRNAs
for CRISPR-Cas9-mediated gene inactivation.
Nature biotechnology 2014,
32, 1262-1267.
[129] Wang, T., Wei, J. J., Sabatini, D. M., Lander, E. S., Genetic screens in human cel s using the CRISPR-
Cas9 system.
Science (New York, N.Y.) 2014,
343, 80-84.
[130] Wu, X., Scott, D. A., Kriz, A. J., Chiu, A. C.
, et al., Genome-wide binding of the CRISPR endonuclease
Cas9 in mammalian cells.
Nature biotechnology 2014,
32, 670-676.
[131] Dang, Y., Jia, G., Choi, J., Ma, H.
, et al., Optimizing sgRNA structure to improve CRISPR-Cas9 knockout
efficiency.
Genome biology 2015,
16, 280.
[132] Zhang, H., Zhang, J., Wei, P., Zhang, B.
, et al., The CRISPR/Cas9 system produces specific and
homozygous targeted gene editing in rice in one generation.
Plant Biotechnol J. 2014,
12, 797 - 807.
[133] Vouillot, L., Thelie, A., Pollet, N., Comparison of T7E1 and surveyor mismatch cleavage assays to
detect mutations triggered by engineered nucleases.
G3 (Bethesda, Md.) 2015,
5, 407-415.
[134] Jacobs, T. B., LaFayette, P. R., Schmitz, R. J., Parrott, W. A., Targeted genome modifications in soybean
with CRISPR/Cas9.
BMC biotechnology 2015,
15, 16.
[135] Duan, Y. B., Li, J., Qin, R. Y., Xu, R. F.
, et al., Identification of a regulatory element responsible for salt
induction of rice OsRAV2 through ex situ and in situ promoter analysis.
Plant Molecular Biology 2015.
[136] Tang, F., Yang, S., Liu, J., Zhu, H., Rj4, a Gene Controlling Nodulation Specificity in Soybeans, Encodes a
Thaumatin-Like Protein, but Not the One Previously Reported.
Plant physiology 2015.
[137] Xu, C., Liberatore, K. L., Macalister, C. A., Huang, Z.
, et al., A cascade of arabinosyltransferases
controls shoot meristem size in tomato.
Nature genetics 2015,
47, 784-792.
[138] Lee, J., Chung, J. H., Kim, H. M., Kim, D. W., Kim, H., Designed nucleases for targeted genome editing.
Plant biotechnology journal 2015.
[139] Slaymaker, I. M., Gao, L., Zetsche, B., Scott, D. A.
, et al., Rationally engineered Cas9 nucleases with
improved specificity.
Science (New York, N.Y.) 2016,
351, 84-88.
[140] Kleinstiver, B. P., Pattanayak, V., Prew, M. S., Tsai, S. Q.
, et al., High-fidelity CRISPR-Cas9 nucleases
with no detectable genome-wide off-target effects.
Nature 2016,
529, 490-495.
[141] Bado, S., Forster, B. P., Nielen, S., Ali, A. M.
, et al., Plant Mutation Breeding: Current Progress and
Future Assessment,
Plant Breeding Reviews: Volume 39, John Wiley & Sons, Inc. 2015, pp. 23-88.
[142] Kurowska, M., Daszkowska-Golec, A., Gruszka, D., Marzec, M.
, et al., TILLING: a shortcut in functional
genomics.
Journal of applied genetics 2011,
52, 371-390.
[143] Curtis, M., DNA Repair Pathways and Genes in Plant, in: Shu, Q. Y., Forster, B. P., Nakagawa, H. (Eds.),
Plant mutation breeding and biotechnology CABI, Wallingford 2012, pp. 57-70.
91
[144] Bleuyard, J. Y., Gal ego, M. E., White, C. I., Recent advances in understanding of the DNA double-
strand break repair machinery of plants.
DNA repair 2006,
5, 1-12.
[145] Shu, H., Forster, B. P., Nakagawa, H., Principles and Applications of Plant Mutation Breeding, in: Shu,
Q. Y., Forster, B. P., Nakagawa, H. (Eds.),
Plant mutation breeding and biotechnology CABI, Wallingford
2012, pp. 301-325.
[146] EFSA, Scientific opinion addressing the safety assessment of plants developed through cisgenesis and
intragenesis.
EFSA Journal 2012,
10.
[147]
Single nucleotide polymorphisms - methods and protocols, Humana Press, New York, NY 2009.
[148] Vogel, B.,
Neue Pflanzenzuchtverfahren - Grundlagen für die Klärung offener Fragen bei der
rechtlichen Regulierung neuer Pflanzenzuchtverfahren, Bundesamt für Umwelt, Bern 2012.
[149] Cooper, J. L., Till, B. J., Laport, R. G., Darlow, M. C.
, et al., TILLING to detect induced mutations in
soybean.
BMC plant biology 2008,
8, 9.
[150] Til , B. J., Cooper, J., Tai, T. H., Colowit, P.
, et al., Discovery of chemically induced mutations in rice by
TILLING.
BMC plant biology 2007,
7, 19.
[151] Chen, L., Huang, L., Min, D., Phil ips, A.
, et al., Development and characterization of a new TILLING
population of common bread wheat (Triticum aestivum L.).
PloS one 2012,
7, e41570.
[152] Arumuganathan, K., Earle, E. D., Nuclear DNA content of some important plant species.
Plant
Molecular Biology Reporter,
9, 208-218.
[153] Schmutz, J., Cannon, S. B., Schlueter, J., Ma, J.
, et al., Genome sequence of the palaeopolyploid
soybean.
Nature 2010,
463, 178-183.
[154] Project, I. R. G. S., The map-based sequence of the rice genome.
Nature 2005,
436, 793-800.
[155] Cheng, Z., Lin, J., Lin, T., Xu, M.
, et al., Genome-wide analysis of radiation-induced mutations in rice
(Oryza sativa L. ssp. indica).
Molecular bioSystems 2014,
10, 795-805.
[156] Kumar, A., Simons, K., Iqbal, M. J., de Jimenez, M. M.
, et al., Physical mapping resources for large
plant genomes: radiation hybrids for wheat D-genome progenitor Aegilops tauschi .
BMC genomics 2012,
13, 597.
[157] Fitzgerald, T., Kazan, K., Li, Z., Morell, M., Manners, J., A high-throughput method for the detection of
homoeologous gene deletions in hexaploid wheat.
BMC plant biology 2010,
10, 1-15.
[158] IWGSC, A chromosome-based draft sequence of the hexaploid bread wheat (Triticum aestivum)
genome.
Science (New York, N.Y.) 2014,
345, 1251788.
[159] Flachowsky, H., Hanke, M. V., Peil, A., Strauss, S. H., Fladung, M., A review on transgenic approaches
to accelerate breeding of woody plants.
Plant Breeding 2009,
128, 217-226.
[160] Sasaki, S., Yamagishi, N., Yoshikawa, N., Efficient virus-induced gene silencing in apple, pear and
Japanese pear using Apple latent spherical virus vectors.
Plant methods 2011,
7, 15.
[161] Yamagishi, N., Sasaki, S., Yamagata, K., Komori, S.
, et al., Promotion of flowering and reduction of a
generation time in apple seedlings by ectopical expression of the
Arabidopsis thaliana FT gene using the
Apple latent spherical virus vector.
Plant Mol Biol 2011,
75, 193-204.
[162] van Nocker, S., Gardiner, S. E., Breeding better cultivars, faster: applications of new technologies for
the rapid deployment of superior horticultural tree crops.
Horticulture Research 2014,
1, 14022.
[163] Blümel, M., Dally, N., Jung, C., Flowering time regulation in crops — what did we learn from
Arabidopsis?
Current opinion in biotechnology 2015,
32, 121-129.
[164] Pena, L., Martin-Trillo, M., Juarez, J., Pina, J. A.
, et al., Constitutive expression of
Arabidopsis LEAFY or
APETALA1 genes in citrus reduces their generation time.
Nature biotechnology 2001,
19, 263-267.
[165] Flachowsky, H., Hattasch, C., Hofer, M., Peil, A., Hanke, M. V., Overexpression of LEAFY in apple leads
to a columnar phenotype with shorter internodes.
Planta 2010,
231, 251-263.
[166] Weigl, K., Wenzel, S., Flachowsky, H., Peil, A., Hanke, M. V., Integration of BpMADS4 on various
linkage groups improves the utilization of the rapid cycle breeding system in apple.
Plant biotechnology
journal 2015,
13, 246-258.
[167] Srinivasan, C., Dardick, C., Callahan, A., Scorza, R., Plum (Prunus domestica) trees transformed with
poplar FT1 result in altered architecture, dormancy requirement, and continuous flowering.
PloS one 2012,
7, e40715.
92
[168] Flachowsky, H., Le Roux, P. M., Peil, A., Patocchi, A.
, et al., Application of a high-speed breeding
technology to apple (Malus x domestica) based on transgenic early flowering plants and marker-assisted
selection.
New Phytologist 2011,
192, 364-377.
[169] Scorza, R., Callahan, A., Dardick, C., Ravelonandro, M.
, et al., Genetic engineering of Plum pox virus
resistance: 'HoneySweet' plum-from concept to product.
Plant Cel Tissue and Organ Culture 2013,
115, 1-
12.
[170] Yamagishi, N., Kishigami, R., Yoshikawa, N., Reduced generation time of apple seedlings to within a
year by means of a plant virus vector: A new plant-breeding technique with no transmission of genetic
modification to the next generation.
Plant biotechnology journal 2014,
12, 60-68.
[171] Fischer, M., Fischer, C., Evaluation of Malus species and cultivars at the Fruit Genebank Dresden-
Pillnitz and its use for apple resistance breeding.
Genetic Resources and Crop Evolution 1999,
46, 235-241.
[172] García-Libreros, T.,
Department of Applied Genetics and Cel Biology, University of Natural Resources
and Life Sciences, Vienna 2012, p. 151.
[173] Gessler, C., Pertot, I., Vf scab resistance of Malus.
Trees - Structure and Function 2012,
26, 1-14.
[174] Wang, A. D., Aldwinckle, H., Forsline, P., Main, D.
, et al., EST contig-based SSR linkage maps for Malus
x domestica cv Royal Gala and an apple scab resistant accession of M. sieversii, the progenitor species of
domestic apple.
Molecular Breeding 2012,
29, 379-397.
[175] Lewis, R. S., Kernodle, S. P., A method for accelerated trait conversion in plant breeding.
TAG.
Theoretical and applied genetics. Theoretische und angewandte Genetik 2009,
118, 1499-1508.
[176] Flachowsky, H., Le Roux, P. M., Peil, A., Patocchi, A.
, et al., Application of a high-speed breeding
technology to apple (Malus×domestica) based on transgenic early flowering plants and marker-assisted
selection.
New Phytologist 2011,
192, 364-377.
[177] Joshi, S., Soriano, J., Schaart, J., Broggini, G. A. L.
, et al., Approaches for Development of Cisgenic
Apples.
Transgenic Plant Journal 2009, 6.
[178] Sansavini, S., Tartarini, S.,
Acta Horticulturae 2013, pp. 43-56.
[179] Iezzoni, A., RosBREED: Enabling Marker-assisted Breeding in the Rosaceae.
Hortscience 2010,
45, S27-
S28.
[180] Baumgartner, I. O., Kellerhals, M., Costa, F., Dondini, L.
, et al., Development of SNP-based assays for
disease resistance and fruit quality traits in apple (Malus x domestica Borkh.) and validation in breeding
pilot studies.
Tree Genetics & Genomes 2016,
12.
[181] Chagne, D., Crowhurst, R. N., Pindo, M., Thrimawithana, A.
, et al., The draft genome sequence of
European pear (Pyrus communis L. 'Bartlett').
PloS one 2014,
9, e92644.
[182] Velasco, R., Zharkikh, A., Affourtit, J., Dhingra, A.
, et al., The genome of the domesticated apple
(Malus x domestica Borkh.).
Nature genetics 2010,
42, 833-839.
[183] Verde, I., Abbott, A. G., Scalabrin, S., Jung, S.
, et al., The high-quality draft genome of peach (Prunus
persica) identifies unique patterns of genetic diversity, domestication and genome evolution.
Nature
genetics 2013,
45, 487-494.
[184] Xu, Q., Chen, L. L., Ruan, X., Chen, D.
, et al., The draft genome of sweet orange (Citrus sinensis).
Nature genetics 2013,
45, 59-66.
[185] Kumar, S., Bink, M. C. A. M., Volz, R. K., Bus, V. G. M., Chagné, D., Towards genomic selection in apple
(Malus × domestica Borkh.) breeding programmes: Prospects, challenges and strategies.
Tree Genetics &
Genomes 2011,
8, 1-14.
[186] McClure, K. A., Sawler, J., Gardner, K. M., Money, D., Myles, S., Genomics: a potential panacea for the
perennial problem.
American Journal of Botany 2014,
101, 1780-1790.
[187] Myles, S., Improving fruit and wine: what does genomics have to offer?
Trends in Genetics 2013,
29,
190-196.
[188] Brown, S., Apple, in: Badenes, M. (Ed.),
Fruit Breeding, Springer, New York 2012, p. 329.
[189] Bannier, H.-J., Modern Apple Breeding: Genetic Narrowing and Inbreeding Tendencies.
Erwerbs-
Obstbau 2011,
52, 85-110.
[190] Liang, W., Dondini, L., De Franceschi, P., Paris, R.
, et al., Genetic Diversity, Population Structure and
Construction of a Core Collection of Apple Cultivars from Italian Germplasm.
Plant Molecular Biology
Reporter 2015,
33, 458-473.
93
[191] Weigel, D., Nilsson, O., A developmental switch sufficient for flower initiation in diverse plants.
Nature 1995,
377, 495-500.
[192] Yamagishi, N., Yoshikawa, N., Expression of FLOWERING LOCUS T from
Arabidopsis thaliana induces
precocious flowering in soybean irrespective of maturity group and stem growth habit.
Planta 2011,
233,
561-568.
[193] Srinivasan, C., Dardick, C., Callahan, A., Scorza, R., Plum (Prunus domestica) trees transformed with
poplar FT1 result in altered architecture, dormancy requirement, and continuous flowering.
PLoS ONE
2012,
7.
[194] Zhang, H., Harry, D. E., Ma, C., Yuceer, C.
, et al., Precocious flowering in trees: The FLOWERING LOCUS
T gene as a research and breeding tool in Populus.
Journal of experimental botany 2010,
61, 2549-2560.
[195] Flachowsky, H., Szankowski, I., Waidmann, S., Peil, A.
, et al., The MdTFL1 gene of apple (Malus ×
domestica Borkh.) reduces vegetative growth and generation time.
Tree Physiology 2012,
32, 1288-1301.
[196] Hoenicka, H., Lehnhardt, D., Nilsson, O., Hanelt, D., Fladung, M., Successful crossings with early
flowering transgenic poplar: interspecific crossings, but not transgenesis, promoted aberrant phenotypes in
offspring.
Plant biotechnology journal 2014,
12, 1066-1074.
[197] Hoenicka, H., Lehnhardt, D., Polak, O., Fladung, M., Early flowering and genetic containment studies
in transgenic poplar.
IForest 2012,
5, 138-146.
[198] Klocko, A. L., Ma, C., Robertson, S., Esfandiari, E.
, et al., FT overexpression induces precocious
flowering and normal reproductive development in Eucalyptus.
Plant biotechnology journal 2015.
[199] Weigl, K., Flachowsky, H., Peil, A., Hanke, M.-V., Heat mediated silencing of MdTFL1 genes in apple
(Malus × domestica).
Plant Cell Tiss Organ Cult 2015,
123, 511-521.
[200] Wenzel, S., Flachowsky, H., Hanke, M. V., The Fast-track breeding approach can be improved by heat-
induced expression of the FLOWERING LOCUS T genes from poplar (Populus trichocarpa) in apple (Malus x
domestica Borkh.).
Plant Cel Tissue and Organ Culture 2013,
115, 127-137.
[201] Wenzel, S., Flachowsky, H., Hanke, M. V., Preliminary results to establish a speed-breed program
based on heat-induced precocious flowering of apple plants containing the flowering locus t gene from
poplar (populus trichocarpa).
Acta Horticulturae 2013,
976, 471-476.
[202] Yamagishi, N., Li, C., Yoshikawa, N., Promotion of Flowering by Apple Latent Spherical Virus Vector
and Virus Elimination at High Temperature Allow Accelerated Breeding of Apple and Pear.
Frontiers in plant
science 2016,
7, 171.
[203] Tamura, A., Kato, T., Taki, A., Sone, M.
, et al., Preventive and curative effects of Apple latent spherical
virus vectors harboring part of the target virus genome against potyvirus and cucumovirus infections.
Virology 2013,
446, 314-324.
[204] Li, C., Sasaki, N., Isogai, M., Yoshikawa, N., Stable expression of foreign proteins in herbaceous and
apple plants using Apple latent spherical virus RNA2 vectors.
Archives of virology 2004,
149, 1541-1558.
[205] Trankner, C., Lehmann, S., Hoenicka, H., Hanke, M. V.
, et al., Over-expression of an FT-homologous
gene of apple induces early flowering in annual and perennial plants.
Planta 2010,
232, 1309-1324.
[206] Jackson, S. D., Hong, Y., Systemic movement of FT mRNA and a possible role in floral induction.
Frontiers in plant science 2012,
3, 127.
[207] Le Roux, P. M., Flachowsky, H., Hanke, M. V., Gessler, C., Patocchi, A., Use of a transgenic early
flowering approach in apple (Malus x domestica Borkh.) to introgress fire blight resistance from cultivar
Evereste.
Molecular Breeding 2012,
30, 857-874.
[208] Le Roux, P. M., Flachowsky, H., Jänsch, M., Kel erhals, M.
, et al., Development of apple pre-breeding
genotypes highly resistant to fire blight by early flowering.
Acta Horticulturae 2014,
1048, 55-64.
[209] Yao, S., FasTracking Plum Breeding.
Agricultural research 2011,
59, 2.
[210] Scorza, R., Callahan, A. M., Dardick, C. D., Srinivasan, C.
, et al., Biotechnological advances in the
genetic improvement of prunus domestica.
Acta Horticulturae 2013,
985, 111-118.
[211] Scorza, R., Callahan, A., Dardick, C., Ravelonandro, M.
, et al., Genetic engineering of Plum pox virus
resistance: 'HoneySweet' plum-from concept to product.
Plant Cel , Tissue and Organ Culture 2013,
115, 1-
12.
[212] Srinivasan, C., Scorza, R., Callahan, A., Dardick, C.,
Development of Very Early Flowering and Normal
Fruiting Plum With Fertile Seeds, WIPO 2012.
94
[213] Park, D., Kim, D., Jang, G., Lim, J.
, et al., Efficiency to Discovery Transgenic Loci in GM Rice Using Next
Generation Sequencing Whole Genome Re-sequencing.
Genomics & informatics 2015,
13, 81-85.
[214] Sastry, K. S.,
Seed-borne plant virus diseases, 2013.
[215] Panattoni, A., Luvisi, A., Triolo, E., Review. Elimination of viruses in plants: Twenty years of progress.
Spanish Journal of Agricultural Research 2013,
11, 173-188.
[216] Elo, A., Lemmetyinen, J., Novak, A., Keinonen, K.
, et al., BpMADS4 has a central role in inflorescence
initiation in silver birch (Betula pendula).
Physiologia plantarum 2007,
131, 149-158.
[217] Cervera, M., Navarro, L., Peña, L., Gene stacking in 1-year-cycling APETALA1 citrus plants for a rapid
evaluation of transgenic traits in reproductive tissues.
Journal of Biotechnology 2009,
140, 278-282.
[218] Endo, T., Shimada, T., Fujii, H., Nishikawa, F.
, et al., Development of a CiFT Co-expression system for
functional analysis of genes in citrus flowers and fruit.
Journal of the Japanese Society for Horticultural
Science 2009,
78, 74-83.
[219] Kotoda, N., Hayashi, H., Suzuki, M., Igarashi, M.
, et al., Molecular characterization of FLOWERING
LOCUS T-like genes of apple (Malus x domestica Borkh.).
Plant & cel physiology 2010,
51, 561-575.
[220] Kotoda, N., Iwanami, H., Takahashi, S., Abe, K., Antisense expression of MdTFL1, a TFL1-like gene,
reduces the juvenile phase in apple.
Journal of the American Society for Horticultural Science 2006,
131, 74-
81.
[221] Flachowsky, H., Peil, A., Sopanen, T., Elo, A., Hanke, V., Overexpression of BpMADS4 from silver birch
(Betula pendula Roth.) induces early-flowering in apple (Malus x domestica Borkh.).
Plant Breeding 2007,
126, 137-145.
[222] Szankowski, I., Waidmann, S., Omar, A. E. D. S., Flachowsky, H.
, et al.,
Acta Horticulturae 2009, pp.
633-636.
[223] Trankner, C., Lehmann, S., Hoenicka, H., Hanke, M. V.
, et al., Note added in proof to: Over-expression
of an FT-homologous gene of apple induces early flowering in annual and perennial plants.
Planta 2011,
233, 217-218.
[224] Rottmann, W. H., Meilan, R., Sheppard, L. A., Brunner, A. M.
, et al., Diverse effects of overexpression
of LEAFY and PTLF, a poplar (Populus) homolog of LEAFY/FLORICAULA, in transgenic poplar and
Arabidopsis.
The Plant journal : for cel and molecular biology 2000,
22, 235-245.
[225] Bohlenius, H., Huang, T., Charbonnel-Campaa, L., Brunner, A. M.
, et al., CO/FT regulatory module
controls timing of flowering and seasonal growth cessation in trees.
Science (New York, N.Y.) 2006,
312,
1040-1043.
[226] Hoenicka, H., Nowitzki, O., Hanelt, D., Fladung, M., Heterologous overexpression of the birch
FRUITFULL-like MADS-box gene BpMADS4 prevents normal senescence and winter dormancy in Populus
tremula L.
Planta 2008,
227, 1001-1011.
[227] Graham, T., Scorza, R., Wheeler, R., Smith, B.
, et al., Over-Expression of FT1 in Plum (Prunus
domestica) Results in Phenotypes Compatible with Spaceflight: A Potential New Candidate Crop for
Bioregenerative Life Support Systems.
Gravitational and Space Research 2015,
3, 12.
[228] Freiman, A., Shlizerman, L., Golobovitch, S., Yablovitz, Z.
, et al., Development of a transgenic early
flowering pear (Pyrus communis L.) genotype by RNAi silencing of PcTFL1-1 and PcTFL1-2.
Planta 2012,
235,
1239-1251.
[229] Martinez de Alba, A. E., Elvira-Matelot, E., Vaucheret, H., Gene silencing in plants: a diversity of
pathways.
Biochimica et biophysica acta 2013,
1829, 1300-1308.
[230] Bologna, N. G., Voinnet, O., The diversity, biogenesis, and activities of endogenous silencing small
RNAs in
Arabidopsis.
Annual review of plant biology 2014,
65, 473-503.
[231] Baulcombe, D. C., VIGS, HIGS and FIGS: small RNA silencing in the interactions of viruses or
filamentous organisms with their plant hosts.
Current opinion in plant biology 2015,
26, 141-146.
[232] Molnar, A., Melnyk, C., Baulcombe, D. C., Silencing signals in plants: a long journey for small RNAs.
Genome biology 2011,
12, 215.
[233] Parent, J. S., Martinez de Alba, A. E., Vaucheret, H., The origin and effect of small RNA signaling in
plants.
Frontiers in plant science 2012,
3, 179.
[234] Xie, M., Zhang, S., Yu, B., microRNA biogenesis, degradation and activity in plants.
Cel ular and
molecular life sciences : CMLS 2015,
72, 87-99.
95
[235] Warthmann, N., Chen, H., Ossowski, S., Weigel, D., Herve, P., Highly specific gene silencing by artificial
miRNAs in rice.
PloS one 2008,
3, e1829.
[236] Rogers, K., Chen, X., Biogenesis, turnover, and mode of action of plant microRNAs.
The Plant cel
2013,
25, 2383-2399.
[237] Liu, Q., Wang, F., Axtell, M. J., Analysis of complementarity requirements for plant microRNA
targeting using a Nicotiana benthamiana quantitative transient assay.
The Plant cel 2014,
26, 741-753.
[238] Ossowski, S., Schwab, R., Weigel, D., Gene silencing in plants using artificial microRNAs and other
small RNAs.
The Plant journal : for cel and molecular biology 2008,
53, 674-690.
[239] Tiwari, M., Sharma, D., Trivedi, P., Artificial microRNA mediated gene silencing in plants: progress and
perspectives.
Plant Molecular Biology 2014,
86, 1-18.
[240] Schwab, R., Ossowski, S., Riester, M., Warthmann, N., Weigel, D., Highly specific gene silencing by
artificial microRNAs in
Arabidopsis.
The Plant cel 2006,
18, 1121-1133.
[241] Carbonell, A., Takeda, A., Fahlgren, N., Johnson, S. C.
, et al., New generation of artificial MicroRNA
and synthetic trans-acting small interfering RNA vectors for efficient gene silencing in
Arabidopsis.
Plant
physiology 2014,
165, 15-29.
[242] Li, J. F., Chung, H. S., Niu, Y., Bush, J.
, et al., Comprehensive protein-based artificial microRNA screens
for effective gene silencing in plants.
The Plant cel 2013,
25, 1507-1522.
[243] Barampuram, S., Zhang, Z. J., Recent advances in plant transformation.
Methods in molecular biology
(Clifton, N.J.) 2011,
701, 1-35.
[244] Voinnet, O., Use, tolerance and avoidance of amplified RNA silencing by plants.
Trends in plant
science 2008,
13, 317-328.
[245] Wesley, S. V., Hel iwel , C. A., Smith, N. A., Wang, M. B.
, et al., Construct design for efficient, effective
and high-throughput gene silencing in plants.
The Plant journal : for cel and molecular biology 2001,
27,
581-590.
[246] Acevedo-Garcia, J., Kusch, S., Panstruga, R., Magical mystery tour: MLO proteins in plant immunity
and beyond.
The New phytologist 2014,
204, 273-281.
[247] Antony, G., Zhou, J., Huang, S., Li, T.
, et al., Rice xa13 recessive resistance to bacterial blight is
defeated by induction of the disease susceptibility gene Os-11N3.
The Plant cel 2010,
22, 3864-3876.
[248] Li, S. H., Ji, R., Dudler, R., Yong, M. L.
, et al., Wheat gene TaS3 contributes to powdery mildew
susceptibility.
Plant Cel Reports 2013,
32, 1891-1901.
[249] Habben, J. E., Bao, X., Bate, N. J., DeBruin, J. L.
, et al., Transgenic alteration of ethylene biosynthesis
increases grain yield in maize under field drought-stress conditions.
Plant biotechnology journal 2014,
12,
685-693.
[250] Pieczynski, M., Marczewski, W., Hennig, J., Dolata, J.
, et al., Down-regulation of CBP80 gene
expression as a strategy to engineer a drought-tolerant potato.
Plant biotechnology journal 2013,
11, 459-
469.
[251] Waltz, E., Beating the heat.
Nat Biotech 2014,
32, 610-613.
[252] Wang, Y., Beaith, M., Chalifoux, M., Ying, J.
, et al., Shoot-specific down-regulation of protein
farnesyltransferase (alpha-subunit) for yield protection against drought in canola.
Mol Plant 2009,
2, 191-
200.
[253] EFSA, Opinion of the Scientific Panel on Genetically Modified Organisms on an application (Reference
EFSA-GMO-UK-2005-14) for the placing on the market of genetically modified potato EH92-527-1 with
altered starch composition, for production of starch and food/feed uses , under Regulation (EC) No
1829/2003 from BASF plant science.
EFSA Journal 2006,
4.
[254] Regina, A., Bird, A., Topping, D., Bowden, S.
, et al., High-amylose wheat generated by RNA
interference improves indices of large-bowel health in rats.
Proc Natl Acad Sci U S A 2006,
103, 3546-3551.
[255] Davuluri, G. R., van Tuinen, A., Fraser, P. D., Manfredonia, A.
, et al., Fruit-specific RNAi-mediated
suppression of DET1 enhances carotenoid and flavonoid content in tomatoes.
Nature biotechnology 2005,
23, 890-895.
[256] Yu, B., Lydiate, D. J., Young, L. W., Schafer, U. A., Hannoufa, A., Enhancing the carotenoid content of
Brassica napus seeds by downregulating lycopene epsilon cyclase.
Transgenic research 2008,
17, 573-585.
96
[257] Ali, N., Paul, S., Gayen, D., Sarkar, S. N.
, et al., Development of low phytate rice by RNAi mediated
seed-specific silencing of inositol 1,3,4,5,6-pentakisphosphate 2-kinase gene (IPK1).
PloS one 2013,
8,
e68161.
[258] Dubois, A. E., Pagliarani, G., Brouwer, R. M., Kollen, B. J.
, et al., First successful reduction of clinical
allergenicity of food by genetic modification: Mal d 1-silenced apples cause fewer allergy symptoms than
the wild-type cultivar.
Al ergy 2015,
70, 1406-1412.
[259] Gilissen, L. J., Bolhaar, S. T., Matos, C. I., Rouwendal, G. J.
, et al., Silencing the major apple allergen
Mal d 1 by using the RNA interference approach.
The Journal of allergy and clinical immunology 2005,
115,
364-369.
[260] Peters, S., Imani, J., Mahler, V., Foetisch, K.
, et al., Dau c 1.01 and Dau c 1.02-silenced transgenic
carrot plants show reduced allergenicity to patients with carrot allergy.
Transgenic research 2011,
20, 547-
556.
[261] Gil-Humanes, J., Piston, F., Altamirano-Fortoul, R., Real, A.
, et al., Reduced-gliadin wheat bread: an
alternative to the gluten-free diet for consumers suffering gluten-related pathologies.
PloS one 2014,
9,
e90898.
[262] Gil-Humanes, J., Piston, F., Tollefsen, S., Sollid, L. M., Barro, F., Effective shutdown in the expression of
celiac disease-related wheat gliadin T-cell epitopes by RNA interference.
Proc Natl Acad Sci U S A 2010,
107,
17023-17028.
[263] Barro, F., Iehisa, J. C., Gimenez, M. J., Garcia-Molina, M. D.
, et al., Targeting of prolamins by RNAi in
bread wheat: effectiveness of seven silencing-fragment combinations for obtaining lines devoid of coeliac
disease epitopes from highly immunogenic gliadins.
Plant biotechnology journal 2016,
14, 986-996.
[264] Raboy, V., Approaches and chal enges to engineering seed phytate and total phosphorus.
Plant
Science 2009,
177, 281-296.
[265] Lott, J. N. A., Ockenden, I., Raboy, V., Batten, G. D., Phytic acid and phosphorus in crop seeds and
fruits: A global estimate.
Seed Science Research 2000,
10, 11-33.
[266] Zhang, C., Wu, Z., Li, Y., Wu, J., Biogenesis, Function, and Applications of Virus-Derived Small RNAs in
Plants.
Frontiers in microbiology 2015,
6, 1237.
[267] Wang, M. B., Abbott, D. C., Waterhouse, P. M., A single copy of a virus-derived transgene encoding
hairpin RNA gives immunity to barley yellow dwarf virus.
Molecular plant pathology 2000,
1, 347-356.
[268] Zhang, X., Li, H., Zhang, J., Zhang, C.
, et al., Expression of artificial microRNAs in tomato confers
efficient and stable virus resistance in a cell-autonomous manner.
Transgenic research 2011,
20, 569-581.
[269] Fahim, M., Millar, A. A., Wood, C. C., Larkin, P. J., Resistance to Wheat streak mosaic virus generated
by expression of an artificial polycistronic microRNA in wheat.
Plant biotechnology journal 2012,
10, 150-
163.
[270] Taylor, N. J., Halsey, M., Gaitan-Solis, E., Anderson, P.
, et al., The VIRCA Project: virus resistant cassava
for Africa.
GM crops & food 2012,
3, 93-103.
[271] Aragao, F. J., Faria, J. C., First transgenic geminivirus-resistant plant in the field.
Nature biotechnology
2009,
27, 1086-1088; author reply 1088-1089.
[272] Aragao, F. J., Nogueira, E. O., Tinoco, M. L., Faria, J. C., Molecular characterization of the first
commercial transgenic common bean immune to the Bean golden mosaic virus.
J Biotechnol 2013,
166, 42-
50.
[273] Koch, A., Kogel, K. H., New wind in the sails: improving the agronomic value of crop plants through
RNAi-mediated gene silencing.
Plant biotechnology journal 2014,
12, 821-831.
[274] Lilley, C. J., Davies, L. J., Urwin, P. E., RNA interference in plant parasitic nematodes: a summary of the
current status.
Parasitology 2012,
139, 630-640.
[275] Niu, J., Jian, H., Xu, J., Chen, C.
, et al., RNAi silencing of the Meloidogyne incognita Rpn7 gene reduces
nematode parasitic success.
European Journal of Plant Pathology 2012,
134, 131-144.
[276] Steeves, R. M., Todd, T. C., Essig, J. S., Trick, H. N., Transgenic soybeans expressing siRNAs specific to a
major sperm protein gene suppress Heterodera glycines reproduction.
Functional Plant Biology 2006,
33,
991-999.
97
[277] Youssef, R. M., Kim, K. H., Haroon, S. A., Matthews, B. F., Post-transcriptional gene silencing of the
gene encoding aldolase from soybean cyst nematode by transformed soybean roots.
Experimental
parasitology 2013,
134, 266-274.
[278] Zhang, H., Li, H. C., Miao, X. X., Feasibility, limitation and possible solutions of RNAi-based technology
for insect pest control.
Insect science 2013,
20, 15-30.
[279] Xu, L., Duan, X., Lv, Y., Zhang, X.
, et al., Silencing of an aphid carboxylesterase gene by use of plant-
mediated RNAi impairs Sitobion avenae tolerance of Phoxim insecticides.
Transgenic research 2014,
23,
389-396.
[280] Baum, J. A., Bogaert, T., Clinton, W., Heck, G. R.
, et al., Control of coleopteran insect pests through
RNA interference.
Nature biotechnology 2007,
25, 1322-1326.
[281] Cheng, W., Song, X. S., Li, H. P., Cao, L. H.
, et al., Host-induced gene silencing of an essential chitin
synthase gene confers durable resistance to Fusarium head blight and seedling blight in wheat.
Plant
biotechnology journal 2015,
13, 1335-1345.
[282] Koch, A., Kumar, N., Weber, L., Kel er, H.
, et al., Host-induced gene silencing of cytochrome P450
lanosterol C14alpha-demethylase-encoding genes confers strong resistance to Fusarium species.
Proc Natl
Acad Sci U S A 2013,
110, 19324-19329.
[283] Nowara, D., Gay, A., Lacomme, C., Shaw, J.
, et al., HIGS: host-induced gene silencing in the obligate
biotrophic fungal pathogen Blumeria graminis.
The Plant cel 2010,
22, 3130-3141.
[284] Lange, M., Yellina, A. L., Orashakova, S., Becker, A., Virus-induced gene silencing (VIGS) in plants: an
overview of target species and the virus-derived vector systems.
Methods in molecular biology (Clifton, N.J.)
2013,
975, 1-14.
[285] Pandey, P., Senthil-Kumar, M., Mysore, K. S., Advances in plant gene silencing methods.
Methods in
molecular biology (Clifton, N.J.) 2015,
1287, 3-23.
[286] Bourque, J. E., Antisense strategies for genetic manipulations in plants.
Plant Science 1995,
105, 125-
149.
[287] Frizzi, A., Huang, S. S., Tapping RNA silencing pathways for plant biotechnology.
Plant biotechnology
journal 2010,
8, 655-677.
[288] Senior, I. J., Uses of plant gene silencing.
Biotechnology & genetic engineering reviews 1998,
15, 79-
119.
[289] Waterhouse, P. M., Graham, M. W., Wang, M. B., Virus resistance and gene silencing in plants can be
induced by simultaneous expression of sense and antisense RNA.
Proc Natl Acad Sci U S A 1998,
95, 13959-
13964.
[290] Schwab, R., Ossowski, S., Warthmann, N., Weigel, D., Directed gene silencing with artificial
microRNAs.
Methods in molecular biology (Clifton, N.J.) 2010,
592, 71-88.
[291] Chuang, C. F., Meyerowitz, E. M., Specific and heritable genetic interference by double-stranded RNA
in
Arabidopsis thaliana.
Proc Natl Acad Sci U S A 2000,
97, 4985-4990.
[292] Jones-Rhoades, M. W., Bartel, D. P., Bartel, B., MicroRNAS and their regulatory roles in plants.
Annual
review of plant biology 2006,
57, 19-53.
[293] Schwab, R., Palatnik, J. F., Riester, M., Schommer, C.
, et al., Specific effects of microRNAs on the plant
transcriptome.
Developmental cel 2005,
8, 517-527.
[294] Jones-Rhoades, M. W., Bartel, D. P., Computational identification of plant microRNAs and their
targets, including a stress-induced miRNA.
Molecular cel 2004,
14, 787-799.
[295] Fahlgren, N., Hill, S. T., Carrington, J. C., Carbonell, A., P-SAMS: a web site for plant artificial microRNA
and synthetic trans-acting small interfering RNA design.
Bioinformatics (Oxford, England) 2016,
32, 157-
158.
[296] Senthil-Kumar, M., Hema, R., Anand, A., Kang, L.
, et al., A systematic study to determine the extent of
gene silencing in Nicotiana benthamiana and other Solanaceae species when heterologous gene sequences
are used for virus-induced gene silencing.
New Phytologist 2007,
176, 782-791.
[297] Xu, P., Zhang, Y., Kang, L., Roossinck, M. J., Mysore, K. S., Computational estimation and experimental
verification of off-target silencing during posttranscriptional gene silencing in plants.
Plant physiology 2006,
142, 429-440.
98
[298] Felippes, F. F., Weigel, D., Triggering the formation of tasiRNAs in
Arabidopsis thaliana: the role of
microRNA miR173.
EMBO reports 2009,
10, 264-270.
[299] Montgomery, T. A., Yoo, S. J., Fahlgren, N., Gilbert, S. D.
, et al., AGO1-miR173 complex initiates
phased siRNA formation in plants.
Proc Natl Acad Sci U S A 2008,
105, 20055-20062.
[300] Ahmed, F., Dai, X., Zhao, P. X., Bioinformatics tools for achieving better gene silencing in plants.
Methods in molecular biology (Clifton, N.J.) 2015,
1287, 43-60.
[301] Senthil-Kumar, M., Mysore, K. S., Caveat of RNAi in plants: the off-target effect.
Methods in molecular
biology (Clifton, N.J.) 2011,
744, 13-25.
[302] Sherman, J. H., Munyikwa, T., Chan, S. Y., Petrick, J. S.
, et al., RNAi technologies in agricultural
biotechnology: The Toxicology Forum 40th Annual Summer Meeting.
Regulatory toxicology and
pharmacology : RTP 2015,
73, 671-680.
[303] EFSA, Scientific Opinion on Guidance for risk assessment of food and feed from genetically modified
plants.
EFSA Journal 2011,
9.
[304] EFSA, Scientific Opinion on Guidance for the risk assessment of genetically modified plants used for
non-food or non-feed purposes.
EFSA Journal 2009,
1164.
[305] EFSA, Scientific Opinion on the assessment of potential impacts of genetically modified plants on non-
target organisms.
EFSA Journal 2010,
8.
[306] Parisi, C., Tillie, P., Rodriguez-Cerezo, E., The global pipeline of GM crops out to 2020.
Nature
biotechnology 2016,
34, 31-36.
[307] EFSA, International scientific workshop ‘Risk assessment considerations for RNAi-based GM plants’.
2014,
EN-705.
[308] Ramon, M., Devos, Y., Lanzoni, A., Liu, Y.
, et al., RNAi-based GM plants: food for thought for risk
assessors.
Plant biotechnology journal 2014,
12, 1271-1273.
[309] OECD,
Safety Assessment of Foods and Feeds Derived from Transgenic Crops, Volume 1, OECD
Publishing.
[310] OECD,
Safety Assessment of Foods and Feeds Derived from Transgenic Crops, Volume 2, OECD
Publishing.
[311] Zhang, L., Hou, D., Chen, X., Li, D.
, et al., Exogenous plant MIR168a specifically targets mammalian
LDLRAP1: evidence of cross-kingdom regulation by microRNA.
Cell Res 2012,
22, 107-126.
[312] Dickinson, B., Zhang, Y., Petrick, J. S., Heck, G.
, et al., Lack of detectable oral bioavailability of plant
microRNAs after feeding in mice.
Nature biotechnology 2013,
31, 965-967.
[313] Zhang, Y., Wiggins, B. E., Lawrence, C., Petrick, J.
, et al., Analysis of plant-derived miRNAs in animal
small RNA datasets.
BMC genomics 2012,
13, 381.
[314] Tosar, J. P., Rovira, C., Naya, H., Cayota, A., Mining of public sequencing databases supports a non-
dietary origin for putative foreign miRNAs: underestimated effects of contamination in NGS.
RNA (New
York, N.Y.) 2014,
20, 754-757.
[315] Yang, J., Hirschi, K. D., Farmer, L. M., Dietary RNAs: New Stories Regarding Oral Delivery.
Nutrients
2015,
7, 3184-3199.
[316] Mazzara, M., Munaro, B., Grazioli, E., Savini, C.
, et al.,
JRC validated methods, reference methods and
measurements, Joint Research Centre – Institute for Health and Consumer Protection, Luxembourg 2013.
[317] Savini, C., Mazzara, M., Delobel, C., Pinski, G., Van den Eede, G.,
JRC validated methods, reference
methods and measurements, JRC- Institute for Health and Consumer Protection, Luxembourg 2012.
[318] Ricroch, A. E., Bergé, J. B., Kuntz, M., Evaluation of genetically engineered crops using transcriptomic,
proteomic, and metabolomic profiling techniques.
Plant physiology 2011,
155, 1752-1761.
[319] Kamthan, A., Chaudhuri, A., Kamthan, M., Datta, A., Small RNAs in plants: recent development and
application for crop improvement.
Frontiers in plant science 2015,
6, 208.
[320] Saurabh, S., Vidyarthi, A. S., Prasad, D., RNA interference: concept to reality in crop improvement.
Planta 2014,
239, 543-564.
[321] Jia, H., Wang, N., Targeted genome editing of sweet orange using Cas9/sgRNA.
PloS one 2014,
9,
e93806.
[322] Gao, J., Wang, G., Ma, S., Xie, X.
, et al., CRISPR/Cas9-mediated targeted mutagenesis in Nicotiana
tabacum.
Plant Molecular Biology 2015,
87, 99-110.
99
[323] Zhou, X., Jacobs, T. B., Xue, L. J., Harding, S. A., Tsai, C. J., Exploiting SNPs for biallelic CRISPR
mutations in the outcrossing woody perennial Populus reveals 4-coumarate:CoA ligase specificity and
redundancy.
The New phytologist 2015,
208, 298-301.
100
Appendix
7 Appendix
7.1 Literature Search
Literature searches were carried out using the databases PubMed, Scopus, Web of ScienceTM Core
Collection and Ovid® (Agris, Agricola, CAB Abstracts and Food Science and Technology Abstracts).
Retrieved references were combined in a library in Endnote X7 software (Thomson Reuters) and
duplicates were eliminated. Remaining references were checked manually for fulfilling the intended
search criteria by title and/or abstract screening. In few instances publications were included from
other sources or searches (webpages, random searches).
Literature searches ended March 2016.
7.1.1 CRISPR-Cas
Database searches to find literature relating to CRISPR-Cas application in plants were carried out
using the fol owing keywords: [(plant OR plants OR plant* OR “plant breeding”) AND crispr].
7.1.2 Rapid cycle breeding
Database searches to find literature relating to accelerated breeding in plants were carried out using
the following keywords: ("high speed breeding" OR "fast breeding" OR "FasTrack breeding" OR "Fast
Track breeding" OR "rapid cycle breeding" OR "accelerated breeding") AND plant*.
101
Appendix
7.2 Definition of GMO according to EU Directive 2001/18/EC on the
deliberate release into the environment of genetical y modified
organisms
Article 2
Definitions
For the purposes of this Directive:
(1) “organism” means any biological entity capable of replication or of transferring genetic material;
(2) ”genetically modified organism (GMO)” means an organism, with the exception of human beings,
in which the genetic material has been altered in a way that does not occur naturally by mating
and/or natural recombination;
Within the terms of this definition:
(a) genetic modification occurs at least through the use of the techniques listed in Annex I A, part 1;
(b) the techniques listed in Annex I A, part 2, are not considered to result in genetic modification;
Article 3
Exemptions
1. This Directive shall not apply to organisms obtained through the techniques of genetic
modification listed in Annex I B.
ANNEX I A
TECHNIQUES REFERRED TO IN ARTICLE 2(2)
PART 1
Techniques of genetic modification referred to in Article 2(2)(a) are inter alia:
(1) recombinant nucleic acid techniques involving the formation of new combinations of genetic
material by the insertion of nucleic acid molecules produced by whatever means outside an
organism, into any virus, bacterial plasmid or other vector system and their incorporation into a host
organism in which they do not naturally occur but in which they are capable of continued
propagation;
(2) techniques involving the direct introduction into an organism of heritable material prepared
outside the organism including micro-injection, macro-injection and micro-encapsulation;
(3) cell fusion (including protoplast fusion) or hybridisation techniques where live cel s with new
combinations of heritable genetic material are formed through the fusion of two or more cells by
means of methods that do not occur naturally.
102
Appendix
PART 2
Techniques referred to in Article 2(2)(b) which are not considered to result in genetic modification,
on condition that they do not involve the use of recombinant nucleic acid molecules or genetically
modified organisms made by techniques/methods other than those excluded by Annex I B:
(1)
in vitro fertilisation,
(2) natural processes such as: conjugation, transduction, transformation,
(3) polyploidy induction.
ANNEX I B
TECHNIQUES REFERRED TO IN ARTICLE 3
Techniques/methods of genetic modification yielding organisms to be excluded from the Directive,
on the condition that they do not involve the use of recombinant nucleic acid molecules or
genetically modified organisms other than those produced by one or more of the
techniques/methods listed below are:
(1) mutagenesis,
(2) cell fusion (including protoplast fusion) of plant cells of organisms which can exchange genetic
material through traditional breeding methods.
103
Appendix
7.3 Tables
Table 7.1 Summary of scientific publications in plants reporting analyses on off-target effects of CRISPR-Cas9 in genome editing (2013 – publications available
November 2015).
Target
Off-target candidate
Nr. of mismatches
locus
locus identification
distribution
Method of detection
Off-target activity
detected
Experimental system
Reference
A. thaliana
RACK1b/c
1 selected based on
2 mm
sequencing
none detected
transient
[93]*
homology
in seed region
cel culture
2 BLASTn searches of
candidate off-target sites were
none detected
stable transformation
[107]
spacer sequence
aligned against whole genome
germline transmission
GAI
against genome:
sequencing data of T1 (n=2) and T2
complete spacer, seed
mm </= 2
(n=1) GE lines
region only
GAI
4 selected based on
1-4 mm
60 T1 plants sequenced at each
none detected
stable transformation
[107]
homology
in/near seed region
locus
germline transmission
GAI
na
na
Are mutated target sites stable?
none detected
stable transformation
[107]
Re-sequencing GE lines in progeny
germline transmission
Cas-OFFinder
4-5 mm
Targeted deep sequencing
none detected
transient delivery of pre-
[27]
PHYB
3 candidates
distributed
assembled
(fewer mm not
ribonucleoprotein complex
detected)
into protoplasts
Cas-OFFinder
4-5 mm
Targeted deep sequencing
none detected
transient delivery of pre-
[27]
BRI1 gRNA1 6 candidates
Distributed
assembled
(fewer mm not
ribonucleoprotein complex
detected)
into protoplasts
Cas-OFFinder
2-5 mm
Targeted deep sequencing
none detected
transient delivery of pre-
[27]
BRI1 gRNA2 4 candidates
Distributed
assembled
(fewer mm not
ribonucleoprotein complex
detected)
into protoplasts
ETC2
Cas-OFFinder
> 4mm
Amplicon sequencing in 2 GE lines
none detected
stable transformation
[108]
3 candidates
in al last 7 seed nt
germline transmission
conserved
FT
Cas-OFFinder
3-4 mm
2 chosen off-target sites with 3 mm none detected
stable transformation
[120]
104
Appendix
Target
Off-target candidate
Nr. of mismatches
locus
locus identification
distribution
Method of detection
Off-target activity
detected
Experimental system
Reference
gRNA1
16 candidates
mm in seed region
were amplicon sequenced (n= 48)
germline transmission
in a T1 plant
FT
Cas-OFFinder
3-4 mm
1 chosen off-target site with 3 mm
none detected
stable transformation
[120]
gRNA2
12 candidates
mm in seed region
was amplicon sequenced (n= 48) in
germline transmission
a T1 plant
C. sinensis
BLASTn search of
4-7 mm
8 off-targets analyzed with
none detected
transient
[321]*
spacer sequence
Distributed
restriction enzyme suppressed PCR
Agroinfiltration of leaves
PDS
against genome
46 off-target sites
included
G. max
12g37050
1 candidate based on
1 mm in PAM
Sequencing in 15 GE lines
Yes (1 line identified)
soybean hairy root system
[122]#
homology
NGGNAG
BLASTn (e value
2-6 mm
Amplicon sequencing
none detected
soybean hairy root system
[134]
07g14530
threshold 5)
Distributed
(n=10)
10 candidate loci
DDM1
BLASTn (e value
4 mm
Amplicon sequencing
none detected
soybean hairy root system
[134]
gRNA1
threshold 5)
Distributed
(n=10)
1 candidate loci
DDM1
BLASTn (e value
2 mm
Amplicon sequencing
Yes, in al experimental
soybean hairy root system
[134]
gRNA2
threshold 5)
seed region
(n=10)
repeats
1 candidate locus
BLASTn (e value
3 mm
Amplicon sequencing
none detected
soybean hairy root system
[134]
Met1
threshold 5)
Distributed
(n=5)
1 candidate locus
BLASTn (e value
6 and 2 mm
Amplicon sequencing
yes, gRNA with 2 mm in
soybean hairy root system
[134]
miR1514
threshold 5)
Non-seed region
(n=4)
non-seed region in al
2 candidate loci
experimental repeats
H. vulgare
2 candidates based on
1 mm in seed region
Sequencing in 93/95 T1 individuals
Yes, gRNA with mm
stable transformation
[102]
HvPM19-1
homology
each
of two independent T0 lines
(further away from PAM
than 2nd off-target) in
seed region, 3/93
105
Appendix
Target
Off-target candidate
Nr. of mismatches
locus
locus identification
distribution
Method of detection
Off-target activity
detected
Experimental system
Reference
individuals
HvPM19-3
2 candidates based on
1mm in seed r.
Sequencing in 76 T1 individuals of
None detected
[102]
homology
3 mm distributed
one T0 line
L. sativa
Cas-OFFinder
2-5 mm
High throughput sequencing of 92
none detected
transient delivery of pre-
[27]
349 candidate loci
candidate sites in 3 GE lines
assembled
BIN2
ribonucleoprotein complex
into protoplasts,
regeneration of plants
N.
benthamian
a
PDS
BLASTn
2-10 mm
None not conclusive
Transient
[96]
98 candidates
for me
Agro-infiltration of leaves
BLASTn search of
5-7 mm
T7EI restriction assay of 13
none detected
Transient
[25]
PDS
spacer against genome
candidate sequences, n=?
Agro-infiltration of TRV
vector in stably expressing
Cas9 plants
3 candidates reported
1, 3, 5 mm
Restriction enzyme suppressed PCR none detected
Transient
[103]
PDS
by Nekrasov
et al.,
n=5
Agro-infiltrated leaves
2013
N. tabacum
BLASTn search of
2 mm
Sequencing of PCR fragment in GE
none detected
stable transformation
[322]*
PDR6
spacer against genome Non-seed region
lines (n=?)
1 candidate found
P. tremula ×
alba
4CL1
1 candidate selected
3 mm
Amplicon sequencing
none detected
stable transformation
[323]
based on homology
Seed region
in 8 GE lines
4CL5
4CL5 in variety with
1 mm seed region
Amplicon sequencing
none detected
stable transformation
[323]
gRNA
natural SNPs
1 mm PAM
in 10 transgenic lines
S.
tuberosum
IAA2
BLASTn search of
1 mm
PCR sequencing of 6 GE lines
none detected
stable transformation
[103]
spacer against genome PAM
106
Appendix
Target
Off-target candidate
Nr. of mismatches
locus
locus identification
distribution
Method of detection
Off-target activity
detected
Experimental system
Reference
1 candidate found
O. sativa
BLASTn search of
3 mm distributed
RE suppressed PCR of 3 selected
Yes, activity detected at
Transient transformation
[99]*
MPK5
spacer against genome 3 mm distributed
candidates
off-target site with 3 mm protoplasts
11 candidates
5 mm distributed
which start furthest from
PAM
BLASTn search of
3 mm
PCR – RE assay
none detected
Transient transformation
[97]*
PDS
spacer against genome distributed
protoplasts
1 candidate
BLASTn search of
1 mm non-seed
PCR – RE assay
Yes, potential y detected Transient transformation
[97]*
MPK2
spacer against genome
##
protoplasts
2 candidates
1 mm seed
PCR – RE assay/sequencing
none detected
Selected based on
3-5 mm
Sequencing at target locus in 20 GE none detected
stable transformation
[124]*
DERF1
homology
2 only in non-seed
lines (T0 and T1, al independent
5 candidates
region
lines)
Selected based on
3-5 mm
Sequencing at target locus in 20 GE none detected
stable transformation
[124]*
MYB1
homology
2 only non-seed region
lines (T0 and T1, al independent
3 candidates
(5 mm)
lines)
Selected based on
1-7 mm
Sequencing at target locus in ~70
Yes, at 1 candidate locus stable transformation
[124]*
homology
2 only non-seed region
Cas9 positive lines (independent T0 7 plants with off-target
YSA1
5 candidates
(1 and 7 mm)
lines)
activity: locus with 1 mm
in non-seed region
SWEET13
Bioinformatics
>/= 16 identical sites
Sequencing of 7 T0 lines at 6
none detected
stable transformation
[125]*
6 candidates
candidate loci
BLASTn search of
1 seed
Sequencing ~ 80 plants
none detected
stable transformation
[123]
BEL1
spacer against genome 3 seed/non-seed
3 candidates detected
3 seed/non-seed
CRISPR-P
3, 4 mm
Sequencing of target locus
none detected (50 plants stable transformation
[109]
AOX1a
Selected 2 highest
distributed
of T0 and T1)
ranked
CRISPR-P
3, 4 mm
Sequencing of target locus
none detected (49 plants stable transformation
[109]
AOX1b
Selected 2 highest
distributed
of T0 and T1)
ranked
AOX1c
CRISPR-P
2, 3 mm
Sequencing of target locus
none detected (60 plants stable transformation
[109]
107
Appendix
Target
Off-target candidate
Nr. of mismatches
locus
locus identification
distribution
Method of detection
Off-target activity
detected
Experimental system
Reference
Selected 2 highest
distributed
of T0 and T1)
ranked
CRISPR-P
1 mm non seed r.
Sequencing of target locus
Yes, activity detected in
stable transformation
[109]
BEL
Selected 2 highest
3 mm distributed
2 plants at locus with 1
ranked
mm (89 plants of T0 and
T1)
3 candidates selected
1 mm non seed r.
CAPS marker, sequencing
Yes, activity detected
stable transformation
[114]
based on homology,
(6/13 regenerated
confirmed by CRISPR-P
plants)
2 mm seed/non-seed
Yes, activity detected
(10/13 regenerated
plants)
2 mm seed/non seed
none detected (0/13):
mm nearest to PAM
(al regenerated plants
CDKB2
from 1 transformation
event (cal us); result
repeatable in 3 further
trasnsformation events
(cal i))
Further 3 candidates
CAPS marker
none detected
ranked 3, 5, 9 by
CRISPR-P
T. aestivum
Set of spacers with
1-11 mm distributed
PCR-RE analysis
Off-target activity
Transient
[98]*
INOX
random mutations
detected in case mm are Protoplast cel culture
in non-seed region
na: not applicable; mm: mismatches
# also report off-target activity with second target, however in that case both loci are 100% identical at spacer and PAM sequence
## off-target site very close to target site
*taken from [90]
108
www.bmgf.gv.at
Document Outline
- Studie_Techniken_Pflanzenzüchtung_engl_rot_20170322.pdf