See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/330812010

Development of an Agrobacterium‐delivered CRISPR/Cas9 system for wheat

genome editing

Article  in  Plant Biotechnology Journal · February 2019

DOI: 10.1111/pbi.13088

CITATIONS READS

8 248

7 authors, including:

Zhengzhi Zhang Lei Hua

University of Missouri South Dakota State University
49 PUBLICATIONS   719 CITATIONS    3 PUBLICATIONS   39 CITATIONS   

SEE PROFILE SEE PROFILE

Ajay Gupta Bing Yang

South Dakota State University University of Missouri
1 PUBLICATION   8 CITATIONS    74 PUBLICATIONS   3,368 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

CRISPR/Cas9 View project

plant-pathogen interaction View project

All content following this page was uploaded by Lei Hua on 28 July 2019.

The user has requested enhancement of the downloaded file.

Plant Biotechnology Journal (2019) 17, pp. 1623–1635 doi: 10.1111/pbi.13088

Development of an Agrobacterium-delivered CRISPR/Cas9

system for wheat genome editing
Zhengzhi Zhang1,†, Lei Hua2,†, Ajay Gupta2, David Tricoli3, Keith J. Edwards4, Bing Yang1,5,* and Wanlong Li2,*
1
Division of Plant Sciences, University of Missouri, Columbia, MO, USA
2
Department of Biology and Microbiology, South Dakota State University, Brookings, SD, USA
3
Plant Transformation Facility, University of California, Davis, CA, USA
4
Functional Genomics, University of Bristol, Bristol, UK
5
Donald Danforth Plant Science Center, St. Louis, MO, USA

Received 26 October 2018; Summary

revised 19 January 2019; CRISPR/Cas9 has been widely used for genome editing in many organisms, including
accepted 22 January 2019. important crops like wheat. Despite the tractability in designing CRISPR/Cas9, efficacy in the
*Correspondence (Tel 6056885743;
application of this powerful genome editing tool also depends on DNA delivery methods. In
fax 6056885624; email
wheat, the biolistics based transformation is the most used method for delivery of the CRISPR/
[email protected]; and
Cas9 complex. Due to the high frequency of gene silencing associated with co-transferred
Tel 5738827705; fax 5738849676;
email [email protected]) plasmid backbone and low edit rate in wheat, a large T0 transgenic plant population are
†
These authors equally contributed to this required for recovery of desired mutations, which poses a bottleneck for many genome
work. editing projects. Here, we report an Agrobacterium-delivered CRISPR/Cas9 system in wheat,
which includes a wheat codon optimized Cas9 driven by a maize ubiquitin gene promoter and
a guide RNA cassette driven by wheat U6 promoters in a single binary vector. Using this
CRISPR/Cas9 system, we have developed 68 edit mutants for four grain-regulatory genes,
TaCKX2-1, TaGLW7, TaGW2, and TaGW8, in T0, T1, and T2 generation plants at an average
edit rate of 10% without detecting off-target mutations in the most Cas9-active plants.
Homozygous mutations can be recovered from a large population in a single generation.
Different from most plant species, deletions over 10 bp are the dominant mutation types in
wheat. Plants homozygous of 1160-bp deletion in TaCKX2-D1 significantly increased grain
number per spikelet. In conclusion, our Agrobacterium-delivered CRISPR/Cas9 system provides
Keywords: Agrobacterium mediation, an alternative option for wheat genome editing, which requires a small number of
CRISPR/Cas9, genome editing, grain transformation events because CRISPR/Cas9 remains active for novel mutations through
regulation. mutations, wheat. generations.

et al., 2013a; Wang et al., 2014; Wright et al., 2005; Zhang

Introduction
et al., 2015). In contrast, the CRISPR/Cas9 system consists of a
Targeted mutagenesis or precise genomic alteration has long short guide RNA (sgRNA) and a conditional DNA nuclease Cas9
been a dream of biologists and biotechnologists. This dream has (Jinek et al., 2012). The 20 nucleotides (nt) at the 50 end of the
come true thanks to the rapid development of genome editing sgRNA guide the Cas9/sgRNA complex to search for the target
technologies in the last decade using engineered nucleases such sequence along the chromosomal DNA until an exact match is
as zinc finger nucleases (ZFN) (Carroll, 2011), TAL effector found. If an NGG trinucleotide, the protospacer adjacent motif
nuclease (TALEN) (Cermak et al., 2011; Li et al., 2011) and (PAM), is located immediately downstream of the target site,
Cluster Regularly Interspaced Short Palindromic Repeats (CRISPR)- Cas9 undergoes a conformational change, which activates two
associated (Cas) systems (Jinek et al., 2012). These systems separate nuclease domains in Cas9 and leads to cleavage of the
operate in a similar fashion to create a double strand break (DSB) target (Jinek et al., 2014). Since the first success in editing genes
at a preselected and defined genomic site. The DSB is subse- in cultured human cell lines (Cong et al., 2013; Mali et al., 2013),
quently repaired by the error-prone non-homologous end joining application of the CRISPR/Cas9 system has been rapidly expanded
(NHEJ) or by the error-free homologous recombination (HR) in the into plants and ushered a flood of publication in last several years
presence of a homologous DNA template (Symington and (reviewed in references: Soyars et al., 2018; Weeks et al., 2016).
Gautier, 2011). While HR leads to gene correction or replace- The predominant use of CRISPR/Cas9 technology in plants is to
ment, NHEJ causes insertion, deletion and other mutations at the target genes of interest for knockout. More recently, the CRISPR
cleavage locus. Engineered ZFN and TALEN nucleases involve system was engineered by Cas9 nickase fused with cytidine
selection and assembly of modular DNA binding units and have deaminase or adenosine deaminase for highly precise base editing
been successfully applied to plants (Cai et al., 2009; Christian (Gaudelli et al., 2017; Komor et al., 2016; Nishida et al., 2016),
et al., 2013; Clasen et al., 2015; Haun et al., 2014; Li et al., which has also been adopted in plants. With these features, the
2011, 2012; Maeder et al., 2008; Mahfouz et al., 2011; Shan engineered CRISPR systems have been playing an important role

ª 2019 The Authors. Plant Biotechnology Journal published by Society for Experimental Biology and The Association of Applied Biologists and John Wiley & Sons Ltd. 1623
This is an open access article under the terms of the Creative Commons Attribution License, which permits use,
distribution and reproduction in any medium, provided the original work is properly cited.
1624 Zhengzhi Zhang et al.

in the basic understanding of plant biology and crop (68.5%  7.9%) than the rest three U6 promoters (P < 0.01).
improvement. TaU6.1 ranked second in editing efficiency, but it was not
Common wheat, bread wheat, or hexaploid wheat (Triticum significantly different from TaU6.2 (P > 0.08) and TaU6.5
aestivum L., genomes AABBDD) is the most widely cultivated (P = 0.28). In term of average GFP fluorescence intensity per
crop. It provides ~20% of our daily diets and thus plays an cell, TaU6.3 also showed the best efficiency among the
important role in global food security and rural economy. To meet promoters tested (P < 0.05). This result indicated that the Cas9
the demand for the increasing population, we must increase functions properly in wheat cells, and TaU6.1 and TaU6.3 are
wheat yield by 50% by the year 2034 (http://iwyp.org/). In this best candidates for expressing guide RNAs in wheat. Thus,
respect, the CRISPR technology is expected to contribute greatly TaU6.1 was used to drive the sgRNA1 inserted into the BtgZI
to creating novel variations in wheat. Wheat was one of the plant sites and TaU6.3 to drive the sgRNA2 inserted into the BsaI sites
species first modified by CRISPR/Cas9 (Shan et al., 2013b), and in pTagRNA4 (Figure S2).
several traits have been targeted for mutations, including To establish an Agrobacterium-based wheat CRISPR/Cas9
powdery mildew resistance (Wang et al., 2014), grain size (Wang system, we chose a binary vector pLC41 (Japan Tobacco, Tokyo,
et al., 2018; Zhang et al., 2018b), and grain quality (Liang et al., Japan) that was modified as a Gateway recipient vector. A
2017; Sanchez-Leo n et al., 2018; Zhang et al., 2018b). In all pENTR derived intermediate vector was constructed to express
these researches, the Cas9 and sgRNA transgenes were delivered Cas9 under the strong maize ubiquitin gene promoter and
by the biolistic bombardment. In recent years, the efficiency of for the construction of guide RNA genes under U6 promoters
the Agrobacterium-mediated transformation has been improved (Figure S3).
significantly in wheat (Ishida et al., 2015; Richardson et al., 2014;
Identification of grain regulatory genes and design of
Wang et al., 2017a). Compared to the biolistics approach, the
guide RNAs
improved Agrobacterium-mediation systems increased transfor-
mation efficiency and expanded the transformability of wheat In a previous study, we identified 45 orthologus loci as candidate
genotypes. These improvements have not been integrated into genes for wheat grain regulatory genes, 23 of which are negative
CRISPR/Cas9-based genome editing or other genome editing regulators or expression is negatively regulated by microRNAs (Li
platforms in wheat. During the preparation of this manuscript, a and Yang, 2017). From these 23 genes, we selected two negative
report was published on Agrobacterium-delivered CRISPR/Cas9 regulators TaCKX2-1 and TaGW2, and two microRNA-regulated
for wheat gene editing in which one gene (DA1) was tested in T0 positive regulators TaGLW7 and TaGW8 for the present research,
transgenic plants with a high frequency of not yet proven two guide RNAs were designed for each gene to target the
inheritable mutations (Zhang et al., 2018a). In the present sequences conserved among the A, B and D genomes of the
research, we developed an Agrobacterium-delivered CRISPR/ hexaploid wheat Fielder (Table S1). For TaCKX2-1 and TaGW2,
Cas9 system for genome editing in wheat and applied it to guide RNA sequences are located in a conserved region of their
target four grain-size regulatory candidate genes. Our result first exon, and perfectly matched their targets. An SNP was found
showed that a small number of transformation events are enough in the first position of PAM motif (50 -NGG-30 ) in TaCKX2-D1
for recovering desired mutations in T0, T1, and T2 plants (Figures 2a and 3b), it should not affect the editing efficiency
attributed to the continuing activity of CRISPR/Cas9 transgene because the nucleotide at that position is degenerate. For
through generations. TaGLW7 and TaGW8, the guide RNAs target the microRNA
recognition sites (Figure 2c,d), the two guide RNA target
Results sequences are overlapped on two strands. For TaGLW7 gene, a
single nucleotide mismatch is found in the guide RNA target
Development of a CRISPR/Cas9 system for wheat
sequences among the homoeologous genes. Accordingly, two
genome editing
guide RNA genes were designed with sgRNA1 targeting the
We first sought for strong U6 promoters from wheat genome to TAGLW7-A and sgRNA2 targeting TaGLW7-B and TaGLW7-D
express guide RNA genes. Promoters from four wheat U6 (Figure 2c).
RNA coding genes, i.e., TaU6.1, TaU6.2, TaU6.3, and TaU6.5
Agrobacterium-delivery of CRISPR/Cas9 system into
(Appendix S1), were tested individually for their abilities to
wheat cells
express the guide RNA gene gGFP, which targets the 1-bp
insertion site in the mutated GFP gene in the pUC-35S:GFP+1 Four CRISPR constructs were individually introduced into Fielder
for restoration of GFP, together with wheat codon-optimized immature embryos through Agrobacterium-mediated gene trans-
Cas9 in wheat protoplasts (Figure 1a). To do so, we first fer. After selection and regeneration, eight plants for TaCKX2-1,
evaluated the transfection rate of wheat protoplasts using a four for TaGW2, two for TaGLW7, and eight plants for TaGW8
pOsUbi-GFP construct as a control. On average, 52.8% of were obtained. Plants grow normally in growth chamber or
protoplasts were successfully transfected by pOsUbi-GFP (Fig- greenhouse without obvious morphological alteration. PCR
ure 1b). When transfected with the pUC-35S:GFP+1 construct, screening with Cas9 and sgRNA specific primers identified that
in contrast, no GFP signal was observed in the protoplasts all plants except one from TaCKX2-1, contain both Cas9 and
(Figure 1c), indicating that the 1-bp frameshift insertion com- sgRNA (Table S2). Gene-specific primers flanking the target
pletely abolished the GFP-coding capacity. When transfected region were used successfully to PCR-amplify the expected
together with the U6 promoter-driven guide RNA gene gGFP fragments from all three loci of each gene. However, a T7
and the ZmUbi-Cas9, the GFP signal was detected in the endonuclease 1 (T7E1) assay with PCR products did not reveal the
protoplasts (Figure 1d–g). After normalization with the transfec- expected digested pattern for the InDel (insertion/deletion)
tion rate from pOsUbi-GFP, the desired editing occurred mutation. Sanger sequencing of the PCR products confirmed
at 47.4%–68.5% in four U6 promoters tested (Figure S1). that all examined plants carry only the wild type allele for
Among them, TaU6.3 showed significantly higher editing rate TaCKX2-1, TaGW2, TaGLW7, and TaGW8.

ª 2019 The Authors. Plant Biotechnology Journal published by Society for Experimental Biology and The Association of Applied Biologists and John Wiley & Sons Ltd., 17, 1623–1635
 Agrobacterium-delivered wheat CRISPR/Cas9 system 1625

Figure 1 Expression of GFP in protoplasts transfected with Cas9, sgGFP, and non-functional GFP. (a) Strategy to test the wheat CRISPR-Cas9 system in
protoplasts. (b). pOsUbi-GFP; (c) pUC-35S:GFP+1; (d) pTaU6.1-gGFP; (e) pTaU6.2-gGFP; (f) pTaU6.3-gGFP; (g) pTaU6.5-gGFP. BF: Bright field; FITC:
Fluorescein isothiocyanate. Scale bars indicate 50 lm.

4.0%, 4.6%, and 3.1% for TaCKX2-A1, TaCKX2-B1 and

Detection of mutations in T0 plants by deep sequencing
TaCKX2-D1, respectively. By contrast, insertion frequency at
We were not able to detect a mutation in T0 plants by T7E1 assay these loci are 1.2%, 0.5%, and 0.2%, respectively. There are also
and Sanger sequencing directly on PCR products. This result could rare cases of complex editing (simultaneous deletion and inser-
be due to low edit frequency in wheat as shown in prior studies tion) in TaCKX2-B1 and TaCKX2-D1, but they all happened at
(Wang et al., 2014, 2018), but somatic mutations may have <0.05% (Figure 3).
occurred in the T0 plants. To test this hypothesis, we sequenced The lengths of the deletion range from 1 to 37 bp in TaCKX2-
the PCR products derived from the TaCKX2-1 target sites by the A1, 1 to 47 bp in TaCKX2-B1, and 1 to 72 bp in TaCKX2-D1.
HiSeq 2000 platform. A total 1 101 752 raw reads were obtained Interestingly, a 37-bp deletion from position 136 to position 172
from the deep sequencing, of which 486 890 reads meet the downstream of the translation start site (ATG) occurs at a high
quality standards. Approximately 48% and 47% of the quality frequency. It is the most abundant mutant allele in TaCKX2-A1
reads were mapped to the TaCKX2-B1 and TaCKX2-D1, respec- and TaCKX2-B1, with 22.5% and 19.3% of the total mutations,
tively, but only 0.2% of them were mapped to the TaCKX2-A1. respectively. This deletion, occurred at 10% in TaCKX2-D1, is the
Both TaCKX2-A1 and TaCKX2-B1 showed 5.2% of mutation rate, third most abundant mutation next to the 1-bp deletion at
while TaCKX2-D1 had a lower mutation frequency of 3.3% position 165 (16.7%) and 2-bp deletion at positions 172 and 173
(Figure 3). Deletions occurred at much higher frequencies than (12.6%). In TaCKX2-A1 and TaCKX2-B1, the 1-bp and 2-bp
insertions in all three homoeologous genes. Deletion frequency is deletions at 3 bp upstream of the PAM sites of two sgRNAs rank

ª 2019 The Authors. Plant Biotechnology Journal published by Society for Experimental Biology and The Association of Applied Biologists and John Wiley & Sons Ltd., 17, 1623–1635
1626 Zhengzhi Zhang et al.

(a)
TaCKX2-1
XcmI
TaCKX2-A1: 5’- CCCAACGCCACCCTGGCGGCCTCCACGGACTTCGGCAACATCACGGCGG- 3’
TaCKX2-B1: 5’- CCCAACGCCACCCTGGCGGCCTCCACGGACTTCGGCAACATCACGGCGG- 3’
TaCKX2-D1: 5’- CCAAACGCCACCCTGGCGGCCTCCACGGACTTCGGCAACATCACGGCGG- 3’
BglI 5’- GGACUUCGGCAACAUCACGG- 3’
3’- UUGCGGUGGGACCGCCGGAG- 5’ sgRNA2
sgRNA1

(b)
TaGW2

TaGW2-A: 5’- CCAGAAGAAGCTACGCAAGTTGATCCTCGAGGCCAAGCTCGCGCCC- 3’

TaGW2-B: 5’- CCAGAAGAAGCTACGCAAGTTGATCCTCGAGGCCAAGCTCGCGCCC- 3’
TaGW2-D: 5’- CCAGAAGAAGCTACGCAAGTTGATCCTCGAGGCCAAGCTCGCGCCC- 3’
3’- CUUCUUCGAUGCGUUCAAC GCUCCGGUUCGAGCGCGGG- 5’
sgRNA1 sgRNA2

(c)
sgRNA1
3’- AGAGAAGACAGUUCGUCGG- 5’
TaGLW7-A:5’- TGCTCCCTCTCTTCTGTCAAGCAGCCAT- 3’

TaGLW7

TaGLW7-B: 5’- TGCTCCCTCTCTTCTGTCATGCAGCCAT- 3’

TaGLW7-D: 5’- TGCTCCCTCTCTTCTGTCATGCAGCCAT- 3’
3’- GAGAAGACAGUACGUCGGUA- 5’
sgRNA2

(d)
sgRNA2
3’- AGAAGACAGUAGGGGCCGG- 5’
TaGW8-A: 5’- CCGACTGTGCTCTCTCTCTTCTGTCATCCCCGGCC- 3’
TaGW8-B: 5’- CCGACTGTGCTCTCTCTCTTCTGTCATCCCCGGCC- 3’
TaGW8-D: 5’- CCGACTGTGCTCTCTCTCTTCTGTCATCCCCGGCC- 3’

TaGW8

TaGW8-A: 5’- CCGACTGTGCTCTCTCTCTTCTGTCATCCCCGGCC- 3’

TaGW8-B: 5’- CCGACTGTGCTCTCTCTCTTCTGTCATCCCCGGCC- 3’
TaGW8-D: 5’- CCGACTGTGCTCTCTCTCTTCTGTCATCCCCGGCC- 3’
3’- UGACACGAGAGAGAGAAGAC- 5’
sgRNA1

Figure 2 Schematics of gene structures with exons (black bars), introns (black lines) and sequences of target sites and guide RNAs. (a and b) Structures of
TaCKX2-1 and TaGW2 with target sites in exon 1. Nucleotides in red and green indicate target sites and PAM sequences, respectively. Underlined
nucleotides represent the single nucleotide polymorphism (SNP) in three homeologues. Boxed letters indicate the restriction enzyme site. The solid red bars
are the conserved region of the target genes. (c) Structures of TaGLW7 with target sites located in 30 UTR and (d) structures of TaGW8 with the target sites
in exon 3. Highlighted yellow regions indicate microRNA recognition sites.

ª 2019 The Authors. Plant Biotechnology Journal published by Society for Experimental Biology and The Association of Applied Biologists and John Wiley & Sons Ltd., 17, 1623–1635
 Agrobacterium-delivered wheat CRISPR/Cas9 system 1627

products size being ~1.7 kb to screen the 44 Cas9-positive plants in

the #4 T1 family (Figure S5a,b). The PCR and Sanger sequencing
results showed that 16 mutant plants carry an identical mutation, a
complex mutation with deletion of 1160 bp in TaCKX2-D1
(Figure S7), of which three are homozygous, and 13 are heterozy-
gous (Table S5), indicating the deletion occurred in T0 and
transmitted to T1 generation. It was not detected by deep
sequencing because the deletion was much larger than the insert
size of the sequencing library. Thus, the edit rate in the T0 plants is
12.5% (1 in 8 T0 plants) for TaCKX2-D1. Four new mutations were
detected in TaCKX2-A1 and TaCKX2-B1, each of which contained
two heterozygous deletion mutants (Figure S6; Table S3). Thus,
mutagenesis frequencies of the TaCKX2-A1 and TaCKX2-B1 were
4.5% (2/44) in the T1 generation.
For TaGW2, a 1-bp deletion in TaGW2-A, and a 17-bp deletion
in TaGW2-D were detected in 15 Cas9-positive T1 plants with an
edit efficiency of 13.3% (Table S3; Figure 5b; Figure S5c). For
TaGLW7, we found two mutants from 74 T1 plants derived from
two T0 plants with an edit efficiency of 2.7%, and both mutations
Figure 3 Types and frequencies of mutation in T0 plants discovered by were deletions within the microRNA recognition site of TaGLW7-
PCR amplicon deep sequencing.
A (Figure 5c). Of 93 TaGW8 T1 plants, one mutant with a 16-bp
deletion was identified (Figure 5d), with an edit frequency of
in abundance just next to the 37-bp deletion (Figure 4). Overall, 1.1%. Taken together, edit efficiency in T1 progeny varied among
the deletion frequency at the sgRNA2 target site is higher than the targets. Of the 25 mutated plants among four target genes,
that at the sgRNA1 target site (Figure S4), suggesting that TaU6.3 ten types of mutations were detected, and all of them were
may be more active in transcription of sgRNA, or sgRNA2 for deletions ranging from 1 to 1160 bp (Figure 5a). Because both
TaCKX2-1 had the better capability in directing the Cas9-sgRNA target sites and gene-specific PCR primer binding sites were
binding to the target region. These results showed that our within the region of the 1160-bp deletion (complex edit with
CRISPR/Cas9 system did work in the T0 transgenic plants. 880- and 280-bp deletions and a 1-bp substitution) (Table S4) in
TaCKX2-D1, it is possible that the mutated allele was initially
Screening mutations in T1 generation
missed with the primers in the T0 plant #4.
Considering that the CRISPR/Cas9 system functions in the somatic We also screen T1 populations for transgene segregation.
cells of the T0 transgenic plants, we expected mutations would Transgenes segregated into 15 (present) to 1 (absent) in
also occur in gametes and embryonic cells, and screened T1 TaGLW7_T1-32 and _T1-38 families (P > 0.55698) and into 63
progenies carrying the Cas9 transgene for targeted mutations. To (present) to 1 (absent) in TaGW2 T1-12 family (P = 0.824132),
investigate CRISPR/Cas9 induced mutation at the sgRNA target indicating two and three copies of transgenes in the T0 transgenic
sites, we used three gene-specific primer pairs to amplify each plants, respectively. From these populations, we identified two
homoeologue, which contained the genomic region of two sgRNA transgene-free mutants, one for TaGLW7-A and one for TaGW2-D.
target sites, and Sanger sequencing of PCR amplicons for all T1
Inheritance of induced mutations in T2 progenies
plants was performed. The PCR restriction enzyme digestion assay
(PCR-RE) and T7E1 cleavage assay were not conducted for To confirm inheritance of TaCKX2-1 mutations in T2 generation,
independent T1 progenies, as only one restriction site existed in two T1 lines, #4-1 that contained a 24-bp deletion and a 1-bp
the target sequence of TaCKX2-1 sgRNA1, while T7E1 assay substitution mutation in TaCKX2-A1, and #4-35 that carried a 1-
sometime failed to screen homozygous mutants and heterozy- bp deletion in TaCKX2-B1, were allowed to self-pollinate for
gous mutants with small deletion or insertion mutation. Twenty- generation of T2 populations. Totally, 24 T2 individuals were
five plants containing 10 types of mutations were identified for Cas9-positive and carried mutations at the TaCKX2-A1 and
four target genes from 226 T1 plants (Tables S3 and S5). TaCKX2-B1. Of 17 T2 plants from #4-1, two were homozygous
For TaCKX2-1 gene, we screened three T1 populations derived mutants, eight heterozygous mutants and seven wild type plants
from T0 plants #1, #2, and #4. From the 44 T1 plants of the #4 (Figure S8a), while two homozygous mutants, four heterozygous
family, we detected two mutations in TaCKX2-A1 and two mutants, and one wild type plant were identified from the T2
mutations in TaCKX2-B1. A 24-bp deletion and 1-bp substitution plants of #4-35 by using PCR-RE assay (Figure S8b). The
(#4-1) and a 499-bp deletion (#4-2) were found in TaCKX2-A1 segregation ratios fit the 1:2:1 expectation (P > 0.22313). Sanger
(Figure 5a; Figure S6b), and a 11-bp deletion (#4-31) and 1-bp sequencing results showed that all the T2 mutation types in the
deletion (#4-35) were identified in TaCKX2-B1 (Figure 5a; Fig- TaCKX2-A1 and TaCKX2-B1 were consistent with their T1
ure S6c). Two plants #4-31 and #4-35 showed no PCR amplifica- mutations. We screened 48 T2 progenies from the T1 mutants
tion in TaCKX2-D1, while PCR amplicons were amplified in both for TaGW2-A and TaGW2-D. Their segregation also followed a
TaCKX2-A1 and TaCKX2-B1 (Figures S6b, S6c, and S6d). Using a 1:2:1 ratio (P > 0.14370). These results indicate that the CRISPR/
long-range PCR primer pair, we amplified a product of ~2.5 kb in Cas9 induced mutations are heritable in wheat.
the wild type and ~1.3 kb in homozygous mutant plants (Fig-
Edit mutations in the T2 and T3 generations
ure S6d), suggesting that ~1.2 kb deletion occurred in TaCKX2-D1.
As the ~2.5 kb band is weak in wild type and heterozygous plants We found one new mutation in the T2 progenies of #4-1 (#4-1-9),
(Figure S6d), a new primer pair were designed with the PCR which was due to a complex deletion of 218, 18 and 13 bp at the

ª 2019 The Authors. Plant Biotechnology Journal published by Society for Experimental Biology and The Association of Applied Biologists and John Wiley & Sons Ltd., 17, 1623–1635
1628 Zhengzhi Zhang et al.

TaCKX2-A1
GCGACCCCAACGCCACCCTGGCGGCCTCCACGGACTTCGGCAACATCACGGCGGCG WT
GCGACCCCAACGCCACCCTGGCGGCCTCCACGGACTTCGGCAACATCACtGGCGGCG +1 bp (20%)
GCGACCCCAAC-CCACCCTGGCGGCCTCCACGGACTTCGGCAACATCACGGCGGCG -1 bp (17%)
GCGACCCCAACGCCACCCTGGCGGCCTCCACGGACTTCGGCAACATC-CGGCGGCG -1 bp (17%)
GCGACCCCAACGCCACCCTGGCGGCCTCCACGGACTTCGGCA--------GCGGCG -8 bp (2.8%)
GCGACCCCAACG------------------------------------CGGCGGCG -36 bp(8.5%)
GCGACCCCAAC-------------------------------------CGGCGGCG -37 bp(22.5%)

TaCKX2-B1
GCGACCCCAACGCCACCCTGGCGGCCTCCACGGACTTCGGCAACATCACGGCGGCG WT
GCGACCCCAACGtCCACCCTGGCGGCCTCCACGGACTTCGGCAACATCACGGCGGCG +1 bp (4.4%)
GCGACCCCAAC-CCACCCTGGCGGCCTCCACGGACTTCGGCAACATCACGGCGGCG -1 bp (4.4%)
GCGACCCCAACGCCACCCTGGCGGCCTCCACGGACTTCGGCAACATC-CGGCGGCG -1 bp (11.3%)
GCGACCCCAACGCCACCCTGGCGGCCTCCACGGACTTCGGCAACATC--GGCGGCG -2 bp (8.9%)
GCGACCCCAACGCCACCCTGGCGGCCTCCACGGACTTCGGCAACA---CGGCGGCG -3 bp (4.0%)
GCGACCCCAACGCCACCCTGGCGGCCTCCACGGACTTCGGCAAC-----GGCGGCG -5 bp (4.7%)
GCGACCCCAACG------------------------------------CGGCGGCG -36 bp(4.3%)
GCGACCCCAAC-------------------------------------CGGCGGCG -37 bp(19.3%)
GCGACCCCAAC--------------------------------------GGCGGCG -38 bp(1.0%)

TaCKX2-D1
GCGACCCAAACGCCACCCTGGCGGCCTCCACGGACTTCGGCAACATCACGGCGGCG WT
GCGACCCAAAC--CACCCTGGCGGCCTCCACGGACTTCGGCAACATCACGGCGGCG -2 bp (3.7%)
GCGACCCAAAC-----CCTGGCGGCCTCCACGGACTTCGGCAACATCACGGCGGCG -5 bp (1.3%)
GCGACCCAAACGCCACCCTGGCGGCCTCCACGGACTTCGGCAACATC-CGGCGGCG -1 bp (16.7%)
GCGACCCAAACGCCACCCTGGCGGCCTCCACGGACTTCGGCAACATC--GGCGGCG -2 bp (12.6%)
GCGACCCAAACGCCACCCTGGCGGCCTCCACGGACTTCGGCAAC-----GGCGGCG -5 bp (6.2%)
GCGACCCAAACGCCACCCT------------------------------GGCGGCG -30 bp(2.8%)
GCGACCCAAAC-------------------------------------CGGCGGCG -37 bp(10.0%)

Figure 4 Representative InDel mutations in TaCKX2-1 from T0 transgenic plants. Green letters indicate the PAM sequences. The dashed lines represent
nucleotide deletions. Insertion is shown by lower case letters in black. The number in the parenthesis represents the frequency of the mutation type in the
total mutations detected.

TaCKX2-D1 locus. No additional mutation was detected in the T2 sperm cells. From T2 and T3 populations, we recovered 35
progenies of line #4-35 due to small population size. To further homozygous double mutants (mutations in two of the three
investigate whether new mutations were induced by CRISPR/Cas9 homoeologous genes) and 33 triple mutants (mutations in all
system in the T2 generation, we screened progenies of three three homeologous genes), ten of which are homozygous in two
Cas9-positive T1 plants, #4-6, #4-10, and #4-18 for analysis of homoeologous loci of TaCKX2-1 (Table S7).
TaCKX2-1 targets because they produced more seeds than other We also screened 46 T2 plants from #1-32 line for TaGLW7
plants. The T1 plant #4-6 is heterozygous for the 1160-bp mutations using PCR-RNP assay (Figure S10) and found novel
deletion in TaCKX2-D1 and is wild type in TaCKX2-A1 and mutations in the TaGLW7-A and TaGLW7-D gene. Two indepen-
TaCKX-B1 (Figure S7). A total of 180 T2 plants from line #4-6 dent mutant plants carried the same mutation type of 1-bp
were screened using PCR-RE (Figure S9). Fifty-one new mutations insertion in TaGLW7-A, and a novel 5-bp deletion was found in
were identified including 27 in TaCKX2-A1, 18 in TaCKX2-B1, TaGLW7-D. The results suggest that the occurrence of novel
and 6 in TaCKX2-D1 homoeologous loci in addition to the 1160- mutations may be associated with the continuing activity of the
bp deletion in TaCKX2-D1 (Figure 6; Table S4). While most of the Cas9/gRNA complex.
newly identified mutations were heterozygous, two mutations, 1-
No off-target mutation detected via Sanger sequencing
bp deletion in #4-6-226 and 2-bp deletion in #4-6-47 in TaCKX2-
of PCR-amplicons in a CRISPR active population
A1 (Figure S9a; Table S6), are homozygous. Because no
heterozygotes were found in the T2 population, the homozygous We next assessed the potential off-target effect by the wheat
mutations were most likely derived from fertilization of the two CRISPR/Cas9 system. Considering that Cas9 and the guide RNA
independent deletions at the same positions from the egg and for TaCKX2-1 were most active to induce on-target mutations in

ª 2019 The Authors. Plant Biotechnology Journal published by Society for Experimental Biology and The Association of Applied Biologists and John Wiley & Sons Ltd., 17, 1623–1635
 Agrobacterium-delivered wheat CRISPR/Cas9 system 1629

TaCKX2-3.1-A # 4-1 TaGW2-A #1-14

5’-CCCAACGCCACCCTGGCGGCCTCCACGGACTTCGGCAACATCACGGCGG-3’ WT 3’-GGGCGCGAGCTTGGCCTCGAGGATCAACTTGCGTAGCTTCTTCTGG-5’ WT
5’-CCCAAC------------------------TGCGGCAACATCACGGCGG-3’ -24 3’-GGGCGCGAGCTTGGC-----------------GTAGCTTCTTCTGG-5’ -17

TaCKX2-3.1-A # 4-2 TaGW2-D #1-12

5’-CCCA.....................................GCATGCAC-3’ WT 3’-GGGCGCGAGCTTGGCCTCGAGGATCAACTTGCGTAGCTTCTTCTGG-5’ WT
5’-CCCA...............499bp.................GCATGCAC-3’ -499 3’-GGGCGCGAGCTTGGC-TCGAGGATCAACTTGCGTAGCTTCTTCTGG-5’ -1

TaCKX2-3.1-B # 4-31 TaGLW7-A #1-32

5’-CCCAACGCCACCCTGGCGGCCTCCACGGACTTCGGCAACATCACGGCGG-3’ WT 3’-TATATAGATATACATAGCTAGTAGCATGGCTGCTTGACAGAAGAGAGGG-5’ WT
5’-CC-----------TGGCGGCCTCCACGGACTTCGGCAACATCACGGCGG-3’ -11 3’-TATATAGATATACATACC-------------------------AGAGGG-5’ -25

TaCKX2-3.1-B # 4-35 TaGLW7-A #1-38

5’-CCCAACGCCACCCTGGCGGCCTCCACGGACTTCGGCAACATCACGGCGG-3’ WT 3’-TATATAGATATACATAGCTAGTAGCATGGCTGCTTGACAGAAGAGAGGG-5’ WT
5’-CCCAAC-CCACCCTGGCGGCCTCCACGGACTTCGGCAACATCACGGCGG-3’ -1 3’-TATATAGATATACATAGCTAGTAGCATGGCTG----------GAGAGGG-5’ -10

TaCKX2-3.1-D # 4-6 TaGW8-D #1-73

5’-CAATAC...........ACGCATACGTTTATGT............ACCCTGG-3’ WT 5’-GACGCACGTGCTCGACTCCGACTGTGCTCTCTCTCTTCTGTCATCCCCGGCC-3’ WT
5’-CAATAC...880bp...ACGCATACGTTTATAT...280bp....ACCCTGG-3’ Mix 5’-GACGCACGTGCTC----------------TCTCTCTTCTGTCATCCCCGGCC-3’ -16

Figure 5 CRISPR/Cas9 induced mutations in target genes. Sequencing chromatograms of PCR amplicons from the selected T1 mutants and corresponding
regions are shown in (a–d) TaCKX2-1, TaGW2, TaGLW7, and TaGW8. The PAM sequences and target sequences are highlighted in green and red,
respectively. The dashes lines indicate deletion, and the blue letters represent substitution. Black dots indicate nucleotides not shown.

the #4-6 T2 family (i.e., 51 new mutations in 180 plants), we

Grain numbers increase in TaCKX2-1 mutants
selected progeny, 24 TaCKX2-1 mutants and 24 wild type plants,
of this family to assess the off-target effect by this most active While combining the mutations in the homoeologous copies
gRNA. The TaCKX2-1 target sequences plus the PAM motifs were of an orthologous gene or different orthologous loci, we
used to search the wheat reference genome, resulting in eight evaluated the effect of mutation of the 1160-bp deletion in
potential off-target sites with 2 and 3 bp mismatches compared TaCKX2-D1 on grain number per spikelet and spikelet density
to the target sequences. Two of the eight off-targets are unique per spike. Results showed that there was no significant
in the wheat genome, while the remaining six contain two sites, difference in spike length, spikelet number/spike, and spikelet
each of which are triplicated in the wheat genome. We selected density between the mutant and the wild type segregants
the top four of eight loci to design gene and allele specific primers from the same T2 population (P > 0.76805). Compared to the
for PCR amplification of relevant regions (Table S8). Individual wild type, the mutation increased grain number/spike from 42
PCR amplicons (n = 192) were subjected to Sanger sequencing, to 55 (P = 0.03374; Figure 7c,e), grain number/spikelet from
and the sequencing chromatograms were manually analysed for 2.45 to 3.25 (P = 0.00198; Figures 7b and 9d) and grain
any sign of double tracing peaks at the target regions. The results weight per spike from 1.12 to 1.69 g (P = 0.01717; Fig-
showed that no mutation occurred at these four off-target sites ure 7f). These results indicate that TaCKX2-1 negatively
via the targeted Sanger sequencing method. However, it is regulates the trait of grain number in wheat.
possible that unintended mutations occur in some somatic cells,
rare mutations that can be detected only by using more sensitive Discussion
methods such as deep sequencing of relevant PCR amplicons. It is
Agrobacterium delivery of CRISPR/Cas9 system
also possible that genetic variations from parental plants can alter
PAMs and increase the potential off-target effect in more CRISPR/Cas9 has been the primary choice for plant genome
complex and larger polyploid genomes like cotton (Li et al., editing, and Agrobacterium-mediated transformation is the most
2018). Nevertheless, the results indicate that, with careful design common method for delivery of the CRISPR/Cas9 DNA compo-
of guide RNA and better knowledge of the genome, the CRISPR/ nents to plants (Soyars et al., 2018). Mainly due to historical
Cas9 system can specifically target the pre-selected sites for reason, all genome editing systems in wheat were implemented
mutations. through the biolistics-mediated transformation (Liang et al.,

ª 2019 The Authors. Plant Biotechnology Journal published by Society for Experimental Biology and The Association of Applied Biologists and John Wiley & Sons Ltd., 17, 1623–1635
1630 Zhengzhi Zhang et al.

TaCKX2-3.1-A
GCGACCCCAACGCCACCCTGGCGGCCTCCACGGACTTCGGCAACATCACGGCGGCGCTCC WT/WT
GCGACCCCAAC-CCACCCTGGCGGCCTCCACGGACTTCGGCAACATCACGGCGGCGCTCC -1/WT
GCGACCCCAAC--CACCCTGGCGGCCTCCACGGACTTCGGCAACATCACGGCGGCGCTCC -2/WT
GCGACCCCA--------------GCCTCCACGGACTTCGGCAACATCACGGCGGCGCTCC -14/WT
GCG--------------------GCCTCCACGGACTTCGGCAACATCACGGCGGCGCTCC -20/WT
GCGACC--------------------TCCACGGACTTCGGCAACATCACGGCGGCGCTCC -20/WT
GCGACCCCAAC--------------------GGACTTCGGCAACATCACGGCGGCGCTCC -20/WT
GCGACCCCA---------------------CGGACTTCGGCAACATCACGGCGGCGCTCC -21/WT
GCGACCCCAAC---------------------------------ATCACGGCGGCGCTCC -33
GCGACCCCAACGCCAC------------------------------------------CC -41
GGGCAA----------------------63bp-----------------------GCTCC -63
GTTTAC----------------------164bp----------------------GCTCC -164
GCGACCCCAACtGCCACCCTGGCGGCCTCCACGGACTTCGGCAACATCACGGCGGCGCTCC +1/WT
CCGCGT---122bp---CTTCGGCAACAT---507bp---GTA--52bp---CAACTCGC Mix

TaCKX2-3.1-B
GCGACCCCAACGCCACCCTGGCGGCCTCCACGGACTTCGGCAACATCACGGCGGCGCTCC WT/WT
GCGACCCC-----------GGCGGCCTCCACGGACTTCGGCAACATCACGGCGGCGCTCC -11/WT
GCGACCC-----------TGGCGGCCTCCACGGACTTCGGCAACATCACGGCGGCGCTCC -11/WT
GCGACCCCAACGCCACCCTGGCGGCCTCCACGGACTT-----------CGGCGGCGCTCC WT/-11
GCGACCCCAAC--------------------GGACTTCGGCAACATCACGGCGGCGCTCC -20/WT
GCGACCCCA---------------------CGGACTTCGGCAACATCACGGCGGCGCTCC -21/WT
GCGACCCCAA---------------------GGACTTCGGCAACATCACGGCGGCGCTCC -21/WT
GCGACCCCAAC---------------------------------ATCACGGCGGCGCTCC -33/WT
CGCGAA---104bp---GCTTGCCG---267bp---GCATCA--97bp---GTGGACGGA Mix

TaCKX2-3.1-D
GCGACCCAAACGCCACCCTGGCGGCCTCCACGGACTTCGGCAACATCACGGCGGCGCTCC WT/WT
GCGACCC-----------TGGCGGCCTCCACGGACTTCGGCAACATCACGGCGGCGCTCC -11/WT
-----17bp-----ACCCTGGCGGCCTCCACGGACTTCGGCAACATCACGGCGGCGCTCC -17/WT
G---------------------GGCCTCCACGGACTTCGGCAACATCACGGCGGCGCTCC -21/WT
GCGACCCAAAC------------------------------------------GCGCTCC -42

Figure 6 Sequence information of new TaCKX2-A1, TaCKX2-B1 and TaCKX2-D1 mutations in the T2 progeny plants from event #4-6. The mutation types
include deletions (dashed line), insertions (lower-case letters in black) and substitutions (blue letters). PAM sequences are in green.

2017; Sanchez-Leo n et al., 2018; Wang et al., 2014, 2018; and expanding the spectrum of transformable genotypes (Wang
Zhang et al., 2018a,b). Co-transfer of large DNA fragments et al., 2017a) makes it a more promising delivery method for
from the vector backbone by biolistic transformation leads to the wheat genome editing. In the present research, we developed
high frequency of transgene silencing (Anand et al., 2003; Tassy an Agrobacterium-delivered CRISPR/Cas9 system and success-
et al., 2014). A large number of transgenic plants are required fully generated targeted mutations in four grain-regulatory
for recovery of desired edit mutations in T0 generation. Thus, genes. The transgenes remain active and continuously induce
transformation poses a bottleneck for many wheat genome site-specific mutations in T1 and T2 generations. Compared to a
editing projects and application in wheat breeding. By contrast, limited number of T0 transgenic plants, it is much easier to
Agrobacterium mediation usually generates simple insertion generate large pools of T1 and T2 plants and screen for
events of the transfer DNA (T-DNA), a DNA fragment defined independent and novel mutations using the efficient and cheap
by the border (left and right) sequences in a binary plasmid, with Cas9/gRNA RNP assay. Thus, the Agrobacterium-mediated
reduced frequency of transgene silencing (Dai et al., 2001). system provides an alternative option for wheat genome editing,
Recent progress in improving Agrobacterium-mediated transfor- which requires a small number of transformation events but
mation efficiency (Ishida et al., 2015; Richardson et al., 2014) additional generations.

ª 2019 The Authors. Plant Biotechnology Journal published by Society for Experimental Biology and The Association of Applied Biologists and John Wiley & Sons Ltd., 17, 1623–1635
 Agrobacterium-delivered wheat CRISPR/Cas9 system 1631

(a) (d)

(e)

Figure 7 Effect the 1160-bp deletion in TaCKX2-

D1 on grain number. (a) Spikes of wild type (left)
and the homozygous mutant (right) segregated
from the T2 family #4-6. (b) Spikelets of the wild
type (left) and the homozygous mutant (right). (c) (b)
Seeds harvested from the main-stem spikes of the
wild type (left) and the homozygous mutant
(right). (d) Grain number per spikelet of the wild (f)
type (left) and the homozygous mutant (right).
The numbers in the y-axis indicate grain number
per spikelet. (e) Grain number per spike of the
wild type (left) and the homozygous mutant (c)
(right). The numbers in the y-axis indicate grain
number per spike. (f) Grain weight per spike of the
wild type (left) and the homozygous mutant
(right). The numbers in the y-axis indicate grain
weight (gram) per spike. The error bars indicate
the standard deviation of the mean calculated
from five biological replicates.

rate of 28.3%. The estimated edit rate is much lower in TaCKX2-

Types and spectrum of edit mutations in wheat
D1 as compared to TaCKX-A1 and TaCKX-B1 because the T0
Deep sequencing of the target region from the T0 plants and plant #4 carried a 1160-bp deletion allele and the T1 plant #4-6 is
screening a large population of T1 and T2 plants give us the heterozygous for this deletion. Therefore, the potential edit rate
opportunity to investigate the spectrum and types of the edit in TaCKX2-1 could be even higher. This is approximately
mutations in wheat. Our results show that deletions are the threefold higher than the previously reports in wheat by Biolistic
dominant type of edit mutations in wheat. In T0 plants, deep delivery (Liang et al., 2017; S
anchez-Leo n et al., 2018; Wang
sequencing indicates that over 99% of the mutations are due to et al., 2014, 2018). Compared to TaCKX2-1 and TaGW2, the
deletions, and our genotyping data from T1 and T2 plants show mutation rate in TaGLW7 and TaGW8 is much lower. This is
that only five insertions were found of the 68 mutations because we intended to target the microRNA recognition sites of
detected, all of which are insertions of 1-bp A or T. Remaining these two genes, which are highly conserved among the
63 mutations are deletions (Figure S11). Similar scenarios were squarosa promoter binding-like transcription factors (Xie et al.,
also found in other wheat studies (Liang et al., 2017; S anchez- 2006). Thus, there is no other option for designing gene-specific
Leon et al., 2018; Wang et al., 2014, 2018; Zhang et al., 2018a, guide RNA genes. Despite the low edit efficiency compared to
b). Thus, deletions are the dominant type of edit mutations in the model plants, we detected two homozygous mutants in
wheat. This scenario greatly contrasts with what was found in TaCKX2-A1 and two triple mutants in a population of 231 T2
the model grass rice, where insertion is the primary type of edit plants (Table S4 and S6).
mutations (Zhang et al., 2014). In the 68 mutants identified,
Use of edit mutations in wheat breeding
deletions over 10 bp accounts for 73.5% and complex edit
mutations account for 25% (Table S4; Figure S11). This contrasts In addition to studies of gene function, CRISPR technology can
the deep sequencing results from the T0 plants, where the 1-, 2-, have a great impact on plant breeding by manipulating the
and 37-bp deletions are most abundant. This discrepancy might agriculturally important traits such as grain quality and grain yield.
be associated with cell types wherein mutations occurred. The Our result showed that 1160-bp deletion in TaCKX2-D1 signif-
DNA used for deep sequencing was from leaf somatic cells, and icantly increased the grain number per spikelet. This is consistent
the mutations recovered in the T1 and T2 plants occurred in germ with a prior study in rice, where reduction-of-function or loss-of-
cells or embryonic cells. Out of 180 T2 plants from the #4-6 line, function mutations in OsCKX2 (Gn1) increased grain number per
we detected 51 new mutations by PCR-RE with an estimated edit panicle (Ashikari et al., 2005). The effect of TaCKX2-1 on grain

ª 2019 The Authors. Plant Biotechnology Journal published by Society for Experimental Biology and The Association of Applied Biologists and John Wiley & Sons Ltd., 17, 1623–1635
1632 Zhengzhi Zhang et al.

number needs to be further validated by using other independent Two versions of Cas9 were used. The first Cas9 was codon-
homozygous mutants of single double or triple mutations of the optimized for expression in wheat and made from six gBlocks
same orthologous gene. In addition to TaCKX2-1, we also through the Gibson assembly using Gibson Assembly Master
obtained mutation in three other grain regulatory genes, Mix (New England Biolabs, Ipswich, MA). The assembled Cas9
TaGLW7, TaGW2, and TaGW8. Two recent studies on TaGW2 was cloned into pMCG1005 in which the selection marker Bar
showed that knockout mutations of this gene could significantly gene was replaced with hygromycin phosphotransferase (hpt)
increase wheat grain size and grain weight (Wang et al., 2018; gene. After confirmation of the hpt activity in rice callus through
Zhang et al., 2018a,b). Combining and transferring the muta- Agrobacterium-mediated transformation, the construct was
tions of TaCKX2-1 and TaGW2 into the elite wheat cultivars are inserted with the AttR1-ccdB-AttR2 Gateway cassette to produce
expected to have an additive effect on increasing the sink plasmid pTaCas9-Hyg-GW. The selection efficiency of the ccdB
strength. gene was examined by the introduction of the plasmid into
Wheat cultivars Bobwhite and Fielder are the most trans- Escherichia coli XL1-blue (Agilent Genomics, Santa Clara, CA) for
formable genotypes, but they are outdated and not currently minimal colony formation. pTaCas9-Hyg-GW was used for testing
used in wheat production and by wheat breeding. For the use of the activity of U6 promoters in protoplasts. The second Cas9
the edit mutations in wheat improvement, two approaches may gene, Cas9:P2A:H2A:GFP and referred to as Cas9-GFP here-
be considered. A straightforward approach would be transferring inafter, was also wheat-codon optimized and derived from the
the desired mutation alleles into elite genotypes and purging the fusion of Cas9 with the histone 2A (H2A) and GFP coding
CRISPR/Cas9 transgene by backcrossing. Another approach sequences, the Cas9 coding sequence was separated by a
would be transferring the CRISPR/Cas9 transgene into the elite sequence encoding P2A, a translational slipping signal. This
cultivars and screen new mutations in the improved background. Cas9-GFP under the maize ubiquitin 1 gene promoter was used
Compared to the former, the latter approach would reduce the for wheat transgenics.
risk of genetic vulnerability due to the reduction in diversity in the To transiently and quickly identify active U6 promoters for
region surrounding the mutant alleles. One prerequisite for this gRNA, a mutated green fluorescent protein (GFP) gene was
approach is the inheritance of active CRISPR/Cas9 transgene constructed. To insert a 1-bp downstream of the start codon of
through the backcrossing generations. In this respect, the GFP, a forward primer GFP+1-F, with the frameshift insertion,
Agrobacterium-mediated delivery system has an advantage over was paired with the reverse primer GFP-R (Table S9) to amplify
the biolistic delivery as demonstrated by the present research that the open reading frame (ORF) of the GFP gene. The PCR product
the CRISPR/Cas9 transgene remains high activity in the T1 and T2 was digested with BamHI and SpeI and cloned into pUC19 under
generation. Both approaches can be coupled with genomic the 35S promoter, resulting in pUC-35S:GFP+1. A sgRNA
selection for acceleration of the transgene-free mutants as novel targeting the 1-bp insertion site in pUC-35S:GFP+1 was synthe-
germplasm or cultivars. sized and cloned into the gRNA vector to produce the construct
pTagRNA4-gGFP.
Experimental procedures To make the constructs targeting the grain regulatory genes in
wheat, two complementary oligonucleotides corresponding to
Plants and growing conditions
the protospacer sequences of each of two sgRNAs, sgRNA1 and
Wheat cultivar Fielder and its transgenic plants were grown in 6″ sgRNA2 for the same gene, were designed and synthesized.
pots containing Sunshine potting mix (Green State Gardener, Double-stranded oligonucleotides were sequentially cloned into
Burlington, VT) mixed with Perlite and Vermiculite (Hummert, St BtgZI sites and BsaI sites in pTagRNA4, in which sgRNA1 is driven
Louis, MO) in equal ratio and supplied with Multicote controlled by TaU6.1 and sgRNA2 by TaU6.3. gRNA cassette was then PCR-
release fertilizer (Haifa Group, Haifa, Israel), one plant per pot. amplified and integrated into pCas9-GFP (Cas9-GFP in pENTR4) at
Plants were grown in greenhouse in the fall, winter and spring the BssHII site by Gibson assembly. Through Gateway recombi-
seasons and in growth chambers in summer to increase tillers. nation reaction, the Cas9-TagRNA4 cassette was finally incorpo-
The temperature of the greenhouse and growth chambers was rated into pLC41-GW (a Gateway binary vector) that is derived
set at 22 °C with a 16 h of photoperiod, and the plants were from pLC41_Exp1 Icat-GUS.
watered every other day. To make an in vitro transcription system for guide RNA that
can be used along with Cas9 protein to in vitro screen for
Construction of plasmids
mutated plants, a gBlock containing the T7 promoter, guide
Wheat U6 snRNA (Marshallsay et al., 1992) was used as a query to RNA scaffold, and two BsaI cloning sites in between was
BLAST search against wheat reference genome IWGSC (v1.0) synthesized (Integrated DNA Technology, Iowa City, IA). The
(International Wheat Genome Sequencing Consortium, 2018) to gBlock was inserted into pENTR4 between NcoI and XbaI
retrieve their promoter sequences. gBclocks for promoters of through Gibson assembly replacing the ccdB cassette, resulting
TaU6.1 and TaU6.2 were synthesized (Integrated DNA Technology, in pT7-sgRNA. To make in vitro transcribed guide RNA
Iowa City, IA) while TaU6.3 and TaU6.5 were PCR amplified from molecule, a guide RNA gene was assembled in pT7-sgRNA
wheat genomic DNA. U6 promoters were then cloned into pENTR4 vector. Specifically, two complementary oligonucleotides for the
together with guide RNA (gRNA) scaffold and U6 terminator, which spacer sequence of sgRNAs were synthesized, annealed to form
is flanked by attL1 and attL2 sites. Two BtgZI restriction sites or two double stranded DNA fragment, and then cloned into BsaI sites
BsaI restriction sites were introduced during the gBlock synthesis for in pT7-sgRNA.
cloning of the sgRNA spacer sequences. After testing the efficiency
Isolation and transfection of wheat protoplasts
of U6 promoters (Appendix S1) in wheat protoplasts, two most
active U6 promoters, i.e., U6.1 and U6.3 promoters, were Leaves from 1-week-old seedlings of wheat cultivar Fielder
assembled into the same construct to facilitate the construction were chopped latitudinally into 0.5-mm strips using razor
of two sgRNA genes in an intermediate vector, pTagRNA4. blades. The leaf strips were used for protoplast isolation

ª 2019 The Authors. Plant Biotechnology Journal published by Society for Experimental Biology and The Association of Applied Biologists and John Wiley & Sons Ltd., 17, 1623–1635
 Agrobacterium-delivered wheat CRISPR/Cas9 system 1633

following the procedure described by Zhou et al. (2014). The

Wheat transformation
protoplasts were resuspended in a W5 solution (2 mM MES (pH
5.7), 154 mM NaCl, 125 mM CaCl2, 5 mM KCl) at a concen- Fielder was cultivated in a growth chamber. Immature seeds were
tration of 2 9 106 per mL. The pTaCas9-Hyg-GW, pUC-35S: harvested from panicles approximately 2 weeks after anthesis,
GFP+1, and pTaU6-gGFP were mixed at a concentration of surface-sterilized for 1 min in 70% ethanol followed by 10 min in
400 ng/lL each plasmid. An aliquot of 200 lL protoplast 1.2% sodium hypochlorite and were washed three times with
solution was sequentially mixed with 30 lL of plasmid DNA sterilized water. Immature embryos were isolated from the
and 230 lL of freshly prepared PEG solution (40% PEG-4000, sterilized seeds under a stereoscopic microscope. The CRISPR/
0.2 M Mannitol, 0.1 M CaCl2) in a 2-ml microcentrifuge tube by Cas9 constructs were transferred into Agrobacterium tumefa-
inverting three times. After the mixture was incubated at room ciens strain EHA105 by electroporation. The Agrobacterium-
temperature in the dark for 20 min, 900 lL of W5 solution mediated transformation on the isolated immature embryos and
was added to the tube and mixed by inverting the tube three plant regeneration were conducted following the protocol
times. The transfected protoplasts were collected by centrifu- described by Ishida et al. (2015). The transgenic plants were
gation at 250 g for 5 min, washed once with W5 solution, re- further confirmed by PCR assays of the Cas9 and sgRNA
suspended in 800 lL of W5 solution, and incubated at room transgenes.
temperature. Constructs pOsUbi-GFP and pUC-35S:GFP+1
Purification of Cas9 protein, in vitro transcription of
served as a positive and negative control, respectively, to
guide RNA and preparation of Cas9/sgRNA
transfect the protoplasts. Forty hours after transfection, proto-
ribonucleoprotein
plasts were examined under a fluorescence microscope for the
proportion of protoplasts with GFP fluorescence and images Escherichia coli strain Rosetta 2 DE3 carrying plasmid pMJ806 was
were captured. Transfection was conducted with two replicates obtained from Addgene. The Cas9 protein was prepared follow-
for each construct, and five random microscope views under ing the protocol as described (Anders and Jinek, 2014). For
209 magnification were calculated. Images were analysed with in vitro transcription of guide RNA, pT7-sgRNA derived gRNA-
Fiji-ImageJ (Schindelin et al., 2012), and integrated density was specific construct was linearized at a unique restriction enzyme
measured to quantify the fluorescence density in the selected site NcoI, followed by dephosphorylation with Shrimp Alkaline
view. Integrated density was then averaged by the total Phosphatase (rSAP) (New England Biolabs, Ipswich, MA), and
number of protoplasts in that view. then purified with DNA Clean and Concentrator Kit (Zymo
Research, Irvine, CA). sgRNA was synthesized by using the
Primers and guide RNA design
HiScribe T7 Quick High Yield RNA Synthesis Kit (New England
Total genomic DNA was isolated from the wheat leaf tissue Biolabs) following the manufacturer’s instructions and purified
following the protocol described by Li et al. (2008) except using with RNA Clean and Concentrator (Zymo Research, Irvine, CA).
the CTAB buffer (2% CTAB (hexadecyl trimethyl-ammonium For each DNA/ribonucleoprotein reaction, 1 lg of sgRNA and
bromide), 100 mM Tris, 20 mM EDTA, 1.4 M NaCl, 1% PVP 1 lg of Cas9 protein were mixed in Cas9 reaction buffer and
(polyvinyl pyrrolidone (vinylpyrrolidine homopolymer) Mw incubated at room temperature for 15 min to form a ribonucle-
40 000), and pH5.0). Sequences of the homoeologous genes oprotein (RNP) complex. The resulting RNP was used for 2–3
of T. aestivum cv Chinese Spring (CS) were retrieved from the reactions each with about 1 lg of PCR product in a total volume
Ensembl Plants database (https://plants.ensembl.org/index.html), of 10 lL at 37 °C for 3 h. After inactivated at 65 °C for 15 min,
and gene-specific primers (Table S9) were designed for the the reactions were analysed with electrophoresis in 2% agarose
homoeologous genes using online program GSP (https://probes. gel.
pw.usda.gov/GSP/; Wang et al., 2016) and used to PCR-amplify
Mutant screening
the relevant genomic regions (~1 kb). Specificity of the primers
was validated by use of DNA from the nulli-tetrasomic stocks The PCR amplicons were digested with restriction enzymes (XcmI
(Sears, 1966), and the PCR products were separated through and BglI), T7E1 or Cas9/sgRNA ribonucleoprotein and separated
electrophoresis in 1% agarose gel. All the primers are listed in by agarose electrophoresis. For digestion with restriction enzymes
Table S9. The PCR reaction was performed for each homoeol- or T7E1, 1 unit of enzyme was used in 15 lL reaction including
ogous copy of an orthologous gene in the A, B, and D genome, PCR product, following the manufacturer’s instruction. After
which included 100 ng genomic DNA, 59 GoTaq Green PCR being denatured (95 °C for 5 min) and re-annealed (ramp down
buffer (Promega, Madison, WI), 125 lM dNTPs, 0.5 lM primers, to 25 °C at 5 °C/min), the PCR products were subjected to T7E1
and 0.2 lL DNA Taq polymerase. PCR cycling conditions were as (New England Biolabs, Ipswich, MA) digestion (1 unit, 0.2 lL) at
follows: an initial denaturation at 95 °C for 5 min, followed by 37°C for 1.5 h and separated by agarose gel electrophoresis.
35 cycles of 95 °C for 30 s, 55 °C for 30 s, and 72 °C for
DNA sequencing and sequence analyses
1.5 min, and a final extension at 72 °C for 7 min. The gene-
specific PCR products were cleaned up with ExoSAP-IT (Thermo The PCR products were deep-sequenced by the high-throughput
Fisher Scientific, Waltham, MA) treatment and subjected to sequencing platform HiSeq 2000 (Illumina, San Diego, CA) or
sequencing. The sequences from the A, B, and D genome were individually sequenced by Sanger sequencing in the ABI DNA
aligned using the online program Multiple Sequence Alignment analysers (Thermo Fisher Scientific, Waltham, MA). Genomic DNA
(MUSCLE, https://www.ebi.ac.uk/Tools/msa/muscle/) and con- samples from T0 plants transformed with Cas9 and TaCKX2-1
served sequences in the first exon of the genes were further sgRNA was used in amplicon sequencing. First round PCR was
inspected for selection of 20 nucleotides immediately upstream conducted using gene-specific primers with universal tags for 35
of a PAM motif (50 -NGG-30 ) and a GC content of 40%–60% for cycles. PCR products were separated in 2.0% agarose gel, and
guide RNA design. All guide RNA target sequences are listed in the target bands were excised and column-purified. Approxi-
Table S7. mately 10 ng of PCR product was used for a second round of PCR

ª 2019 The Authors. Plant Biotechnology Journal published by Society for Experimental Biology and The Association of Applied Biologists and John Wiley & Sons Ltd., 17, 1623–1635
1634 Zhengzhi Zhang et al.

with primers matching the universal tags and containing a Christian, M., Qi, Y., Zhang, Y. and Voytas, D.F. (2013) Targeted mutagenesis
sample-specific barcode within the forward primer. PCR was of Arabidopsis thaliana using engineered TAL effector nucleases. G3
conducted for 12 cycles. PCR products were separated on 2.0% (Bethesda), 3, 1697–16705.
Clasen, B.M., Stoddard, T.J., Luo, S., Demorest, Z.L., Li, J., Cedrone, F., Tibebu,
agarose gel, and target bands were cut and column-purified.
R. et al. (2015) Improving cold storage and processing traits in potato
Purified amplicons were quantified with Qubit 2.0 fluorometer
through targeted gene knockout. Plant Biotechnol. J. 14, 169–176.
(Life Technologies, Carlsbad, CA) and pooled equally to make a Cong, L., Ran, F.A., Cox, D., Lin, S., Barretto, R., Habib, N., Hsu, P.D. et al.
library at a concentration of 4 nmol/L. Paired-end (2 9 250) (2013) Multiplex genome engineering using CRISPR/Cas systems. Science,
sequencing was conducted with a total of 500 cycles. Sequence 339, 819–823.
analysis was conducted with CRISPR-DAV (Wang et al., 2017b) to Dai, S., Zheng, P., Marmey, P., Zhang, S., Tian, W., Chen, S., Beachy, R.N. et al.
discover the modification. (2001) Comparative analysis of transgenic rice plants obtained by
For Sanger sequencing, PCR amplicons were treated with Agrobacterium-mediated transformation and particle bombardment. Mol.
ExoSAP-ITTM PCR Product Cleanup Reagent (Thermo Fisher) Breed. 7, 25–33.
following the manufacturer’s manual before subjected to Gaudelli, N.M., Komor, A.C., Rees, H.A., Packer, M.S., Badran, A.H., Bryson,
D.I. and Liu, D.R. (2017) Programmable base editing of A•T to G•C in
sequencing. The sequencing chromatograms were fed to the
genomic DNA without DNA cleavage. Nature, 551, 464–471.
computer program DSDecodeM (http://skl.scau.edu.cn/dsdecode/;
Haun, W., Coffman, A., Clasen, B.M., Demorest, Z.L., Lowy, A., Ray, E.,
Liu et al., 2015) to determine the allelic mutations. Retterath, A. et al. (2014) Improved soybean oil quality by targeted
Phenotyping mutagenesis of the fatty acid desaturase 2 gene family. Plant Biotechnol. J.
12, 934–940.
The spikes from the main stems were scored for spikelet and International Wheat Genome Sequencing Consortium (2018) A chromosome-
grain numbers; spikelet length, density and average grain number based draft sequence of the hexaploid bread wheat (Triticum aestivum)
per spikelet were measured. The total grains from individual spike genome. Science, 345, 1251788.
were weighed using a PB602-S digital precision scale (Mettler Ishida, Y., Tsunashida, M., Hiei, Y. and Komari, T. (2015) Wheat (Triticum
Toledo, Columbus, OH). Five biological replicates were used for aestivum L.) transformation using immature embryos. Methods Mol. Biol.
1223, 189–198.
each genotype.
Jinek, M., Chylinski, K., Fonfara, I., Hauer, M., Doudna, J.A. and Charpentier, E.
Data analysis (2012) A programmable dual-RNA-guided DNA endonuclease in adaptive
bacterial immunity. Science, 337, 816–821.
The Student’s t-test was performed using pooled standard Jinek, M., Jiang, F., Taylor, D.W., Sternberg, S.H., Kaya, E., Ma, E., Anders, C.
deviations to evaluate the statistical significance of the difference et al. (2014) Structures of Cas9 endonucleases reveal RNA-mediated
between different U6 promoters or between the mutant and wild conformational activation. Science, 343, 1247997.
type. Chisq test was implemented for fitness of segregation of Komor, A.C., Kim, Y.B., Packer, M.S., Zuris, J.A. and Liu, D.R. (2016)
mutations in a T2 population to the 1 : 2 : 1 ratio. Both the Programmable editing of a target base in genomic DNA without double-
Student’s and Chisq tests were conducted using Microsoft Excel. stranded DNA cleavage. Nature, 533, 420–424.
The cut-off for statistical significance was set to a P-value of 0.05. Li, W. and Yang, B. (2017) Translational genomics of grain size regulation in
wheat. Theor. Appl. Genet. 130, 1765–1771.
Li, W., Huang, L. and Gill, B.S. (2008) Recurrent deletions of puroindoline genes
Acknowledgements at the grain hardness locus in four independent lineages of polyploid wheat.
Plant Physiol. 146, 200–212.
This research is supported by USDA NIFA-IWYP (award number: Li, T., Huang, S., Jiang, W.Z., Wright, D., Spalding, M.H., Weeks, D.P. and
2017-67008-25934 to WL and BY). We thank Xianran Li at Iowa Yang, B. (2011) TAL nucleases (TALNs): hybrid proteins composed of TAL
State University for helping analyse the deep sequencing data. effectors and FokI DNA-cleavage domain. Nucleic Acids Res. 39, 359–
372.
Li, T., Liu, B., Spalding, M.H., Weeks, D.P. and Yang, B. (2012) High-efficiency
Conflict of interest TALEN-based gene editing produces disease-resistant rice. Nat. Biotechnol.
The authors declare no conflicts of interest. 30, 390–392.
Li, J., Manghwar, H., Sun, L., Wang, P., Wang, G., Sheng, H., Zhang, J. et al.
(2018) Whole genome sequencing reveals rare off-target mutations and
References considerable inherent genetic or/and somaclonal variations in CRISPR/Cas9-
edited cotton plants. Plant Biotechnol. J. https://doi.org/10.1111/pbi.13020.
Anand, A., Zhou, T., Trick, H.N., Gill, B.S., Bockus, W.W. and Muthukrishnan, S. Liang, Z., Chen, K., Li, T., Zhang, Y., Wang, Y., Zhao, Q., Liu, J. et al. (2017)
(2003) Greenhouse and field testing of transgenic wheat plants stably Efficient DNA-free genome editing of bread wheat using CRISPR/Cas9
expressing genes for thaumatin-like protein, chitinase and glucanase against ribonucleoprotein complexes. Nat. Commun. 8, 14261.
Fusarium graminearum. J. Exp. Bot. 54, 1101–1111. Liu, W., Xie, X., Ma, X., Li, J., Chen, J. and Liu, Y.G. (2015) DSDecode: a web-
Anders, C. and Jinek, M. (2014) In vitro enzymology of Cas9. Methods Enzymol. based tool for decoding of sequencing chromatograms for genotyping of
546, 1–20. targeted mutations. Mol. Plant, 8, 1431–1433.
Ashikari, M., Sakakibara, H., Lin, S., Yamamoto, T., Takashi, T., Nishimura, A., Maeder, M.L., Thibodeau-Beganny, S., Osiak, A., Wright, D.A., Anthony, R.M.,
Angeles, E.R. et al. (2005) Cytokinin oxidase regulates rice grain production. Eichtinger, M., Jiang, T. et al. (2008) Rapid “open-source” engineering of
Science, 309, 741–745. customized zinc-finger nucleases for highly efficient gene modification. Mol.
Cai, C.Q., Doyon, Y., Ainley, W.M., Miller, J.C., Dekelver, R.C., Moehle, E.A., Cell, 31, 294–301.
Rock, J.M. et al. (2009) Targeted transgene integration in plant cells using Mahfouz, M.M., Li, L., Shamimuzzaman, M., Wibowo, A., Fang, X. and Zhu, J.-
designed zinc finger nucleases. Plant Mol. Biol. 69, 699–709. K. (2011) De novo-engineered transcription activator-like effector (TALE)
Carroll, D. (2011) Genome engineering with zinc-finger nucleases. Genetics, hybrid nuclease with novel DNA binding specificity creates double-strand
188, 773–782. breaks. Proc. Natl Acad. Sci. USA, 108, 2623–2628.
Cermak, T., Doyle, E.L., Christian, M., Wang, L., Zhang, Y., Schmidt, C., Baller, Mali, P., Yang, L., Esvelt, K.M., Aach, J., Guell, M., DiCarlo, J.E., Norville, J.E.
J.A. et al. (2011) Efficient design and assembly of custom TALEN and other et al. (2013) RNA-guided human genome engineering via Cas9. Science,
TAL effector-based constructs for DNA targeting. Nucleic Acids Res. 39, e28. 339, 823–826.

ª 2019 The Authors. Plant Biotechnology Journal published by Society for Experimental Biology and The Association of Applied Biologists and John Wiley & Sons Ltd., 17, 1623–1635
 Agrobacterium-delivered wheat CRISPR/Cas9 system 1635

Marshallsay, C., Connelly, S. and Filipowicz, W. (1992) Characterization of the Zhang, H., Gou, F., Zhang, J., Liu, W., Li, Q., Mao, Y., Botella, J.R. et al. (2015)
U3 and U6 snRNA genes from wheat: U3 snRNA genes in monocot plants are TALEN-mediated targeted mutagenesis produces a large variety of heritable
transcribed by RNA polymerase III. Plant Mol. Biol. 19, 973–983. mutations in rice. Plant Biotechnol. J. 14, 186–194.
Nishida, K., Arazoe, T., Yachie, N., Banno, S., Kakimoto, M., Tabata, M., Zhang, S., Zhang, R., Song, G., Gao, J., Li, W., Han, X., Chen, M. et al. (2018a)
Mochizuki, M. et al. (2016) Targeted nucleotide editing using hybrid Targeted mutagenesis using the Agrobacterium tumefaciens-mediated
prokaryotic and vertebrate adaptive immune systems. Science, 353, aaf8729. CRISPR-Cas9 system in common wheat. BMC Plant Biol. 18, 302.
Richardson, T., Thistleton, J., Higgins, T.J., Howitt, C. and Ayliffe, M. (2014) Zhang, Y., Li, D., Zhang, D., Zhao, X., Cao, X., Dong, L., Liu, J. et al. (2018b)
Efficient Agrobacterium transformation of elite wheat germplasm without Analysis of the functions of TaGW2 homoeologs in wheat grain weight and
selection. Plant Cell Tiss. Organ. Cult. 119, 647–659. protein content traits. Plant J. 94, 857–866.
Sanchez-Leo n, S., Gil-Humanes, J., Ozuna, C.V., Gimenez, M.J., Sousa, C., Zhou, H., Liu, B., Weeks, D.P., Spalding, M.H. and Yang, B. (2014) Large
Voytas, D.F. and Barro, F. (2018) Low-gluten, nontransgenic wheat chromosomal deletions and heritable small genetic changes induced by
engineered with CRISPR/Cas9. Plant Biotechnol. J. 16, 902–910. CRISPR/Cas9 in rice. Nucleic Acids Res. 42, 10903–10914.
Schindelin, J., Arganda-Carreras, I., Frise, E., Kaynig, V., Longair, M., Pietzsch,
T., Preibisch, S. et al. (2012) Fiji: an open-source platform for biological-
image analysis. Nat. Methods, 9, 676–682. Supporting information
Sears, E.R. (1966) Nullisomic-tetrasomic combinations in hexaplold wheat. In
Additional supporting information may be found online in the
Chromosome Manipulation and Plant Genetics (Lewis, D.R., ed.), pp. 29–47.
Supporting Information section at the end of the article.
London: Oliver and Boyd.
Shan, Q., Wang, Y., Chen, K., Liang, Z., Li, J., Zhang, Y., Zhang, K. et al. Figure S1 Comparison in editing efficiency of gGFP driven by
(2013a) Rapid and efficient gene modification in rice and Brachypodium different wheat U6 promoters in terms of the proportion of
using TALENs. Mol. Plant, 6, 365–368. fluorescence cell (a) and average fluorescence intensity per cell (b).
Shan, Q., Wang, Y., Li, J., Zhang, Y., Chen, K., Liang, Z., Zhang, K. et al.
Figure S2 Structure of pTagRNA4 in the pENTR4 backbone. Two
(2013b) Targeted genome modification of crop plants using a CRISPR-Cas
sgRNAs are inserted into BtgZI-BtgZI and BsaI-BsaI sites sequentially.
system. Nat. Biotechnol. 31, 686–688.
Soyars, C.L., Peterson, B.A., Burr, C.A. and Nimchuk, Z.L. (2018) Cutting edge
Figure S3 Structure of pCas9-GFP used in the transgene. pTaU6-
genetics: CRISPR/Cas9 editing of plant genomes. Plant Cell Physiol. 59, 1608–1620. gRNA cassette is inserted into BssHII sites through Gibson assembly.
Symington, L.S. and Gautier, J. (2011) Double-strand break end resection and Figure S4 Deletion distribution along the amplicon in TaCKX2-A1
repair pathway choice. Annu. Rev. Genet. 45, 247–271. (a), TaCKX2-B1 (b), and TaCKX2-D1 (c).
Tassy, C., Partier, A., Beckert, M., Feuillet, C. and Barret, P. (2014) Biolistic Figure S5 Identification of Cas9-positive transgenic plants in T1
transformation of wheat: increased production of plants with simple progenies.
insertions and heritable transgene expression. Plant Cell, Tissue Organ Cult. Figure S6 Targeted TaCKX2-1 homoeologues using the CRISPR/
119, 171–181. Cas9 system in transgenic T1 plants.
Wang, Y., Cheng, X., Shan, Q., Zhang, Y., Liu, J., Gao, C. and Qiu, J. (2014)
Figure S7 Screen mutant plants in TaCKX2-D1 # 4 T1 plants using
Simultaneous editing of three homoeoalleles in hexaploid bread wheat confers
PCR assay.
heritable resistance to powdery mildew. Nat. Biotechnol. 32, 947–951.
Wang, Y., Tiwari, V.K., Rawat, N., Gill, B.S., Huo, N., You, F.M., Coleman-Derr,
Figure S8 Screen mutations in TaCKX2-1 of plants from the # 4-1
D., et al. (2016) GSP: a web-based platform for designing genome-specific and # 4-35 T2 family by the PCR-RE assay.
primers in polyploids. Bioinformatics 32, 2382–2383. Figure S9 Screen mutations in TaCKX2-1 of the T2 plants from the #
Wang, K., Liu, H., Du, L. and Ye, X. (2017a) Generation of marker-free 4-6 family by the PCR-RE assay.
transgenic hexaploid wheat via an Agrobacterium-mediated co- Figure S10 Screening of mutations in TaGLW7 of the T2 plants from
transformation strategy in commercial Chinese wheat varieties. Plant the # 1-32 family by the PCR-RNP assay.
Biotechnol. J. 15, 614–623. Figure S11 Types and spectrum of 68 edit mutations.
Wang, X., Tilford, C., Neuhaus, I., Mintier, G., Guo, Q., Feder, J.N. and Kirov, S. Table S1 sgRNA target sites used in this study.
(2017b) CRISPR-DAV: CRISPR NGS data analysis and visualization pipeline.
Table S2 Verification of transgenes in T0 plants.
Bioinformatics, 33, 3811–3812.
Table S3 Frequencies of CRISPR/Cas9-induced novel mutations in
Wang, W., Pan, Q., He, F., Akhunova, A., Chao, S., Trick, H.A. and Akhunov, E.
(2018) Transgenerational CRISPR-Cas9 activity facilitates multiplex gene
T0, T1, and T2 populations.
editing in allopolyploid wheat. CRISPR J. 1, 65–74. Table S4 Types of CRISPR/Cas9-induced mutations for target
Weeks, D.P., Spalding, M.H. and Yang, B. (2016) Use of designer nucleases for genes in T0, T1, and T2 generations.
targeted gene and genome editing in plants. Plant Biotechnol. J. 14, 483–495. Table S5 Genotypes of mutant plants induced by CRISPR/Cas9
Wright, D.A., Townsend, J.A., Winfrey, R.J.J., Irwin, P.A., Rajagopal, J., Lonosky, system in T1 population.
P.M., Hall, B.D. et al. (2005) High-frequency homologous recombination in Table S6 Genotypes of mutant plants induced by CRISPR/Cas9
plants mediated by zinc-finger nucleases. Plant J. 44, 693–705. system in TaCKX2 T2 family #4-6.
Xie, K., Wu, C. and Xiong, L. (2006) Genomic organization, differential Table S7 A summary of the double and triple mutants recovered
expression, and interaction of SQUAMOSA promoter-binding-like
in TaCKX2-1.
transcription factors and microRNA156 in rice. Plant Physiol. 142, 280–293.
Table S8 Off-targeting of designed sgRNA for TaCKX2-1 gene.
Zhang, H., Zhang, J., Wei, P., Zhang, B., Gou, F., Feng, Z., Mao, Y. et al. (2014)
The CRISPR/Cas9 system produces specific and homozygous targeted gene
Table S9 PCR primers used in this study.
editing in rice in one generation. Plant Biotechnol. J. 12, 797–807. Appendix S1 Sequences of U6 promoters.

ª 2019 The Authors. Plant Biotechnology Journal published by Society for Experimental Biology and The Association of Applied Biologists and John Wiley & Sons Ltd., 17, 1623–1635

View publication stats