Gun Li,Rong Liu,Rongng Xu,Rjv K.Vrshny,Hnng Ding,Mngwi Li,Xin Yn,Shuxin Hung,Jun Li,Dong Wng,Yishn Ji,Chnyu Wng,Jungung H,Yingng Luo,Shnghn Go,Pnghng Wi,*,Xuxio Zong,*,To Yng,*

a National Key Facility for Crop Gene Resources and Genetic Improvement/Institute of Crop Sciences,Chinese Academy of Agricultural Sciences,Beijing 100081,China

b Key Laboratory of Rice Genetic Breeding of Anhui Province,Rice Research Institute,Anhui Academy of Agricultural Sciences,Hefei 230031,Anhui,China

c State Agricultural Biotechnology Centre,Centre for Crop and Food Innovation,Food Futures Institute,Murdoch University,Murdoch,WA 6150,Australia

d Institute of Crop Germplasm Resources,Shandong Academy of Agricultural Sciences/Shandong Provincial Key Laboratory of Crop Genetic Improvement,Ecology and Physiology,Jinan 250100,Shandong,China

e Zhejiang Xinnan Chemical Industrial Group Co.,Ltd.,Hangzhou 311300,Zhejiang,China

f State Key Laboratory of Microbial Resources,Institute of Microbiology,Chinese Academy of Sciences,Beijing 100101,China

Keywords:Agrobacterium-mediated transformation CRISPR/Cas9 Pea Genome editing

ABSTRACT Pea(Pisum sativum L.)is an annual cool-season legume crop.Owing to its role in sustainable agriculture as both a rotation and a cash crop,its global market is expanding and increased production is urgently needed.For both technical and regulatory reasons,neither conventional nor transgenic breeding techniques can keep pace with the demand for increased production.In answer to this challenge,CRISPR/Cas9 genome editing technology has been gaining traction in plant biology and crop breeding in recent years.However,there are currently no reports of the successful application of the CRISPR/Cas9 genome editing technology in pea.We developed a transient transformation system of hairy roots,mediated by Agrobacterium rhizogenes strain K599,to validate the efficiency of a CRISPR/Cas9 system.Further optimization resulted in an efficient vector,PsU6.3-tRNA-PsPDS3-en35S-PsCas9.We used this optimized CRISPR/Cas9 system to edit the pea phytoene desaturase(PsPDS)gene,causing albinism,by Agrobacterium-mediated genetic transformation.This is the first report of successful generation of gene-edited pea plants by this route.

1.Introduction

Pea(Pisum sativum L.,2n=14),belonging to the family Fabaceae(Leguminosae),originated in and around Central Asia,Ethiopia,the Near East,and the Mediterranean[1].Historically,peas played a pivotal role in the discovery of Mendel’s laws of inheritance,which paved the way for modern genetics[2].However,today the progress of molecular biology and genetics research on pea is far behind that of other legumes,such as Medicago truncatula and Lotus japonicus[3].Peas have a narrow genetic base owing to self-pollination[4],making it difficult to develop cultivars with excellent agronomic traits,especially traits with complex inheritance[5].

Peas are planted in more than 90 countries[6]and continue to be in demand with both farmers and consumers as both a rotation and a cash crop.In 2019,over 7 Mha of dry peas and close to 3 Mha of green peas were sown globally[6].However,the pea yield for 2019 was,on average,only about 2000 kg ha-1,much lower than that of common bean[6,7].Neither traditional nor modern breeding techniques have produced highly improved agronomic traits in pea.Conventional breeding methods,historically relied upon for crop improvement,are inefficient,time-consuming,and complicated in pea[8].Transgenic breeding methods continue to be dogged by safety and regulatory concerns,and commercialization of genetically modified(GM)crops has been limited[9].The complete pea reference genome was published in 2019,making it possible to develop genome-assisted breeding methods[10].

Compared with these other breeding approaches,precision genome editing is attractive for its speed,flexibility,and transgenefree nature[11].Gene editing is a novel genetic engineering technology that relies on engineered nucleases,also known as‘‘molecular scissors”,to insert or delete specific genes at any desired location[12].Clustered regularly interspaced short palindromic repeat(CRISPR)/CRISPR-associated nuclease 9(Cas9)is the third-generation gene editing tool,after zinc-finger nucleases(ZFN)and transcription activator-like effector nucleases(TALENs)[13–16].The CRISPR/Cas9 system comprises a single guide RNA(sgRNA)and Cas9 nuclease,which together form a complex.The sgRNA guides the Cas9 nuclease to anchor to a specific site,inducing targeted DNA double-stranded breaks(DSBs),which are then repaired via homologous recombination(HR)or non-homologous end joining(NHEJ)[17].Owing to its high accuracy and efficiency,low cost,and ease of use,the CRISPR/Cas9-mediated genome editing system has been widely adopted and used in many crops,including rice,maize,wheat,cotton,and soybean[11,18–20].

For the CRISPR/Cas9 system to work,reagents must be delivered in vivo.Harnessing the power of Agrobacterium spp.to insert genetic material into host plants,Agrobacterium-mediated genetic transformation(AMGT)is the currently preferred method of reagent delivery for gene editing[21].However,in comparison with other legumes,it has proved difficult to establish a system for AMGT in pea[22–26].Many researchers[27–30]have tried to develop an efficient genetic transformation system for pea using various tissue culture methods,Agrobacterium strains,infection methods,and exogenous hormones,but these attempts have been characterized by low transformation efficiency and reproducibility.The number of successful transgenic events reported in pea is low[31].There has been no report of successful genome editing by Agrobacterium-mediated CRISPR/Cas9 gene editing in pea.

To validate any new gene editing methodology,it is prudent to start with a gene that produces an obvious phenotype such as albinism[32].The mutation or knockout of the phytoene desaturase(PDS)gene impairs photosynthesis and carotenoid biosynthesis,leading to albinism and plant growth retardation[33],and these effects can be observed in the T0generation[34].The PDS gene has been used as a model gene for CRISPR/Cas9 gene editing platform construction in many crops,such as rice,tobacco,citrus,alfalfa,and banana[35–39].In this study,we also selected PDS as the target gene and developed pea plants with the albino phenotype in the T0generation.To our knowledge,this is the first successful CRISPR/Cas9 gene editing system established in pea,which connects this historical genetic model to the modern gene functional era and paves the way for further pea crop improvement.

2.Materials and methods

2.1.Plant material

The dry pea cultivar Zhongwan 6(Chinese gene bank number:G0005527)was obtained from the National Crop Genebank of China.

2.2.Agrobacterium strains

Agrobacterium rhizogenes strains K599 and Ar.Qual were used for transient expression in hairy roots.Agrobacterium tumefaciens strain EHA105 was used for stable genetic transformation.

2.3.Preparation of culture medium

The media(g L-1)used in this study were as follows.(1)Tryptone yeast(TY)liquid medium:tryptone,5.0;yeast extract,3.0;calcium chloride(CaCl2),1.1;pH 7.0.(2)TY solid medium:tryptone,5.0;yeast extract,3.0;CaCl2,1.1;agar,15;pH 7.0.(3)Yeast extract peptone(YEP)liquid medium:yeast extract,5;tryptone,10;sodium chloride(NaCl),5;pH 7.0.(4)YEP solid medium:yeast extract,5;tryptone,10;NaCl,5;agar,15;pH 7.0.(5)Germination culture medium(GM):1/2 Murashige&Skoog(MS)medium(with vitamins),2.29;sucrose,20;phytagel,3.2;pH 5.8.(6)Infection liquid medium(R1):1/10 MS medium(with vitamins),0.44;sucrose,30;2-morpholinoethanesulfonic acid(MES),3.9;acetosyringone(AS),0.02;pH 5.4.(7)Co-cultivation medium(R2):1/10 MS medium(with vitamins),0.44;sucrose,30;MES,3.9;AS,0.02;dithiothreitol(DTT),0.15;agar,8;pH 5.4.(8)Washing culture medium(WM):1/2 MS medium(with vitamins),2.29;sucrose,30;timentin,0.25;cefotaxime,0.25;pH 5.8.(9)Induction culture medium(R3):1/2 MS(with vitamins),2.29;sucrose,30;MES,0.6;phytagel,3.2;timentin,0.25;cefotaxime,0.25;pH 5.8.(10)Infection liquid medium(P1):1/10 MS medium(with vitamins),0.44;sucrose,30;MES,3.9;AS,0.02;6-benzylaminopurine(6-BA),0.002;gibberellin A3(GA3),0.00025;pH 5.4.(11)Co-cultivation medium(P2):1/10 MS medium(with vitamins),0.44;sucrose,30;MES,3.9;AS,0.02;DTT,0.15;6-BA,0.002;GA3,0.00025;sodium thiosulfate(Na2S2O3),0.16;L-cysteine(Cys),0.4;phytagel,3.2;pH 5.8.(12)Recovery culture medium(P3):MS medium(with vitamins),4.4;sucrose,30;MES,0.6;casein hydrolyzate,0.3;glutamine,0.5;proline,0.5;phytagel,3.2;6-BA,0.002;timentin,0.25;cefotaxime,0.25;pH 5.8.(13)Bud induction culture medium(P4):MS medium(with vitamins),4.4;sucrose,30;MES,0.6;casein hydrolysate,0.3;glutamine,0.5;proline,0.5;phytagel,3.2;6-BA,0.002;timentin,0.25;cefotaxime,0.25;hygromycin,0.01;pH 5.8.(14)Bud elongation culture medium(P5):MS medium(with vitamins),4.4;sucrose,30;MES,0.6;casein hydrolysate,0.3;glutamine,0.5;proline,0.5;phytagel,3.2;GA3,0.0005;3-indoleacetic acid(IAA),0.0001;zeatin(ZT),0.001;asparagine,0.05;pyroglutamic acid,0.05;timentin,0.25;cefotaxime,0.25;hygromycin,0.01;pH 5.8.(15)Rooting culture medium(P6):1/2 MS medium(with vitamins),2.29;sucrose,15;MES,0.6;casein hydrolysate,0.3;3-indolebutyric acid(IBA),0.001;asparagine,0.05;pyroglutamic acid,0.05;timentin,0.25;cefotaxime,0.25;agar,8;pH 5.8.

2.4.Preparation of explants

Pea seeds were sterilized with chlorine gas for at least 8–10 h and then placed on GM at 25 °C(16 h light/8 h darkness)for 5–7 days.Cotyledons were cut in half with 3–5 mm hypocotyls remaining.

2.5.Construction of vectors for CRISPR/Cas9

The pHUN411 binary vector[40],which contains an Oryza sativa U3 promoter cassette(OsU3)for sgRNA expression and an O.sativa codon-optimized Cas9(OsCas9)under the control of a Zea mays ubiquitin promoter(ZmUbi),was used as the backbone vector(provided by Dr.Pengcheng Wei from the Rice Research Institute,Anhui Academy of Agricultural Sciences).We modified the backbone vector as follows:the sequence of enhanced TM2-pd35S-dMac promoter(en35S)[41,42],a native P.sativum U6.3 promoter(PsU6.3)and P.sativum codon-optimized Cas9(PsCas9)were synthesized in the PUC57 vector(General Biosystems(Anhui)Co.,Ltd.,Hefei,Anhui,China).A.thaliana U6.26 promoter(AtU6.26)-SpR-sgRNA expression cassette was constructed from the pHUN 4c01 vector[43]by Hind III digestion.AtU6.26 and PsU6.3 were used to replace OsU3 to drive the expression of sgRNA,en35S was used to replace ZmUbi to drive PsCas9,and pHUN-AtU6.26-SpR-sgRNA-en35S-PsCas9-t35ST and pHUN-PsU6.3-SpR-sgRNA-en35S-PsCas9-t35ST intermediate vectors were constructed.The primer sequences used in these two intermediate vectors are listed in Supplementary Table S1.We designed five specific sgRNAs targeting the three exons of the PsPDS gene.tRNA was infused with U3/U6 promoters and PsPDS sgRNAs.SpR was digested with Bsa I and five PsPDS sgRNAs were inserted into intermediate vectors by Golden Gate Assembly[44,45].Finally,tRNA-absent and tRNA-present CRISPR/Cas9 gene editing vectors,which targeted PsPDS genes OsU3-PsPDS-ZmUbi-OsCas9,AtU6.26-PsPDS-en35S-PsCas9,PsU6.3-PsPDS-en35S-PsCas9,OsU3-tRNA-PsPDS-ZmUbi-OsCas9,AtU6.26-tRNA-PsPDS-en35S-PsCas9 and PsU6.3-tRNA-PsPDS-en35S-PsCas9,were constructed.The primer sequences used for PsPDS sgRNA in tRNA-absent and tRNApresent CRISPR/Cas9 gene editing vectors are listed in Table S2.

2.6.Agrobacterium rhizogenes-mediated transient expression in hairy roots

The plasmid vector with the GUS reporter gene(Fig.1A)was mobilized separately into K599 and Ar.Qual via the freeze–thaw method[46].After two days,single bacterial colonies were streaked in TY solid medium for the first activation at 28 °C.One day later,bacterial colonies were inoculated into TY liquid medium for a secondary activation followed by shaking at 28 °C overnight.Then the bacteria were collected by centrifugation at 5000 r min-1for 10 min,suspended in R1 to make the bacterial OD600to 0.8,and incubated at 4°C for at least 0.5 h to activate.Explants were placed into the re-suspended and activated bacterial solution and infected for 30 min.The explants were then placed onto sterile filter paper above R2 in the darkness at 28°C for 3–4 days.After co-cultivation,the explants were rinsed five times with WM and transferred onto R3 at 28 °C.The growth of hairy roots was observed 2–3 weeks later.The hairy roots were immersed into GUS staining solution(Beijing O’BioLab Co.,Ltd.,Beijing,China)[47]and incubated at 37°C for 24 h.The expression of GUS from the use of A.rhizogenes strains was photographed under an SZX16 stereo microscope(Olympus Corporation,Tokyo,Japan).

2.7.Detecting target site mutations by Hi-TOM next-generation sequencing

Specific primers for the first round of PCR were designed(Table S3)with the target sites situated within a range of 10–100 base pairs(bp)from the forward or reverse primer,and amplified product lengths of about 150–300 bp.To ensure that the target sites were amplified,5 uL of products were taken for agarose gel electrophoresis detection.The remaining PCR products were used for the second round of PCR using the Hi-TOM Kit(AHT001,Xi’an Qingxue Biotechnology Co.,Ltd.,Xi’an,Shaanxi,China)[48].The sequencing information for each sample is accessible at https://www.hi-tom.net/hi-tom/.

2.8.Efficiency of CRISPR/Cas9 gene editing systems

The final CRISPR/Cas9 vectors were transformed into A.rhizogenes strain K599 for transient genetic transformation of pea as described above.After three weeks of cultivation,genomic DNA of hairy roots was extracted by the CTAB method[49]for further detection and analyzed by Hi-TOM next-generation sequencing.

2.9.Stable Agrobacterium-mediated CRISPR/Cas9 gene editing

The most efficient vector from the pea hairy roots system,PsU6.3-tRNA-PsPDS3-en35S-PsCas9,was mobilized into EHA105 via freeze–thaw to produce a stable Agrobacterium-mediated CRISPR/Cas9 gene editing system.Two days later,single bacterial colonies were streaked in YEP solid medium for the first activation at 28 °C.Then,bacterial colonies were inoculated into YEP liquid medium for secondary activation followed by shaking at 28 °C overnight.The bacteria were collected by centrifugation at 5000 r min-1for 10 min,suspended in P1 to make the bacterial OD600to 0.2,and incubated at 4 °C for at least 0.5 h to activate.The explants were placed into the resuspended and activated bacterial solution and infected for 30 min.The explants were placed onto sterile filter paper above P2 in the darkness at 28°C for three days.After co-cultivation,the explants were rinsed five times with WM and transferred onto P3 at 28 °C(16 h light/8 h darkness)for 4–5 days.Cluster buds were induced on P4 at 28 °C(16 h light/8 h darkness)for four weeks.Buds were elongated on P5 at 28 °C(16 h light/8 h darkness)for about eight weeks.Roots were generated on P6 at 28 °C(16 h light/8 h darkness)for about six weeks.The phenotype of cluster buds was then observed,the seedlings were regenerated,and the genomic DNA of regenerated plants was extracted for Hi-TOM next-generation sequencing to detect the mutations of PsPDS.

3.Results

3.1.Screening of Agrobacterium rhizogenes strains

Fig.1.Agrobacterium rhizogenes-mediated transient transformation assay.(A)The map of the pC1391-UBI-gusplus vector.(B)Pea hairy root growing status and GUS staining mediated by different A.rhizogenes strains.The two images on the left and two on the right represent hairy root GUS staining mediated by A.rhizogenes Ar.Qual and K599 strains,respectively.Scale bars,1 cm.(C)Hairy root efficiency and GUS staining efficiency mediated by A.rhizogenes K599 and Ar.Qual strains.The orange column represents hairy root efficiency,and the green column represents GUS staining efficiency.**and***indicate statistical significance at P＜0.01 and P＜0.001,respectively.

The A.rhizogenes strains K599 and Ar.Qual were used to test hairy roots efficiency and GUS staining efficiency.Three weeks after infection and co-cultivation,most of the explants were induced with hairy roots,approximately 5 cm in length.The number and length of hairy roots were recorded and they were stained with GUS staining solution with the purpose of identifying a suitable A.rhizogenes strain for pea.The hairy roots efficiency induced by K599(66.67%)was higher than that of Ar.Qual(50%)(P＜0.01)(Fig.1C).The number and length of hairy roots per K599-transformed explant were also higher than those of Ar.Qual(Fig.1B).The GUS staining efficiency of hairy roots induced by K599(62.50%)was higher than that of Ar.Qual(28.57%)(P＜0.001)(Fig.1C),with more intense blue staining(Fig.1B).Accordingly,K599 was determined to be more suitable for further experimentation than Ar.Qual.

3.2.Target selection and vector construction for CRISPR/Cas9 systems

To test the efficiency of our CRISPR/Cas9 systems in pea,the PsPDS gene was used as target gene.First,Basic Local Alignment Search Tool(BLAST)alignment of the M.truncatula PDS(MtPDS)gene sequence with the pea reference genome was performed.We found a single-copy gene with a full length of 6582 bp and 13 exons as a match in the pea reference genome(Fig.2A).Five target sites were selected,of which three targets were located on exon 1 and one target was located on exons 2 and 3(Fig.2A).Then,we inserted the five sgRNAs into the intermediate vectors respectively to construct the final OsU3-PsPDS-ZmUbi-OsCas9,AtU6.26-PsPDS-en35S-PsCas9,and PsU6.3-PsPDS-en35S-PsCas9 vectors(Fig.2B).All sgRNAs and tRNAs were fused together and inserted into the intermediate vectors to construct the final OsU3-tRNAPsPDS-ZmUbi-OsCas9,AtU6.26-tRNA-PsPDS-en35S-PsCas9,and PsU6.3-tRNA-PsPDS-en35S-PsCas9 vectors(Fig.2B).Thus,we constructed six sets of CRISPR/Cas9 vectors to identify the one with the highest editing efficiency in pea.

3.3.Efficiency of CRISPR/Cas9 gene editing systems

The final six sets of CRISPR/Cas9 vectors were transformed into pea by A.rhizogenes-mediated transient genetic transformation.Extracted genomic DNA of T0pea hairy roots,which were induced by each CRISPR/Cas9 gene editing vector.The targeted sequence regions were amplified to detect mutations by Hi-TOM NGS sequencing.We aligned the sequencing results and recorded the proportion found with mutations in the targeted sequence in T0hairy roots.The number of T0hairy roots examined for each vector was large(from 65 to 318),the number of mutations was from 0 to 84 and the mutation rate varied from 0% to 52.38%(Fig.3A;Table S4).The mutation rate of tRNA-present vectors was generally higher than that of tRNA-absent vectors.Among them,the mutation efficiency of PsU6.3-tRNA-PsPDS3-en35S-PsCas9 was highest,reaching 52.38%(Fig.3A;Table S4).We also determined the editing efficiency of each set of vector at different target sites.The editing efficiency of the same vector varied greatly among target sites.The respective editing efficiencies of PsU6.3-en35S-PsCas9 were 17.21%,0.9%,and 4.82%at PsPDS1,PsPDS2,and PsPDS3 target sites,but there was no mutation at PsPDS4 and PsPDS5 sites(Fig.3B).The editing efficiency of PsU6.3-en35S-PsCas9 was higher than that of OsU3-ZmUbi-OsCas9 or AtU6.26-en35S-PsCas9 at the PsPDS1,PsPDS2,and PsPDS5 sites(Fig.3B).The editing efficiency of different vectors at the same target site was also very different,but in general,the tRNA-present vectors were more efficient than the tRNA-absent vectors(Fig.3B).Neither of the three tRNA-absent vectors worked at the PsPDS5 target site,but all worked when the tRNA element was present,and the editing efficiency ranged from 16.65% to 45.27%(Fig.3B).The vector with highest editing efficiency(PsU6.3-tRNA-PsPDS3-en35S-PsCas9)reached an efficiency of 87.61%in the hairy root system(Fig.3B).We accordingly suggest that the PsU6.3 promoter and tRNA processing enzymes are effective genetic engineering elements that can increase the efficiency of gene editing of CRISPR/Cas9 gene editing systems in pea.

Fig.2.Illustration of the gene editing systems.(A)Maps of the five gRNA target sites on the genomic regions of PsPDS.Blue lines represent introns;green boxes represent exons;Arabic numerals indicate the order of exons;red highlighting represents the target sequences;green highlighting represents the PAM motif(NGG).(B)Diagrams of six sets of CRISPR/Cas9 constructs.Upper three vectors,OsU3,AtU6.26,and PsU6.3,were used to drive the corresponding targeted sgRNAs.ZmUbi and en35S promoters were used to drive OsCas9 and PsCas9 expression,respectively.The bottom three vectors were based on the above three vectors,the tRNA was fused to the gRNA,and expressed as a transcription driven by OsU3,AtU6.26,and PsU6.3 promoters,respectively.

Fig.3.Mutation rate and editing efficiency of different CRISPR/Cas9 genome editing systems.(A)Mutation rate of T0 hairy roots in five PsPDS target sites induced by the six sets of CRISPR/Cas9 vectors.(B)Editing efficiency of T0 hairy roots in the five PsPDS target sites induced by the six sets of CRISPR/Cas9 vectors.The orange column represents tRNA absent,and the green column represents tRNA present.***shows statistical significance at P＜0.001,and ns indicates non-significant.

3.4.Mutation types and frequency analysis

We counted the mutation types and frequencies in different CRISPR/Cas9 systems in five PsPDS target sites.At the same target site,the tRNA-present vectors produced more abundant mutation types than the tRNA-absent vectors(Fig.S1).One of the most common forms of mutation was deletions,whose length ranged from 1 to 47 bp(Fig.S1).Single-bp insertion was also a common mutation type(Fig.S1).The mutation types and frequency of the most efficient vector(PsU6.3-tRNA-PsPDS3-en35S-PsCas9)were also examined(Fig.4).This system produced eight types of mutations,all deletions,with 1-bp deletion as the most frequent mutation type(44.7%).Large(greater than10 bp)deletions were also observed.Overall,the CRISPR/Cas9 systems were highly efficient for generating different types of mutations in pea.

3.5.Stable Agrobacterium-mediated CRISPR/Cas9 gene editing

The most efficient vector from the pea hairy root system,PsU6.3-tRNA-PsPDS3-en35S-PsCas9,was used to proceed with stable Agrobacterium-mediated CRISPR/Cas9 gene editing.As Fig.5A shows,bud induction,bud elongation,and plant regeneration were available.A total of 264 lines were generated,with 150 of these being transgenic.These 150 T0transgenic lines were used to detect mutations by Hi-TOM NGS sequencing.We found that 27 lines were mutated,including one homozygous line,one biallelic line,17 chimeric lines,and 8 heterozygous lines(Table 1).Among these,10 homozygous or non-WT-lines contained compound heterozygous plants showing albino phenotype.The two homozygous and biallelic mutants were two pairs of nucleotides deletion(Fig.5B),and other mutation types were also different deletions.

Table 1 Targeted mutagenesis frequency induced by PsU6.3-tRNA-PsPDS3-en35S-PsCas9 vector in T0 transgenic plants.

Fig.4.Mutation types and frequency induced by the PsU6.3-tRNA-PsPDS3-en35SPsCas9 vector.In the x-axis,d represents occurrence of deletion,and # represents the number of base pairs(bp)deleted from the target site.

4.Discussion

The CRISPR/Cas9 genome editing technology has become an important tool for functional genomics research and molecular breeding of plants[50]and has wide applications in genomic exploration and genetic improvement of crops[11].This approach is characterized by precise editing,simple technical operation,high efficiency,and low cost[51].Edited plants can readily lose exogenous DNA fragments by hybridization or self-pollination with wild-type plants,permitting the development of improved and gene-edited crops without transgene traces,potentially increasing regulatory and market acceptance[52].

Fig.5.Flow diagrams of stable genetic transformation in pea and sequencing results of induced mutations in PsPDS3 target site in T0 generation.(A)Flow diagrams of stable genetic transformation in pea.Images from left to right represent explants,clustered buds,bud elongation,and the successfully edited albino lines.(B)Detailed sequences and sequencing chromatograms in the PsPDS3 target site of the PsPDS3-6 and PsPDS3-15 T0 lines.WT represents the wild-type sequence.The target sequences are highlighted in red and the PAM sequences are highlighted in green.Red minus indicates deletion.d represents occurrence of deletion.#represents the number of base pairs(bp)deleted from the target site.

Transfection,AMGT and particle bombardment are the primary methods of gene editing in germline cells[21].Gene editing in plants requires genetic transformation via Agrobacterium or biolistic bombardment(the gene gun method)[53].However,gene gun delivery is expensive and Agrobacterium-mediated genetic transformation is historically inefficient and labor-intensive in pea.Compared with A.tumefaciens-mediated transformation,the A.rhizogenes-mediated hairy root transformation system has the advantages of high transformation efficiency,a short transformation period,and production of fewer chimeras and does not require the use of exogenous hormones[54].In order to quickly and efficiently deliver our CRISPR/Cas9 reagents into peas,we developed an A.rhizogenes-mediated transient expression system of hairy roots.We then optimized our CRISPR/Cas9 systems for pea genome editing,which makes it possible to accomplish gene knockouts.

From our results,although the hairy root editing efficiency of PsU6.3-en35S-PsCas9 was higher than that of OsU3-ZmUbi-OsCas9 and AtU6.26-en35S-PsCas9,the highest editing efficiency was still＜20%and no editing occurred at PsPDS4 and PsPDS5 target sites.We accordingly set out to improve the efficiency of our genome editing tools.Yang et al.[55]developed a polycistronic tRNAgRNA(PTG)system by harnessing the endogenous tRNA processing and maturation mechanism in plants,which used a pol III promoter to simultaneously transcribe and generate multiple sgRNAs.By targeting multiple sites in rice,they increased the mutation rate by up to 100%[55].Because tRNA and its processing system are conserved in all organisms,we hypothesized that it would work in pea,and in the event the editing efficiency of the tRNA–gRNA system increased from 17.21% to 87.61%,greatly increasing the number of successful editing events.Moreover,this system creates the possibility of development of multi-gene locus editing in pea in the future.

Although the PsPDS gene was successfully edited to produce the albino phenotype,the mutant lines showed severely reduced survivability and it was difficult to validate mutation stability in the offspring.Among the 27 regenerated lines of the T0generation,only two lines were biallelic or carried homozygous mutations,whereas the other 25 lines were all chimeric or carried heterozygous mutations.In order to obtain homozygous mutants,several generations of labor-intensive and time-consuming screening and validation would be needed.It will be necessary to optimize the genome editing delivery system to improve transformation efficiency and solve the problem of chimerism.

Great progress has been made in basic plant research and crop breeding with the use of the CRISPR/Cas9 genome editing technology[52].The CRISPR/Cas9 gene editing technology has the potential to increase global food security and sustainable agricultural development[56].In future research,we plan to use this advanced technology to modify peas to improve yield and biotic and abiotic stress resistance,as well as to broaden genetic diversity and create new pea germplasm for further breeding efforts.We hope that our study furthers the understanding of the potential applications of the CRISPR/Cas9 system and provides tools for the functional study and genetic improvement in pea.

5.Conclusions

By attempting to optimize the engineering reagents of the CRISPR/Cas9 constructs,we developed an Agrobacteriummediated CRISPR/Cas9 system in pea.Using this novel system,we obtained albino-phenotype mutants in the T0generation.We suggest that our system will contribute to functional genomics research and release the potential of gene editing technology to improve agronomic traits in pea.Further,our work constructs a bridge to connect this basic genetic model to the modern gene functional era.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

CRediT authorship contribution statement

Guan Li:Data curation,Formal analysis,Investigation,Writing–original draft.Rong Liu:Software,Supervision.Rongfang Xu:Methodology,Supervision.Rajeev K.Varshney:Writing–review& editing.Hanfeng Ding:Supervision.Mengwei Li:Data curation,Validation.Xin Yan:Data curation,Validation.Shuxian Huang:Data curation,Validation.Juan Li:Methodology,Supervision.Dong Wang:Visualization.Yishan Ji:Visualization.Chenyu Wang:Visualization.Junguang He:Methodology.Yingfeng Luo:Software.Shenghan Gao:Software.Pengcheng Wei:Methodology,Writing–review & editing.Xuxiao Zong:Resources,Funding acquisition,Project administration,Writing–review&editing.Tao Yang:Conceptualization,Project administration,Writing–review & editing.

Acknowledgments

We acknowledge the financial support of the China Agriculture Research System of MOF and MARA-Food Legumes(CARS-08)and the Agricultural Science and Technology Innovation Program(ASTIP)of the Chinese Academy of Agricultural Sciences.

Appendix A.Supplementary data

Supplementary data for this article can be found online at https://doi.org/10.1016/j.cj.2022.04.011.