Li Chen,Shn Yun,Yupeng Ci,Weiwei Yo,Qing Su,b,Yingying Chen,b,Jiling Zhng,b,Wensheng Hou,b,*
a National Center for Transgenic Research in Plants,Institute of Crop Sciences,Chinese Academy of Agricultural Sciences,Beijing 100081,China
b Ministry of Agriculture Key Laboratory of Soybean Biology(Beijing),Institute of Crop Sciences,Chinese Academy of Agricultural Sciences,Beijing 100081,China
Keywords:
ABSTRACT A fast and efficient recognition method of transgenic lines will greatly improve detection efficiency and reduce cost.In this study,we successfully identified the transgenic soybean plants by the color.We isolated a GmW1 gene encoding a flavonoid 3′5′-hydroxylase from a soybean cultivar ZH42(purple flower).We found that purple flowers occurred in the overexpression lines in the Jack and Williams 82 backgrounds(white flower).All plants with purple flowers were positive,and this trait seems stably inherited in the offspring.We have also obtained the editing plants,which were classified into three types according to the different flower colors appeared.We analyzed the phenotype and the homozygous types of the T1 mutants.We also found that a correspondence between flower color and stem color.This study provides a visible color reporter on soybean transformation.It can be quickly and early to identify the transgenic soybean plants by stem color of seedlings,which substantially reduces the amount of labor and cost.
Genetic engineering represents a key strategy to the production of genetic variation in crop plants that will foster improvements in agriculture productivity and sustainability[1,2].Specifically,two of the most powerful genetic engineering strategies available to improve crops are transgene and genome editing modifications.In the particular case of genome editing,the recent years have witnessed its wide application in plants[3,4].However,the detection of transgenic lines is laborious,time consuming and encompasses high financial costs.A simplicity and efficiency reporter will be great helpful for its detection.
At present,the Agrobacterium-mediated cotyledon nodes method is primarily used for soybean transformation.The transgenic soybean plants by this transformation system are directly generated from adventitious bud,and not from the callus,which induced most of the T0transgenic plants obtained are chimeric.So it is possible to obtain homozygous plants from,at least,T1generations,which involve a substantial amount of work to identify.Hence,the development of a highly efficient visual system that allows for screening transgene plants would represent a transformative technology for soybean cultivation.Importantly,the frequently used reporter systems,including the green fluorescent protein gene(GFP)[5],the β-glucuronidase gene(GUS)[6],the red fluorescent protein gene(DsRed2)[7],are widely used in rice,maize,or other plant,which can be easily detected in these transformed tissues.We have attempted the GFP and DsRed2 reporters,but they are hard to detect in soybean tissues.The above reporters are not convenient nor intuitive to use in the detection of transformed tissues in soybean,whereby we focus on those that are.
The soybean flower colors and stem colors are two very recognizable phenotypes.Soybeans have mostly either purple or white flowers,although there are some variations such as purple throat,near white,pink and magenta.Six genes(W1,W2,W3,W4,Wm,and Wp)primarily control soybean flower color.The W1,W3,W4,Wm,and Wp encode flavonoid 3′5′-hydroxylase(F3′5′H),dihydroflavonol-4-reductase 1(DFR1),DFR2,flavonol synthase,and flavanone 3-hydroxylase(F3H),respectively;W2 is an MYB transcription factor that regulates the pH of petal vacuolar sap[8].The W1 is a key enzyme in the biosynthetic pathway of flavonoids[9,10].The W1 and w1 display purple and white flower color,respectively.The white flower color of Williams 82 is due to the 65-bp insertion and 12-bp deletion in the third exon of F3′5′H,which results in premature translation and thus non-functional F3′5′H proteins[11].Regulating the W1 can be used as a phenotype marker of color and thus facilitate the identification of transgenic plants in transformed soybean.
Total RNA was extracted using Trizol reagent from the trifoliolate leaves of soybean cv.‘ZH42′(purple)seedlings.cDNA was synthesized with Superscript II reverse transcriptase(TransGen Biotech,Beijing,China)and used as a template for GmW1 cDNA cloning.Amplification was performed via PCR using KOD-plus-Neo DNA polymerase(Toyobo,Tokyo,Japan).The sequences of the primers used for amplifying the full-length GmW1 cDNA are 5′-ATGGACTCATTGTTACTTCT-3′(forward)and 5′-CTAAGAAATG TAAGCACTTG-3′(reverse).To construct the overexpression vector,the CDS of GmW1 was inserted into the PTF101 vector containing a CaMV 35S promoter and a bar gene.
A 20-bp sgRNA sequence was designed using the web tool CRISPR-P(https://cbi.hzau.edu.cn/crispr/),which was driven by the Arabidopsis U6 gene promoter.The Cas9 sequence was inserted downstream of the CaMV 2X 35S promoter.The bar gene driven by a CaMV 35S promoter was used as a screening marker[3].
The overexpression vector and CRISPR/Cas9 expression vector plasmids were transformed into Agrobacterium tumefaciens strains EHA101 and EHA105 via electroporation,respectively.The tissue culture and transformation according to a previously-published protocol[12].The soybean cultivars‘Williams 82’and‘Jack’(white flower)were used for transformation for overexpression GmW1,while the soybean cultivar‘ZH42′was used for knockout GmW1.
The GmW1-overexpressing transgenic plants were screened by LibertyLink strip detection.The LibertyLink strips were used to determine the presence of the PAT protein in the transgenic plants.The potential GmW1 mutants were then screened by PCR and DNA sequencing analysis.Genomic DNA was extracted from the leaves of each individual plant,and the regions spanning the target sites were amplified by PCR using Phanta Super Fidelity DNA Polymerase(Vazyme Biotech,Nanjing,Jiangsu,China)and sequenced.DNA from plants that were heterozygous for the mutations showed overlapping peaks on the sequencing chromatograms from the target sites to the end of the DNA fragment.The wild-type and homozygous mutations had no overlapping peaks at the target sites.The homozygous mutant types were identified by sequence alignment against the wild-type gene sequence[3].
Total RNA was extracted using Trizol reagent from the overexpression GmW1 plants.First-strand cDNA was synthesized with Superscript II reverse transcriptase(TransGen Biotech).Quantitative RT-PCR(qRT-PCR)was performed using an ABI 7500 Real-Time PCR System.The primers used for GmW1 amplification were 5′-CAACACCCCTAAACCTGCCT-3′(forward) and 5′-GAACCAACAC AATCCCCATCCT-3′(reverse).The primers used for GmActin amplification were 5′-GAGCTATGAATTGCCTGATGG-3′(forward)and 5′-CGTTTCATGAATTCCAGTAGC-3′(reverse).All PCRs were run with three biological replicates each.The relative expression level was calculated using the comparative 2-ΔΔCT method.
We cloned a 1530 bp GmW1 from the soybean cultivar ZH42(purple flower)and inserted it into the PTF101 vector,using the bar gene as a herbicide resistance marker.The two soybean cultivars Jack(white flower)and Williams 82(white flower)were used for the transformation.
We obtained 9 and 6 overexpression T0plants with the purple flower phenotype in the Jack and Williams 82 backgrounds,respectively(Fig.1a).All of these T0plants were bar gene positive,as demonstrated by PAT strips(Fig.1b).We found that purple flowers occurred in the T0overexpression lines.We then randomly selected three independent T0transgenic lines with purple flowers from each cultivar in order to examine the heritability of the phenotype in the T1generation.The seeds collected from the T0transformants were planted in two pots for each line and for the test.The results show that 23 and 24 T1seedlings germinated from the Jack and Williams 82 backgrounds,respectively.Furthermore,11 T1transgenic plants showed purple flowers in the Jack,while 12 plants contained white flowers.Analogously,we detected a total of 14 and 10 T1transgenic plants with purple and white flowers,respectively,in the Williams 82(Fig.1c).All plants were examined for the presence of the bar gene by PAT strips,which showed that all the plants with purple flower were bar positive,while those with white flowers were bar negative(Fig.1c).Given that all plants with purple flowers were positive,this trait seems stably inherited in the offspring.Hence,this confirms that the W1 gene can be used as visual reporter to improve the detection of soybean transformation.
In addition,we found the color intensity was dependent on the GmW1 expression level in overexpression GmW1 plants.The higher level of the GmW1 expression,the deeper the color.According to the color intensity,the purple was classified into light pink,pink and purple.The expression level of GmW1 gene was about 6.5 times higher in the pink flowers than in the light pink flowers,while was 11.4 times higher in the purple flowers(Fig.1d).
The target site for the knockout of GmW1 was selected in the first exon of the gene.The cultivar ZH42(purple flower)was used for transformation with the CRISPR/Cas9 vector.
Fig.1.Identification of transgenes or editing mutants based on color through the regulation of GmW1.(a)The overexpression GmW1 transgenic plants showed purple flowers.(b)The detection of T0 overexpression plants with purple flowers by PAT strips.WT,wild-type;1–9 and 10–15 are T0 overexpression lines with purple flowers in the Jack and Williams 82 background,respectively.The presence of two bands in the PAT strips was indicative of positive transformation.(c)Table showing the inheritance patterns of flower color in two cultivars.(d)Flower color and expression level of GmW1 in overexpression GmW1 transgenic plants.(e)Pie chart showing the ratio of edited mutants and wide-type plants in the T0 generation.KO represents the CRISPR/Cas9 knockout.(f)Pie chart showing the ratio of the three types of flower color in the T0 edited plants.(g)The phenotype of three types of flower colors in the T0 edited plants.Type I for purple flowers,type II for purple and white flowers,and type III for white flowers.The red and yellow arrows indicate white and purple flowers,respectively.(h)The target site and the sequence of the six homozygous mutants.(i)The white flower in homozygous T1 mutants.(j)Table showing the inheritance patterns of flower color in three types of mutants.(k)The flower and stem color of seedlings in OE plants and mutants.(l)Table showing the correspondence between flower color and stem color.
We obtained a total of 33 T0transgenic plants with the CRISPR/Cas9 vector.Genomic DNA was extracted from trifoliolate leaves at three nodes,and used for PCR amplification.We confirmed the successful editing of the T0plants after the Sanger sequencing results showed chimeric nucleotide changes at the target site.The nonedited plants showed the same sequence as the wild-type.Among the 33 T0plants,76%(25 of 33)were edited at the target site,while the remaining 24%(8 of 33)were wild-type(Fig.1e).Three different flower colors appeared in the 25 edited plants,which allowed us to classify them accordingly.Specifically,44%(11 of 25)of plants with purple flowers were classified as type I,12%(3 of 25)containing both purple and white flowers as type II,and 44% with white flowers only were regarded as type III(Fig.1f,g).From this,it can be demonstrated that the editing of soybean plants in the T0generation resulted in chimeric plants.We subsequently selected three lines from each plant type(I,II and III),planted approximately 60 seeds for each line and evaluated the phenotype and the homozygous types of the T1mutants.We sampled the trifoliolate leaves at any node of each T1plant and used them for DNA extraction and PCR.The sequencing results revealed a total of six types of homozygous mutations at the target site,including four types of deletions and two types of insertions(Fig.1h).All homozygous mutations exhibited white flowers(Fig.1i).In the offspring of type I plants,white flower plants were separated from the last generation of purple flowers,with 8 out of 19 white flower mutants being biallelic homozygous,and 11 out of 19 biallelic heterozygous.In the offspring of type II,we obtained both purple and white flower mutants.Of the latter,11 out of 24 were biallelic homozygous while the remaining 13 were biallelic heterozygous.Finally,in the offspring of type III plants,all of the mutants showed white flowers and were identified as biallelic homozygous(Fig.1j).Hence,it seems that the homozygous T1mutants might be selected from plants with white flowers.
The seeds germinated in the vermiculite.After 7 days,all the seedlings were cleaned from vermiculite and taken pictures.The wild type of Williams 82 had green stem and white flower,while all the GmW1 overexpression plants in the Williams 82 background showed purple stem and purple flower.As for the w1 mutants,the stem color were changed from purple to green,and the flower color were from purple to white in the ZH42 background.We found that a correspondence between flower color and stem color.The purple flower plants had purple stem,while the white flower plants had green stem(Fig.1k,l).So it can be quickly identified the transgene plants by stem color of seedlings without waiting for blooming.
The GmW1 gene can be utilized as a visible color reporter on soybean transformation.As for the overexpression transgenic plants,positive transgenic plants can be confirmed by evaluating flower or stem pigmentation.However,and similarly to what was observed for the mutants created by GmW1 editing,this is not a simple task,as both biallelic homozygous and biallelic heterozygous w1 mutants can exhibit the white flower and green stem phenotype.It’s not directly to determine the homologous mutants only by flower or stem pigment in the early generation.But it can effectively identify the occurrence of gene editing by analyzing flower or stem pigmentation.At least,a proportion of the mutants can be deleted by their visible color,which substantially reduces the amount of labor and cost.The homozygous mutants can first be screened by the color of their stems in early stage and then sequenced.
Subsequently,the 35S::GmW1 expression cassette can be applied and constructed into a overexpression target gene vector or target gene editing vector.For the overexpression vector,the color intensity of transgenic plants is dependent on the GmW1 expression level.The higher level of the GmW1 expression,the deeper the color.In our study,we have found the color with light pink,pink,or purple in overexpression GmW1 plants.Although,there is the possibility that the color is too weak to be recognized,we always select the relative strong expression transgenic plants to study,and these too weak expression transgenic plants may be eliminated.In this new reporter system,color intensity may be used as a reflection to expression level of overexpressed transgenic plants.For the gene editing vector,it could be possible to detect transgenic T0plants based on the purple flower phenotype after incorporating the 35S::GmW1 expression element into a CRISPR vector.The transgene-free edited plants could also be easily to isolate in the offspring by the effective phenotype marker.
In addition,this visible marker could be quickly and effectively used in the crop breeding and genome editing technologies.For the crop breeding,the haploid breeding technology has an important and valuable advantage over traditional approaches.A reliable haploid identification marker is essential for enhancing haploid breeding efficiency.Dong had successfully developed a doublefluorescence protein(DFP)marker to identify maternal haploid seeds in multiple cereal crops[13].Our visible marker system would be applicable in soybean or other dicotyledons haploid breeding.For the genome editing technologies,this visible marker is also a good system for quickly detecting the editing efficiency of the multiple targets,and validating the feasibility of a new gene editing system.
CRediT authorship contribution statement
Li Chen:Formal analysis,Writing–original draft,Project administration.Shan Yuan:Writing–original draft.Yupeng Cai:Writing–original draft.Weiwei Yao:Resources.Qiang Su:Software.Yingying Chen:Resources.Jialing Zhang:Resources.Wensheng Hou:Conceptualization,Writing–review & editing,Funding acquisition.
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.
Acknowledgments
This work was supported by the Agricultural Science and Technology Innovation Program of Chinese Academy of Agriculture Sciences(S2022ZD03).