Systematic identification of endogenous RNA polymerase III promoters for efficient RNA guidebased genome editing technologies in maize

Xinto Qi,Le Dong,Chnglin Liu,Long Mo,Fng Liu,Xin Zhng,Beijiu Cheng,Chunxio Xie,*

aInstitute of Crop Science,Chinese Academy of Agricultural Sciences,Beijing 100081,China

bAnhui Agricultural University,Anhui Province,Hefei 230036,Anhui,China

1.Introduction

RNA-guided engineered editing effectors derived from the bacteria adaptive immune system,designated as CRISPR(clustered regularly interspaced short palindromic repeat)/Cas(CRISPR-associated)nucleases[1,2],RNA effectors[3,4],and modification enzymes[5,6],enable versatile genome editing and represent exciting tools for application in diverse organisms.All of these systems depend essentially upon efficient nuclear-localized transcription of single-guide RNA(sgRNA)driven by RNA polymerase III(Pol III)promoters[1].However,sgRNA expression is the limiting factor in the optimization of targeting and mutagenesis[7].Recently,an alternative method to produce sgRNA using the Pol II promoter coupled with self-processing ribozyme-flanked RNAs was developed[8].However,high and stable small nuclear RNA transcriptional activity of RNA Pol III,which occupies approximately 40%of total RNA[9],suggests that the Pol III promoter might play a primary role in RNA-guided genome editing strategies.Spliceosomal RNAs,such as U6,are conserved from yeast to mammals[10].However,it is rare that foreign U6 or U3(Pol III)promoters work well in RNA-guided genome-editing technologies applied across species[11].For this reason,the identification of endogenous RNA Pol III promoters is one of the fundamental steps in optimizing RNA-guided genome-editing systems in target organisms.The objective of this study was to perform a systematic evaluation of endogenous Pol III promoters for RNA-guided genome technologies in maize,one of the most important cereal crops in the world.

2.Materials and methods

2.1.DNA extraction and isolation of ZmU6 promoters and ZmU3 promoter

DNA was isolated and purified using the DNeasy plant mini kit(Qiagen,Germany)according to the manufacturer's protocols.ZmU6 promoters and the ZmU3 promoter were amplified from the maize inbred‘Zheng 58'.PCR was performed to amplify the promoters using KOD-plus polymerase(Catalog#:KOD-401,TOYOBO Life Science Department,Osaka,Japan)with specific primers(Table S1).The PCR amplicons were clonedinto the pEASY-blunt vector(Transgene,Beijing,China)and then sequenced following standard protocols.

2.2.Construction of the Traffic Light Reporter(TLR)system vector

The TLR system vector has an artificial target 23 bp DNA sequence, 5′-GAGAGAGCGTGTGTCGTCTCCGGG-3′, between the start codon, ATG, and an eGFP followed by a DsRed-2 reading frame and the two gene (eGFP and DsRed-2) were separated by two nucleotides, CC.

The maize U6 and U3 promoters were used to drive the sgRNA gene,and the promoters and sgRNA genes were cloned into the TLR vector following the manufacturer's suggested protocols.For the U3 promoter,the first base was dropped from the target sequence so that transcription would start at an “A”.Thus,the target sequence for the maize U3 promoter was 5′-AGAGAGCGTGTGTCGTCTCCGGG-3′,in which the last G is in the eGFP open reading frame.

2.3.Mesophyll protoplast culture and transformation

Protoplasts were prepared from the seedling leaves of the maize hybrid Zhongdan 99,based on a previously reported method[12].A 200-μL volume of mesophyll protoplast containing 4 × 105protoplasts with 40 μg plasmid DNA(20 μg of the TLR and 20 μg of the Ubi promoter drived the spCRISPRC as 9 gene vector)was mixed with 220 mL of 45%(4.5 g per 10 mL,[w/v])polyethylene glycol PEG-4000(Sigma-Fluka,catalog number 81242,USA)solution at room temperature in the dark for 15 min.After addition of 880 mL W5 solution(154 mmol L?1NaCl,125 mmol L?1CaCl2,5 mmol L?1KCl,5 mmol L?1glucose,0.03%MES,pH to 5.8 with KOH.)to stop the reaction,the protoplasts were harvested by centrifugation at 100g for 3 min.They were resuspended in 2 mL W5 solution and incubated in 6-well plates in the dark at 28°C for 48 h.

2.4.Confocal laser scanning microscopy and flow cytometry

eGFP and DsRed-2 signals resulting from induced mutations were detected with a Zeiss(LSM700,Germany)microscope using 488-nm(eGFP)and 558-nm(DsRed-2)excitation wavelengths with 10 times eye lens,20 or 40 times objective lens magnification.Flow cytometry of TLR samples was performed on a BD FACSAriaIII instrument(BD,USA)using 488-nm(eGFP,FITC)and 558-nm(DsRed-2)excitation wavelengths.Briefly,protoplasts were harvested after incubation by centrifugation at 100g for 3 min and resuspended in 200 μL W5 solution.The suspension was loaded into a Round-Bottom Tubes(BD,USA).Gates were set to capture positive cells,using a negative control to determine autofluorescence thresholds for nonexpressing cells.

2.5.Construction of the RNA-guided Cas9 vector

The expression cassette of the maize ubiquitin promoter was used to derive the modified coding sequence of SpCas9 gene,which was constructed based on the CPB vector.The nuclear localization signal sequence of SV40 and nucleoplasmin were embedded at either end of the expressed Cas9 protein.The guide RNA sequence 5′-CCGACTACCCGGAGCTGAACCTC-3′was selected to target the maize ZmWx1 gene within the coding region of exon 7 at chromosome 9 from 23,267,684 to 23,271,612 bp(AGP v3.0).The underlined CCG represents the proto-adjacent-motif(PAM).The maize U6-2 promoter described in this report was used to express the sgRNA gene targeting ZmWx1.

2.6.Maize transformation

ZC01,a private inbred line,was transformed based on a modified Agrobacterium tumefaciens(EHA105 strain)-mediated immature embryo transformation protocol[13].The procedure and the characterization of transformation products were essentially as previously described[14].

2.7.Detection of genome modifications

Genomic DNA was extracted from maize T0 and T1 Wx1 mutants.PCR was performed to amplify the genomic regions surrounding the CRISPR target sites using KOD-plus polymerase(TOYOBO,Life Science Department,Osaka,Japan).The PCR amplicons were cloned into the pEASY-blunt vector(Transgen)and sequenced.M13 primers were used for Sanger sequencing on an ABI3730 instrument(Applied Biosystems,California,USA).

Fig.1–A scheme illustrating TLR system for efficient RNA Pol III promoter component evaluation.(A)Stable expression vector for Cas9 nuclease in TLR.(B)The TLR expression vector of single-guide RNA(sgRNA)targeting 23 bp artificial DNA in panel C using maize U6/U3(six U6 and one U3)promoters.The U3 promoter was designed to target a 5′-

2.8.Phenotype of amylopectin in grain endosperm and pollen grains

Observations of grain and pollen amylopectin followed Hunt et al.[15].Kernels were harvested at maturity and soaked overnight.Endosperm was excised,placed under glass slides,and stained with Lugol's solution(10%[w/v]KI,5%[w/v]I2),diluted 100-fold with water immediately prior to use.Pollen was treated with Carnoy's reagent(absolute ethyl alcohol:acetic acid 1:1)for 24 h and stained with Lugol's solution diluted as described above.The starch-granule color of kernel endosperm or pollen grains were observed under 10×magnification under a microscope(MSHOT,Guangzhou,China).

Fig.2–Confocal microscopy(Zeiss,LSM700,Germany)profiles of a maize mesophyll protoplast cell harboring both mutation types leading to eGFP expression(excitation wavelength,488 nm)and DsRed-2 expression(554 nm)mutations in the Traffic Light Reporter(TLR)system.(A)488 nm;(B)bright view control;(C)554 nm;D)488 nm and 554 nm.

Fig.3–Flow cytometry(BD,USA)analysis profiles of seven diverse maize RNA Pol III promoters at excitation wavelengths of 488 nm(FITC and GFP,Q4),554 nm(DsRED,Q1)and both(Q2)in the TLR system.(A)U6-1;(B)U6-2;(C)U6-3;(D)U6-4;(E)U6-5;(F)U6-6;(G)U3;(H)mutation rates of all 7 Pol III promoters evaluated by flow cytometry.

3.Results

3.1.The rationale for the experimental design

We established an in vitro TLR system that can be used in maize mesophyll protoplasts for a rapid test of promoter activity(Fig.1).Briefly,mesophyll protoplasts were isolated from plants stably expressing SpCas9(Fig.1A).The protoplasts were transiently transformed with sgRNA expression constructs(Fig.1B),in which sgRNA was expressed in maize using several different constructs,each with either six U6 promoters or one U3 promoter,transfected by PEG-4000.The sgRNA/Cas9complex,RNA-guidedendonuclease(RGEN),targets an artificial 23-bp DNA,5′-GAGAGAGCGTGTGTCG TCTCCGGG-3′,located between the start codon,ATG,and an eGFP followed by a DsRed-2 reading frame and the two genes(eGFP and DsRed-2)were separated by two nucleotides,CC(Fig.1C).In addition,for the U3 promoter,the first base was dropped from the target sequence so that transcription would start at an “A”.Thus,the target sequence for the maize U3 promoter was 5′-AGAGAGCGTGTGTCGTCTCCGGG-3′,in which the last G is in the eGFP open reading frame.(Fig.1B).This design produces both a frameshifted eGFP and a DsRed-2.Small insertions or deletions(InDels)are directly induced within the 23-bp region(Fig.1C)by the sgRNA/Cas9 RGEN,potentially resulting in eGFP open reading frame restoration(Fig.1D)when 3n+1 or 3n?2 bp InDels are induced,or resulting in DsRed-2 open reading frame restoration(Fig.1E)when 3n?1 or 3n+2-bp InDels are induced in the target 23 bp region.If a cell harbors both eGFP and DsRed-2 restoration mutations,it should fluoresce in yellow.

Fig.4–In vivo RNA-guided Cas9-mediated target mutations in the ZmWx1 locus.(A)The expression cassette of RNA-guided Cas9.Bar,bialaphos resistance marker;CaMV poly(A),CaMV poly(A)signal sequence;E 35S promoter,enhanced 35S promoter;LB,T DNA left border;NLS,nuclear localization signal sequence(SV40 and nucleoplasmin NLS sequences were used at both ends of the Cas9 nuclease);Nos,Nos terminator;RB,T DNA right border;sgRNA1 and sgRNA1 targeted residue;sgRNA,single guide RNA;SpCas9,Streptococcus pyogenes Cas9;Ubi,maize ubiquitin promoter;ZmU6-2,maize U6-2 RNA Pol III promoter.(B)The target region within the ZmWx1 gene structure and the counts of observed target mutations in the T0 generation.PAM,protospacer-adjacent motif sequence(sequence NGG underlined in red);Sequence in yellow shading,expressed sgRNA region complementary to target region DNA.WT,wild-type sequence of the target region;In,insertion mutation;Del,deletion mutation.The number before In or Del indicates the bp number of the mutation site.Counts,the numbers of mutations found.

3.2.Observation of mutations by confocal microscopy

After transformation,many mutations occurred in cells,leading to green(mutations resulting in eGFP restoration),red(mutations resulting in DsRed restoration)or yellow(both mutation types occurring in the same cell,resulting in eGFP and DsRed-2 restoration)fluorescence.Fig.2 shows an example cell showing restoration of mutations in both eGFP and DsRed-2 in bright view(Fig.2B),eGFP excitation(Fig.2A),DsRed-2excitation(Fig.2C),and bothe GFP and DsRed-2excitation(Fig.2D).Cells withe GFP(Fig.2A)restoration emitted green light,those with DsRed-2(Fig.2C)restoration red light,and those with both restorations yellow light(Fig.2D).

3.3.Systematic evaluation of the maize endogenous RNA pol III promoters

Fig.5–Phenotypes of heritable knockout mutations of ZmWx1 in kernel and pollen starch were characterized by potassium iodide(10%KI)staining in T1 homozygous wx1 lines.

Using flow cytometry,cells harboring different mutations with different fluorescence patterns were easily plotted into regions and counted.Our in vitro data from the TLR system using flow cytometry(Fig.3)showed that U6-6(Fig.3F),U6-2(Fig.3B),U6-4(Fig.3D),and U3(Fig.3G)had mutation frequencies of 21.0%(sum of green,red,and yellow fluorescence),19.6%,17.9%,and 9.7%,respectively(Fig.3H).Judging from the distribution of cells with the three colors,theU6-2,U6-4,U6-6,andU3promoters appeared to yield higher mutation efficiency than the remaining three,U6-1(Fig.3A),U6-3(Fig.3C),and U6-5(Fig.3E).Specifically,the U6-6 promoter yielded a 21%mutation rate,whereas the U6-2,U6-4,and U3 promoters yielded 19.6%,17.9%,and 9.7%mutation rates,respectively.The TLR assay indicates the variation in transcription activities of the seven Pol III promoters,from 3.4%(U6-1)to over 21.0%(U6-6).

3.4.In vivo verification of RNA-guided Cas9 targeting activity using U6-2

The in vitro data suggested that all seven Pol III promoters operated on the CRISPR/Cas9 construct,but with mutation varying from 3.4%(U6-1)to over 21.0%(U6-6),indicating different activities.In Fig.4 a detailed example of RNA-guided Cas9 targeting the ZmWx1 locus is shown,wherein a singleguide RNA was constructed that was driven by the promoter U6-2.An expression cassette of RNA-guided Cas9(Fig.4A)was designed to target the ZmWx1 locus(Fig.4B).Among the 66 T0 transformation-positive events from three rounds of transformation,the targeted mutation efficiencies ranged from 48.5%to 97.1%.Among them,36 insertions of a single T and 15 of a single A comprised the majority of the mutation types.The mutation frequencies were high,such that 32 of 66recessive wx1 mutant homozygotes exhibited a waxy phenotype,in which amylose could not be detected in the kernels of T0 plants.Selfing those families revealed that the targeted mutations all heritably produced recessive wx1 homozygous phenotypes.The kernel endosperm starch and pollen phenotypes of four randomly selected T1 families(CNW-D19,CNWD5,CNW-D21,and CNW-D45)were scored based on KI staining(Fig.5)and amylose content determination(Table 1).In addition to the visible phenotypes of the KI staining of both the kernel endosperm starch particles and pollen cells,the amylose contents of the mutants were low to undetectable,compared with amylose contents of approximately 18%in the wild type(Table 1).These in vivo data for ZmWx1 suggest that the U6-2 promoter could be used as a key element of CRISPR/Cas genome editing tools and will exhibit robust activity.

Table1–Average amylose contents of selected T1 homozygous lines.

4.Discussion

For three potential promoters(U6-2,U6-4,and U6-6)showing higher mutation frequencies,in vivo verification was producedfor U6-2.More experiments should be performed to verify all the contrasting frequencies of U6 and U3 promoters from the different maize endogenous Pol III promoters observed in this study.In our previous study[14],RNA-guided Cas9 based on the in vivo U6-6 promoter yielded efficiencies of up to 91.2%.We also obtained in vivo data for the important of Pol III promoters,U6-4,with efficiencies ranging from 42.9%to 78.9%on a RGEN-targeting ZmAGO1 locus(detailed data will be published in a separate paper),an important target gene for small RNA processing[16].Table 2 shows our in vitro data on mutation frequencies and summarizes in vivo mutation frequencies found using maize endogenous RNA Pol III promoters.The few reported results of in vivo gene targeting efficiency are consistent with our findings.Our data support the establishment of an efficient genome editing technology system relying on the RNA Pol III promoters for expressing single-guide RNAs.

Table 2–In vitro TLR-characterized targeting efficiency of U6 or U3 promoters as verified by in vivo data from the literature or our unpublished results.

In summary,we successfully established a TLR system for detecting the editing efficiency of sgRNAs.Such a system can also be used in an efficiency assay for modified Cas proteins or other components of genome editing systems.These results along with the promoter sequences(Table S2)provide an essential basis for developing genome editing technologies in maize.The in vitro TLR system reported here could also serve as an example of a key starting point for the development of genome editing technologies for other species.

Supplementary tables for this article can be found online at https://doi.org/10.1016/j.cj.2018.02.005.

Acknowledgments

This research was supported by the National Science Foundation of China(31771808),Ministry of Science and Technology(2015BAD02B0203),National Engineering Laboratory of Crop Molecular Breeding,and the Chinese Academy of Agricultural Sciences(Y2017XM03).

[1]M.Jinek,K.Chylinski,I.Fonfara,M.Hauer,J.A.Doudna,E.Charpentier,A programmable dual-RNA–guided DNA endonuclease in adaptive bacterial immunity,Science 337(2012)816–821.

[2]B.Zetsche,J.S.Gootenberg,O.O.Abudayyeh,I.M.Slaymaker,K.S.Makarova,P.Essletzbichler,S.E.Volz,J.Joung,J.van der Oost,A.Regev,E.V.Koonin,F.Zhang,Cpf1 is a single RNA-guided endonuclease of a class 2 CRISPR-Cas system,Cell 163(2015)759–771.

[3]O.O.Abudayyeh,J.S.Gootenberg,S.Konermann,J.Joung,I.M.Slaymaker,D.B.T.Cox,S.Shmakov,K.S.Makarova,E.Semenova,L.Minakhin,K.Severinov,A.Regev,E.S.Lander,E.V.Koonin,F.Zhang,C2c2 is a single-component programmable RNA-guided RNA-targeting CRISPR effector,Science 353(2016),aaf5573.

[4]A.A.Smargon,D.B.T.Cox,N.K.Pyzocha,K.Zheng,I.M.Slaymaker,J.S.Gootenberg,O.A.Abudayyeh,P.Essletzbichler,S.Shmakov,K.S.Makarova,E.V.Koonin,F.Zhang,Cas13b is a type VI-B CRISPR-associated RNA-guided RNnase differentially regulated by accessory proteins Csx27 and Csx28,Mol.Cell 65(2017)618–630.

[5]A.Vojta,P.Dobrinic,V.Tadic,L.Bockor,P.Korac,B.Julg,M.Klasic,V.Zoldos,Repurposing the CRISPR-Cas9 system for targeted DNA methylation,Nucleic Acids Res.44(2016)5615–5628.

[6]Y.Zong,Y.P.Wang,C.Li,R.Zhang,K.L.Chen,Y.D.Ran,J.L.Qiu,D.W.Wang,C.X.Gao,Precise base editing in rice,wheat and maize with a Cas9-cytidine deaminase fusion,Nat.Biotechnol.35(2017)438–441.

[7]M.Jinek,A.East,A.Cheng,S.Lin,E.Ma,J.Doudna,RNA-programmed genome editing in human cells,elife 2(2013),e00471.

[8]X.Tang,L.G.Lowder,T.Zhang,A.A.Malzahn,X.Zheng,D.F.Voytas,Z.Zhong,Y.Chen,Q.Ren,Q.Li,E.R.Kirkland,Y.Zhang,Y.Qi,A CRISPR-Cpf1 system for efficient genome editing and transcriptional repression in plants,Nat.Plants 3(2017),17018.

[9]M.R.Paule,R.J.White,Transcription by RNA polymerases I and III,Nucleic Acids Res.28(2000)1283–1298.

[10]D.A.Brow,C.Guthrie,Spliceosomal RNA U6 is remarkably conserved from yeast to mammals,Nature 334(1988)213–218.

[11]L.Bortesi,R.Fischer,The CRISPR/Cas9 system for plant genome editing and beyond,Biotechnol.Adv.33(2015)41–52.

[12]J.Sheen,Signal transduction in maize and Arabidopsis mesophyll protoplasts,Plant Physiol.127(2001)1466–1475.

[13]Y.Ishida,H.Saito,S.Ohta,Y.Hiei,T.Komari,T.Kumashiro,High efficiency transformation of maize(Zea mays L.)mediated by Agrobacterium tumefaciens,Nat.Biotechnol.14(1996)745–750.

[14]C.Li,C.Liu,X.Qi,Y.Wu,X.Fei,L.Mao,B.Cheng,X.Li,C.Xie,RNA-guided Cas9 as an in vivo desired-target mutator in maize,Plant Biotechnol.J.15(2017)1566–1576.

[15]H.V.Hunt,H.M.Moots,R.A.Graybosch,H.Jones,M.Parker,O.Romanova,M.K.Jones,C.J.Howe,K.Trafford,Waxy phenotype evolution in the allotetraploid cereal broomcorn millet:mutations at the GBSSI locus in their functional and phylogenetic context,Mol.Biol.Evol.30(2013)109–122.

[16]D.Xu,H.Yang,C.Zou,W.Li,Y.Xu C.Xie,Identification and functional characterization of the AGO1 ortholog in maize,J.Integr.Plant Biol.58(2016)749–758.

[17]S.Svitashev,J.K.Young,C.Schwartz,H.Gao,S.C.Falco,A.M.Cigan,Targeted mutagenesis,precise gene editing,and sitespecific gene insertion in maize using Cas9 and guide RNA,Plant Physiol.169(2015)931–945.

[18]Z.Liang,K.Zhang,K.Chen,C.Gao,Targeted mutagenesis in Zea mays using TALENs and the CRISPR/Cas system,J.Genet.Genomics 41(2014)63–68.