Genome Editing and Double-Fluorescence Proteins Enable Robust Maternal Haploid Induction and Identification in Maize
Le Dong,Lina Li,Changlin Liu,Chenxu Liu,Shuaifeng Geng,Xinhai Li,Changling Huang,Long Mao,Shaojiang Chen,Chuanxiao Xie
DOI: https://doi.org/10.1016/j.molp.2018.06.011
IF: 27.5
2018-01-01
Molecular Plant
Abstract:Genome editing technologies have paved the way for exciting and novel applications in plant biotechnology. Doubled haploid (DH) technology has a significant and valuable advantage over traditional approaches in crop breeding. Unlike traditional breeding processes, which may take over eight generations to stabilize the genetic background of interest, DH allows homozygous lines to be obtained within two generations (Kelliher et al., 2017Kelliher T. Starr D. Richbourg L. Chintamanani S. Delzer B. Nuccio M.L. Green J. Chen Z. McCuiston J. Wang W. et al.MATRILINEAL, a sperm-specific phospholipase, triggers maize haploid induction.Nature. 2017; 542: 105-109Crossref PubMed Scopus (207) Google Scholar). Recently, a gene called MATRILINEAL (MTL)/ZmPLA1 has been cloned from maize maternal haploid inducer lines and found to be conserved across cereal crops (Kelliher et al., 2017Kelliher T. Starr D. Richbourg L. Chintamanani S. Delzer B. Nuccio M.L. Green J. Chen Z. McCuiston J. Wang W. et al.MATRILINEAL, a sperm-specific phospholipase, triggers maize haploid induction.Nature. 2017; 542: 105-109Crossref PubMed Scopus (207) Google Scholar, Liu et al., 2017Liu C. Li X. Meng D. Zhong Y. Chen C. Dong X. Xu X. Chen B. Li W. Li L. et al.A 4-bp insertion at ZmPLA1 encoding a putative phospholipase A generates haploid induction in maize.Mol. Plant. 2017; 10: 520-522Abstract Full Text Full Text PDF PubMed Scopus (145) Google Scholar). The gene, together with ever-advancing genome editing technology, provide the potential to generate maternal haploid inducer lines among multiple cereal crops. Haploid induction (HI) and haploid identification (HID) along with chromosome doubling are essential components of DH technologies (Geiger, 2009Geiger H.H. Doubled haploids.in: Bennetzen J.L. Hake S. Handbook of Maize: Genetics and Genomics. Springer, New York, NY2009: 641-657Crossref Scopus (69) Google Scholar). Therefore, a reliable HID marker or method is essential to establish any effective DH breeding approach. Currently, an R1-nj anthocyanin marker within haploid inducer has been established in HID based on the purple anthocyanin in both endosperm and embryo. Haploid kernels can be identified according to their scutellum color where no purple pigmentation should present in the scutellum of the maternal haploid embryo due to the lack of R1-nj from inducers (Nanda and Chase, 1966Nanda D.K. Chase S.S. An embryo marker for detecting monoploids of maize (Zea mays L.).Crop Sci. 1966; 6: 213-215Crossref Google Scholar). HID through the R1-nj marker was found to be inefficient in some germplasms due to complete inhibition or poor expression of this gene (Chaikam et al., 2015Chaikam V. Nair S.K. Babu R. Martinez L. Tejomurtula J. Boddupalli P.M. Analysis of effectiveness of R1-nj anthocyanin marker for in vivo haploid identification in maize and molecular markers for predicting the inhibition of R1-nj expression.Theor. Appl. Genet. 2015; 128: 159-171Crossref PubMed Scopus (48) Google Scholar). Large-scale HID through the R1-nj phenotype in maize requires intensive labor with skillful experience in artificial selection. Alternatively, HID can be done through oil content measurements by using high oil inducer lines and screened at high speed and accurately using automatic nuclear magnetic resonance (Li et al., 2009Li L. Xu X. Jin W. Chen S. Morphological and molecular evidences for DNA introgression in haploid induction via a high oil inducer CAUHOI in maize.Planta. 2009; 230: 367-376Crossref PubMed Scopus (66) Google Scholar). This technology, however, is limited to high oil content inducers and low oil content receptors (Li et al., 2009Li L. Xu X. Jin W. Chen S. Morphological and molecular evidences for DNA introgression in haploid induction via a high oil inducer CAUHOI in maize.Planta. 2009; 230: 367-376Crossref PubMed Scopus (66) Google Scholar, Melchinger et al., 2014Melchinger A.E. Schipprack W. Friedrich Utz H. Mirdita V. In vivo haploid induction in maize: identification of haploid seeds by their oil content.Crop Sci. 2014; 54: 1497-1504Crossref Scopus (50) Google Scholar). Importantly, gene introgression of R1-nj or high oil trait into a potential maternal haploid inducer requires a long process with intensive backcrossing efforts. Furthermore, neither R1-nj nor oil content analysis has the potential to be employed generally as an HID marker for other cereals such as rice, wheat, and barley. Thus, it is pivotal to establish a stable and efficient HID system that can be utilized to create a maternal haploid inducer for any genetic background in maize and can also be extended to other important cereal crops. To meet these demands, we developed haploid inducer lines by targeting the ZmMTL/ZmPLA1 gene using CRISPR/Cas9, which were then crossed with an HID line that carried double-fluorescence protein (DFP) marker to identify maternal parthenogenesis haploid seeds. This system takes advantage of the CRISPR/Cas9-mediated gene editing of maize ZmMTL/ZmPLA1 for generating haploid seeds, which are then being screened using two inflorescent protein genes, enhanced green fluorescent protein (eGFP) and DsRED, driven by an embryo-specific promoter (Liu et al., 2014Liu X. Tian J. Zhou X. Chen R. Wang L. Zhang C. Zhao J. Fan Y. Identification and characterization of promoters specifically and strongly expressed in maize embryos.Plant Biotechnol. J. 2014; 12: 1286-1296Crossref PubMed Scopus (17) Google Scholar) and an endosperm-specific promoter, respectively (Kalla et al., 1994Kalla R. Shimamoto K. Potter R. Nielsen P.S. Linnestad C. Olsen O.-A. The promoter of the barley aleurone-specific gene encoding a putative 7 kDa lipid transfer protein confers aleurone cell-specific expression in transgenic rice.Plant J. 1994; 6: 849-860Crossref PubMed Scopus (117) Google Scholar). Upon crossing with receptor lines for HIs, diploid progenies will show red fluorescence in endosperm and green in embryo. Haploids should show red fluorescence in endosperm and no fluorescence in embryo because of the lack of the genome of the inducer. Thus, the fluorescence emission patterns of both red and green colors, endosperm-specific red color only, or no fluorescence in the progeny could be used for identifying diploid, haploid, and outcross kernels in DH breeding practice. DH lines can be obtained by subsequent chemical treatment of the haploid plants. Since MTL/PLA1 is a conserved gene across cereal crops (Kelliher et al., 2017Kelliher T. Starr D. Richbourg L. Chintamanani S. Delzer B. Nuccio M.L. Green J. Chen Z. McCuiston J. Wang W. et al.MATRILINEAL, a sperm-specific phospholipase, triggers maize haploid induction.Nature. 2017; 542: 105-109Crossref PubMed Scopus (207) Google Scholar), this system could be applicable in other major crops. To prove the concept, we firstly transformed the CRISPR/Cas9 construct targeting ZmMTL/ZmPLA1 (Figure 1A) into maize variety ZC01 by using agrobacterium-mediated transformation method as described previously (Li et al., 2017Li C. Liu C. Qi X. Wu Y. Fei X. Mao L. Cheng B. Li X. Xie C. RNA-guided Cas9 as an in vivo desired-target mutator in maize.Plant Biotechnol. J. 2017; 15: 1566-1576Crossref PubMed Scopus (64) Google Scholar). Among 58 independent T0 transgenic plants, the mutation rate was 87.06% (Supplemental Table 1). Analysis of the ZmMTL/ZmPLA1 locus in transgenic lines (referred to as mMTLs) showed that most of the targeted mutations took place about 3 bp upstream of the protospacer-adjacent motif (Supplemental Table 2), which is consistent with the original characterization on the CRISPR/Cas9 system in maize (Jinek et al., 2012Jinek M. Chylinski K. Fonfara I. Hauer M. Doudna J.A. Charpentier E. A programmable dual-RNA–guided DNA endonuclease in adaptive bacterial immunity.Science. 2012; 337: 816-821Crossref PubMed Scopus (9424) Google Scholar). As shown in Figure 1B, the designed target region was very close to the reported natural mutation site (Kelliher et al., 2017Kelliher T. Starr D. Richbourg L. Chintamanani S. Delzer B. Nuccio M.L. Green J. Chen Z. McCuiston J. Wang W. et al.MATRILINEAL, a sperm-specific phospholipase, triggers maize haploid induction.Nature. 2017; 542: 105-109Crossref PubMed Scopus (207) Google Scholar) so that the best mutation effect can be achieved. Four independent homozygous mMTLs T1 lines were identified to have the CRISPR/Cas9 cassette being segregated out and were used as haploid inducer lines (Figure 1B). We then tested whether the DFP cassette was functional. We transiently expressed the cassette in maize maturing seeds by DNA bombardment. As shown in Supplemental Figure 1, the eGFP gene was found to be expressed in embryos (green) and DsRED was expressed in endosperms (red). The DFP cassette was then stably transformed into the maize variety ZC01, the same inbred line as the created haploid inducer line (Figure 1C). Positive transformants were identified by the presence of fluorescence colors (Supplemental Figure 2), which can also be used to check their inheritance patterns in subsequent generations. To further characterize the seed DFP phenotype, kernels of T1 homozygous transformants were sectioned longitudinally and aleurone-specific red fluorescence and embryo-specific red fluorescence were clearly observed (Figure 1D). Tissue-specific patterns of seed DFP were then verified using confocal microscopy (Figure 1E). These data indicate that the DFP construct works in transgenic plants as expected. To generate haploid inducer lines stacked with efficient HID markers, we crossed mMTL transgenic lines carring the mutated MTL gene with those carrying the DFP cassette, and three haploid inducer lines with robust HID marker were developed (Supplemental Table 2). These lines were named DFP-mediated haploid inducer lines (DHILs). To validate the efficiency of our haploid inducing and screening system, two DHILs (T02-8 and T02-59) were used as male parents to cross with three receptor varieties: ZD99, Zeyu7, and Feng8 (Supplemental Table 3). An exemplar of ears carrying diploid, haploid, and outcross kernels that was derived from DHIL 57-1 and the receptor ZD99 was checked under lights of different wavelengths by corresponding light filters. As shown in Supplemental Figure 3, aleurone-specific DsRED (bottom left panel) and embryo-specific eGFP (the right panel) were readily detected under 540 nm and 480 nm filters. The red colors indicated that endosperms of most kernels were fertilized by DHIL pollens. The ear was put in the upright position to reveal green colors in embryos that were obstructed from observation at the horizontal position (bottom left). As indicated, kernels with no fluorescence color were derived from outcross pollens that were mixed during pollination. To give a better view of the induced haploid kernels, we removed the kernels of the three types from the ear and aligned them in order. As shown in Figure 1F, red endosperms were seen in both diploid and haploid kernels, whereas green embryos were only observed in diploid kernels. This was further confirmed by merging the images taken under 540 nm and 480 nm fluorescent light when yellow colors were achieved in diploid kernels and only red colors were shown in the haploids (the right most panel). We then removed embryos from 25-day-old maturing kernels and found that embryos from haploid kernels were smaller than those of diploid ones, which could be identified with green fluorescence (Figure 1G), indicating that our seed DFP markers could be used for HID among seeds with young embryos. The haploid seeds then can be used for chromosome doubling treatment to achieve diploid plants, an important approach in DH technology (Kasha, 2005Kasha K.J. Chromosome doubling and recovery of doubled haploid plants.in: Don Palmer C.E. Keller W.A. Kasha K.J. Haploids in Crop Improvement II. Springer, Berlin, Heidelberg2005: 123-152Crossref Scopus (52) Google Scholar). Thus, our HID marker is robust for enabling HID for both mature seeds and young embryos in a DH breeding program. To verify the haploid nature of the haploid plants generated, we analyzed their DNA contents in cells prepared from three-leaf stage young plants by using flow cytometry as described previously (Liu et al., 2017Liu C. Li X. Meng D. Zhong Y. Chen C. Dong X. Xu X. Chen B. Li W. Li L. et al.A 4-bp insertion at ZmPLA1 encoding a putative phospholipase A generates haploid induction in maize.Mol. Plant. 2017; 10: 520-522Abstract Full Text Full Text PDF PubMed Scopus (145) Google Scholar). As shown in Figure 1H, the left panel shows the DNA content from the haploid cells at phases G1 and G2 with FL2-A values of ∼50 and ∼100, respectively, while those for diploid cells were ∼100 and ∼200 (right panel), indicating that at both cell-cycle phases, diploid cells had twice the amount of DNA as haploids, indicating that the haploids made from our HI system were real haploids. We also measured the HI rate as well as the abortion rate and the aneuploidy rate among seeds from all the receptor varieties. All kernels that successfully germinated were subject to ploidy analysis using flow cytometry. As shown in Supplemental Table 3, we found that the HI rate ranged from 4.7% to 11.0% with an average of 7.47%, while the average abortion and aneuploidy rates were 12.30% and 8.32%, comparable with what has been reported by Kelliher et al., 2017Kelliher T. Starr D. Richbourg L. Chintamanani S. Delzer B. Nuccio M.L. Green J. Chen Z. McCuiston J. Wang W. et al.MATRILINEAL, a sperm-specific phospholipase, triggers maize haploid induction.Nature. 2017; 542: 105-109Crossref PubMed Scopus (207) Google Scholar. Thus, such an HI rate should be sufficient to establish an effective DH breeding practice in maize. To verify genome zygosity of the induced haploids, two polymorphic simple sequence repeat (SSR) markers (Supplemental Table 4) between the receptor ZD99 and the inducer line 57-1 were obtained by screening 40 SSR loci (Supplemental Materials and Methods), which showed clear polymorphism by agarose gel electrophoresis. The homozygosity of haploids was verified at two independent loci: bnlg2082 at maize chromosome bin8.03 (upper panel, Figure 1I) and bnlg2259 at bin 7.04 (lower panel, Figure 1I). The absence of the transgene containment within the induced maternal haploids tested negative by the Bar strip test or by PCR amplification of the Bar gene (Supplemental Figure 4). These data indicate that the maternal haploids produced were “transgene free” and lacked the CRISPR/Cas9 and the DFP expression cassettes. To test whether maize seed DFP marker can be applied to other cereal species since the MTL/PLA1 gene is conserved among major cereal plants Kelliher et al., 2017Kelliher T. Starr D. Richbourg L. Chintamanani S. Delzer B. Nuccio M.L. Green J. Chen Z. McCuiston J. Wang W. et al.MATRILINEAL, a sperm-specific phospholipase, triggers maize haploid induction.Nature. 2017; 542: 105-109Crossref PubMed Scopus (207) Google Scholar, we transiently expressed our DFP cassette in maturing grains of bread wheat (Triticumaestivum L.), rice (Oryza sativa L.), and barley (Hordeum vulgare L.). Immature grains of 20–30 DAP (days after pollination) were used, cultured after bombardment, and observed under confocal microscopy. As shown in Supplemental Figure 5, expression of the fluorescent genes was observed in embryos and aleurones for all three crops with the expected red color in endosperms and green color in embryos. Thus, the HID marker we developed for maize should be applicable as an essential component of DH in other important cereal crops. In summary, we have developed an approach to create a haploid inducer with robust HID marker for a DH breeding system in maize and, possibly, for other important cereals as well. The CRISPR/Cas9-mediated editing of ZmMTL (ZmPLA1) enabled a maternal haploid inducer to be created. The embryo- and endosperm-specific DFP marker was effective for maternal haploid identification for both mature seeds and young embryos. In addition, the system should be robust in an open pollination environment. It also offers the potential for other important cereals such as wheat, rice, and barley, for which there is no efficient HID selection method. With its broad application, this system has great potential for DH breeding and yield increase in major cereal crops. This work was supported by grants from the National Key Program on Transgenic Research (no. 2016ZX08009-001), Beijing Municipal Science and Technology (Major Program D171100007717001), National Science Foundation of China (no. 31771808), the Chinese Academy of Agricultural Sciences (Y2017XM03), the China National Major Research and Development Program (2016YFD0101803) and National Engineering Laboratory for Crop Molecular Breeding.