A parthenogenesis gene of apomict origin elicits embryo formation from unfertilized eggs in a sexual plant

Significance

The molecular mechanisms controlling apomixis, a mode of asexual reproduction in plants leading to clonal seed formation, are largely unknown. In Pennisetum squamulatum, apomixis segregates as a single dominant locus, the apospory-specific genomic region (ASGR). The ASGR contains multiple copies of the PsASGR-BABY BOOM-like (PsASGR-BBML) gene, a member of the BBM-like subgroup of APETALA 2 transcription factors. Expression of a PsASGR-BBML transgene in sexual tetraploid pearl millet promoted both parthenogenesis (embryo formation without fertilization) and the production of haploid offspring. This study presents the first demonstration, to our knowledge, of function for a gene cloned from a naturally occurring apomictic plant that encodes a key component controlling parthenogenesis.

Abstract

Apomixis is a naturally occurring mode of asexual reproduction in flowering plants that results in seed formation without the involvement of meiosis or fertilization of the egg. Seeds formed on an apomictic plant contain offspring genetically identical to the maternal plant. Apomixis has significant potential for preserving hybrid vigor from one generation to the next in highly productive crop plant genotypes. Apomictic Pennisetum/Cenchrus species, members of the Poaceae (grass) family, reproduce by apospory. Apospory is characterized by apomeiosis, the formation of unreduced embryo sacs derived from nucellar cells of the ovary and, by parthenogenesis, the development of the unreduced egg into an embryo without fertilization. In Pennisetum squamulatum (L.) R.Br., apospory segregates as a single dominant locus, the apospory-specific genomic region (ASGR). In this study, we demonstrate that the PsASGR-BABY BOOM-like (PsASGR-BBML) gene is expressed in egg cells before fertilization and can induce parthenogenesis and the production of haploid offspring in transgenic sexual pearl millet. A reduction of PsASGR-BBML expression in apomictic F₁ RNAi transgenic plants results in fewer visible parthenogenetic embryos and a reduction of embryo cell number compared with controls. Our results endorse a key role for PsASGR-BBML in parthenogenesis and a newly discovered role for a member of the BBM-like clade of APETALA 2 transcription factors. Induction of parthenogenesis by PsASGR-BBML will be valuable for installing parthenogenesis to synthesize apomixis in crops and will have further application for haploid induction to rapidly obtain homozygous lines for breeding.

Apomixis is a naturally occurring process of asexual reproduction that results in offspring that are genetically identical to the mother plant. In nature, apomixis can be achieved through multiple developmental pathways, suggesting different molecular mechanisms have evolved to promote apomixis. Apomictic pathways have been categorized as adventitious embryony, a sporophytic type of apomixis, or diplospory and apospory, two gametophytic forms of apomixis. In gametophytic apomixis, a chromosomally unreduced embryo sac develops from the megaspore mother cell (diplospory) or from a nearby nucellar cell (apospory) in a process termed apomeiosis. Parthenogenesis, the development of the unreduced, unfertilized egg into an embryo, constitutes the second step of the apomictic process (1, 2).

In Pennisetum squamulatum, apomixis is transmitted by a physically large, hemizygous, nonrecombining chromosomal region, the apospory-specific genomic region (ASGR) (3). Multiple copies of the PsASGR-BABY BOOM-like (PsASGR-BBML) gene reside within the ASGR (4) and are postulated as strong candidate genes for the apomictic function of parthenogenesis based on linkage to the ASGR, strong conservation of ASGR-BBML sequences between apomictic Pennisetum/Cenchrus species (5, 6), a loss of the CcASGR-BBML genes in a Cenchrus ciliaris ASGR recombinant plant that has lost the ability to undergo parthenogenesis (7), and similarity of the ASGR-BBML genes to BABY BOOM (BBM) genes of Arabidopsis and Brassica (8).

The first BBM gene was identified as a transcript that was induced in microspore cultures of Brassica napus (BnBBM) undergoing somatic embryogenesis (8). BBM genes are part of a large gene family, the APETALA 2/ETHYLENE RESPONSE FACTOR (AP2/ERF) DNA-binding domain family. The AP2/ERF DNA-binding domain family has been identified in a wide group of plants, including mosses, algae, gymnosperms, and angiosperms. The AP2/ERF gene family is divided into ERF-like, which has a single AP2 domain, and AP2-like, which contains two AP2 domains (repeat 1 and 2) that are similar to each other and separated by a linker region. Although the AP2 domains are highly conserved, the N-terminal and C-terminal sequences of AP2 proteins are more distinct while still containing specific motifs. The AP2-like clade is divided into eudicotAP2 and AINTEGUMENTA (ANT) lineages. The ANT lineage is divided between basalANT and eudicotANT (euANT) lineages that contain specific motifs euANT1–4 (9). The euANT lineage contains the PLETHORA-like (PLT), AINTEGUMENTA-like, AINTEGUMENTA-like1, AINTEGUMENTA-like5, and BBM-like subgroups. The BBM, BBM-like, and ASGR-BBM–like proteins share a conserved bbm-1 domain not identified in other members of the euANT lineage (10). Proteins within the euANT subgroups play critical roles in meristem maintenance, cell proliferation, organ initiation and growth, somatic embryogenesis, embryo differentiation, and root formation (11). Ectopic expression of the BBM gene in Arabidopsis led to the formation of somatic embryos on seedlings (8), and deletion of the bbm-1 domain eliminated the ability of transgenic plants to induce somatic embryogenesis on cotyledons in transformed Arabidopsis (10). The inability to recover homozygous bbm/plt mutants from Arabidopsis and the early arrest of ∼25% of embryos from bbm/plt crosses indicated a redundant function in early embryogenesis for these genes (12).

In the present study, we report ASGR-BBML promoter-GUS activity at the cellular level in tetraploid sexual embryo sacs and developing embryos of pearl millet and discover that transformation of tetraploid sexual pearl millet with an ASGR-BBML transgene, expressed under its native promoter and terminator, is able to promote parthenogenesis, embryo development without fertilization, in three independent transgenic lines. Offspring from one ASGR-BBML transgenic line was able to form diploid/dihaploid seedlings, which in turn can also produce haploid embryos based on embryo flow cytometry. Our results define a previously unidentified role in development for an AP2-like transcription factor, that of parthenogenesis or the initiation of embryo development from an unfertilized egg cell. ASGR-BBML will find application for the synthesis of apomixis in crop plants and for haploid induction to rapidly obtain homozygous lines for breeding.

Results

Expression of ASGR-BBML Genes.

To date, three highly conserved genomic duplications of PsASGR-BBML have been identified from ASGR-linked BACs p203, p207 (5), and p208. The p208 PsASGR-BBML sequence is identical to the p207 PsASGR-BBML2 sequence (EU559277.1) except for the number of AT repeats (11 vs. 17) found in intron 1. Promoter sequences of the three PsASGR-BBML sequences are also highly conserved. The conservation of transcribed sequence leaves in question which of the PsASGR-BBML genomic regions are expressed. Two ASGR-linked copies of ASGR-BBML are present in apomictic C. ciliaris (5). CcASGR-BBM-like1 (EU559278.1) is transcribed and 99.7% identical to the PsASGR-BBML across the ORF, whereas CcASGR-BBM-like2 (EU559279.1), also transcribed, contains two nonsense mutations. Expression of ASGR-BBML has been observed by RT-PCR in apomictic P. squamulatum (PI 319196), C. ciliaris (B12-9), and Pennisetum glaucum backcross (BC) 7 (06-63) and BC₈ (06-A-58) lines (13) in unfertilized ovaries starting 1 d before anthesis and assayed up to 2 d after anthesis, at which time ovaries without endosperm development begin to senesce; in pollinated ovaries at day of anthesis through early seed development; in anthers 1 d before anthesis; and in roots. PsASGR-BBML is also expressed in embryogenic callus derived from apomictic P. glaucum BC₈ (06-A-58). Leaf tissue from P. squamulatum and C. ciliaris does not show ASGR-BBML expression; however, expression in leaf was seen in apomictic P. glaucum BC₇ (06-63) and BC₈ (06-A-58) lines. The PsASGR-BBML transcript encodes a 545-amino-acid protein derived from the splicing of eight exons, a 70-bp 5′ UTR, and multiple 3′ UTRs, with lengths ranging from 30 to 258 bp. The PsASGR-BBML gene contains two AP2 DNA-binding domains and thus is predicted to function as a transcription factor.

Reduction of PsASGR-BBML in Apomictic RNAi Lines.

We first attempted to ascertain the role of PsASGR-BBML in apomictic development by evaluating apomictic F₁ transgenic lines carrying an RNAi construct containing the 3′ end of the PsASGR-BBML gene. Based on Southern blot analysis, the region chosen for the RNAi construct did not hybridize to pearl millet genomic DNA or to additional fragments other than those expected in P. squamulatum genomic DNA. As direct transformation and regeneration of apomictic P. squamulatum was not possible, an alternative strategy and screening protocol for generating apomictic F₁ plants with reduced expression of PsASGR-BBML using the construct RNAi-BBM-3p was developed. Tetraploid sexual pearl millet lines (IA4X) were generated, via a biolistic particle delivery system, which contained the RNAi construct. Eight independent RNAi lines were crossed with pollen from P. squamulatum to produce segregating F₁ plants. Seventy-five plants with genotypes ASGR/RNAi, ASGR/–, –/RNAi, and ^−/− from six independent lines were kept from an initial screen of 190 F₁ seedlings. These plants/lines were screened for RNAi transgene expression in leaf tissue. Twenty-five plants from five independent lines were chosen for detection of PsASGR-BBML expression in ovaries with genotypes ASGR/RNAi (14 plants), ASGR/– (five plants), and –/RNAi (six plants). Three of the 14 ASGR/RNAi plants assayed, each derived from a different line, showed a significant reduction in PsASGR-BBML gene expression based on semiquantitative analysis of PsASGR-BBML expression at day of pollination (SI Appendix, Table S1 and Fig. S1) as compared against the control genotype ASGR/– plant. The three PsASGR-BBML reduced-expression F₁ plants contained the same percentage of ovules with aposporous embryo sac formation as the control plant (SI Appendix, Table S1). The plants were pollinated with red tetraploid P. glaucum (Red-IA4X) pollen, and offspring were determined to be derived through apomixis based on lack of red pigmentation of the midrib, uniform phenotypes, and genotyping for the ASGR. Red-IA4X plants are sexual tetraploid lines containing a dominant Rp1 allele, which confers a dark red pigmentation in the midrib and sheath of leaves (14). However, histological observation of ovaries determined that the number of aposporous embryo sacs showing parthenogenetic embryo development and the number of parthenogenic embryos at or past the globular embryo stage 2 d after anthesis without pollination were significantly reduced in the PsASGR-BBML reduced-expression lines (SI Appendix, Table S1 and Fig. S2). The identified correlation between reduced PsASGR-BBML expression and number of embryo sacs showing parthenogenesis and the diminished development of those parthenogenic embryos suggested a potential role for PsASGR-BBML in unfertilized embryo development.

Expression of PsASGR-BBML at the Cellular Level in Sexual Ovaries Based on PsASGR-BBMLpromoter-GUS Lines.

IA4X plants containing the PsASGR-BBMLpromoter-GUS construct were generated, via a biolistic particle delivery system, to evaluate the expression of PsASGR-BBML at the cellular level. Sexual T₁ offspring from five independent lines were initially evaluated. To prevent unwanted fertilization, heads were bagged before stigma emergence, and stigmas were mechanically removed from the individual florets before pollen shed. Four sexual T₁ individuals within each line containing the PsASGR-BBMLpromoter-GUS construct were evaluated. Control plants were sexual, transgene-null siblings from two lines. GUS signal was detected in the egg/synergid complex in all offspring inheriting the PsASGR-BBMLpromoter-GUS construct in three of the five lines, whereas no GUS activity was identified in the control offspring. Two lines showed very low or no GUS activity.

A more stringent study using one line was undertaken to address the possibility that pollination/fertilization triggered the GUS signal in the egg/synergid complex. Additional PsASGR-BBMLpromoter-GUS plants were germinated, genotyped for the PsASGR-BBMLpromoter-GUS insert, and isolated to a single greenhouse. Heads were continually removed before pollen shed to keep the greenhouse free of millet pollen. Approximately 60 florets were collected from four plants 1 d before anthesis and emasculated, and the florets were placed under in vitro conditions. Another 15–20 ovaries were collected and placed in GUS staining solution for the day before anthesis sampling. For the following 3 d, 15–20 emasculated florets were removed from the in vitro conditions, and the ovaries were dissected and placed in GUS staining solution. As shown in Fig. 1 for unfertilized sexual embryo sacs on the day of anthesis, the GUS signal could be observed within the egg cell, with a weaker GUS signal observed in the synergids (Fig. 1 A–C). The synergid signal may be caused either by a lower expression of the PsASGR-BBML gene in the synergid cells or by leakage of the GUS signal from the egg cell. No GUS staining could be visualized in the central cell or antipodal cells of the sexual embryo sac (Fig. 1 A–C) or in the surrounding ovary tissues. No embryo or endosperm development was identified in any emasculated, in vitro-cultured ovaries up to 2 d past anthesis. Florets that were pollinated and placed in the same in vitro conditions as emasculated florets displayed both endosperm and embryo development. GUS activity was identified in cells of developing embryos characterized up to 3 d after fertilization (Fig. 1D) but not in developing endosperm.

Fig. 1.

PsASGR-BBML expression in sexual embryo sacs. Ovaries from three sexual offspring derived from a T₀ PsASGR-BBMLpromoter-GUS line are shown (A–D). The embryo sacs with antipodals have been outlined. Picture Insets are magnified regions of the egg/synergid/central cell region. The Upper Inset in C is the next focal plane of the ovary to show the central cell. Two asterisks indicate the polar nuclei within the central cell. Arrows indicate synergids. GUS expression is detected in the egg cell of unfertilized sexual embryo sacs on the day of anthesis. GUS signal is not detected in the central cell or antipodal cells of the mature sexual embryo sac. GUS staining is detected in cells of the developing embryo (EM) 3 d after fertilization but not in developing endosperm (D). No other staining in ovary tissue is identified. (Scale bar, 50 µm.)

Parthenogenesis in Transgenic Sexual Pearl Millet Lines Carrying the PsASGR-BBML Transgene Under Control of Its Endogenous Promoter.

Nine independent transgenic lines (18 plants total) containing the transgene gPsASGR-BBML were generated from IA4X via a biolistic particle delivery system. Four lines, consisting of six plants, were not analyzed due to lack of flowering or demise of the plant. Approximately 50 ovaries of 12 individual plants derived from five independent lines were examined 2 d after anthesis using a cleared-pistil technique and observation under differential interference contrast (DIC) or phase contrast optics for parthenogenesis (SI Appendix, Fig. S3). Fertilization was prevented by bagging heads before stigma exsertion and the removal of stigmas/styles before anthesis. Structurally mature embryo sacs containing the egg/synergid apparatus, polar nuclei, and antipodals, which indicate complete sexual embryo sac development, could be identified in all lines (Fig. 2A and SI Appendix, Fig. S3). Three individuals from independent lines (g3f, g11a, and g52) exhibited parthenogenesis (Fig. 2 B and C) based on the presence of embryo-like structures at the micropylar end of the sexual embryo sac and a lack of fertilization to create those embryos based on the persistence of polar nuclei in the central cell and an absence of any endosperm development. Endosperm formation can be readily visualized in fertilized embryo sacs in ovaries at the same developmental stage (SI Appendix, Fig. S3F). All three lines demonstrated endosperm formation at day 2, when pollination was not prevented by removal of stigmas. A minimum of three heads and 100 ovaries from all heads were analyzed for g3f, g11a, and g52 plants along with offspring from an untransformed IA4X line. The percentages of structurally mature sexual embryo sacs (Fig. 2A) and embryo sacs containing parthenogenetic embryos at 2 d after anthesis for lines g11a, g52, and g3f and untransformed offspring are summarized in Table 1. Structurally mature sexual embryo sacs ranged from 61% to 78% in the various lines. Parthenogenesis ranged from 35% to 36% in lines g11a, g52, and g3f, whereas the untransformed offspring showed no parthenogenesis. Expression of the gPsASGR-BBML transgene was verified by RT-PCR with RNA extracted from open-pollinated ovaries 2 d after anthesis for lines g52 and g3f (SI Appendix, Fig. S3). To rule out potential ploidy changes induced by tissue culture selection and regeneration, the three T₀ gPsASGR-BBML lines were subjected to flow cytometry analysis (Fig. 3A). All three plants were tetraploid.

Fig. 2.

Parthenogenetic embryo development in ovaries of sexual tetraploid pearl millet containing the gPsASGR-BBML transgene. Images are from unpollinated ovaries collected and fixed 2 d after anthesis, cleared with methyl salicylate, and visualized using a Zeiss LSM 710 Confocal Microscope. Picture Insets are magnified regions of the egg or embryo/synergid/polar nuclei region. Arrows indicate the polar nuclei within the central cell. (A) A control ovary with a structurally mature embryo sac without fertilization derived from an untransformed tissue culture line. No embryo development is seen. Parthenogenesis in unfertilized ovaries is clearly seen in sexual transgenic line g3f carrying the gPsASGR-BBML transgene based on the appearance of an embryo-like structure at the micropylar end of the embryo sac, polar nuclei, and antipodal cells (B and C). (D) Offspring from line g52 #308 carrying the gPsASGR-BBML transgene and also showing parthenogenesis. Antipodal cells have been cropped from the picture. (Scale bar, 50 µm.)

Table 1.

Summary of visual determination of parthenogenesis in cleared ovaries 2 d after anthesis in untransformed T₀ and T₁ lines

Plant designation (number of plants or heads analyzed)	Genotype of plants ORF +/−	Number of ovaries* with parthenogenic embryos, distinct polar nuclei, no endosperm	Number of ovaries* with parthenogenic embryos, no distinct polar nuclei, no endosperm	Number of ovaries* without parthenogenic embryo development	Number of ovaries without structurally mature sexual embryo sacs	Percent ovaries with structurally mature sexual embryo sacs	Percent ovaries* displaying parthenogenesis
Untransformed (9)	–	0	0	392	94	78	0
g11a (4 heads)	+	29	0	53	53	61	35
g3f (5 heads)	+	59	0	110	81	68	35
g52 (3 heads)	+	27	0	49	33	70	36
g52 offspring (21)	–	0	0	791	229	78	0
g52 offspring (9)	+	90	15	268	109	77	28
g3f offspring (28)	–	0	0	951	392	71	0
g3f offspring (23)	+	122	23	903	530	66	14

^*Ovaries with structurally mature sexual embryo sacs.

Fig. 3.

Flow cytometry analysis to determine genome size of T₀ plants and offspring. Examples of genome size analysis using a BD-Accuri flow cytometer of T₀ plants and g3f offspring (T₁) are shown. (A) Peak analysis of sorghum and T₀ line g11a leaf tissue. (B) Peak analysis of sorghum and g3f offspring 108. (C) Peak analysis of sorghum and g3f offspring 101. (D) Peak analysis of g3f offspring 105 and 107. S2 and S4 designate sorghum 2n/2x/2c and 2n/2x/4c peaks, respectively. H2 and H4 designate T₁ diploid/dihaploid offspring (C and D) with 2n/2x/2c and 2n/2x/4c peaks, respectively. T2 and T4 designate tetraploid T₀ pearl millet (A) or tetraploid T₁ offspring (B and D) with 2n/4x/2c and 2n/4x/4c peaks, respectively.

Due to low germination rates and a low seed set for lines g11a and g52, embryo rescue was used on the developing seed 10–15 d after pollination and on the nongerminating mature seed to recover offspring from the three lines. Pollination with Red-IA4X pollen over multiple heads and days was used to help compensate for potential pollen sterility of transgenic lines g11a and g52. Line g3f was also pollinated with Red-IA4X pollen, however pollen fertility of line g3f was much higher than the other two lines and self-fertilization of g3f lines also occurred. Plant g11a set a total of nine seeds, of which two offspring survived to greenhouse planting. Plant g52 set 97 seeds, of which 31 offspring survived to greenhouse planting. Plant g3f set hundreds of seeds, a mix of both self- and cross-pollination, of which 194 were randomly selected and 107 survived to greenhouse planting.

All offspring were analyzed for inheritance of a 3,694-bp amplicon that covers the gPsASGR-BBML ORF starting 5 bp downstream from the start codon and amplifying into the 3′ UTR (ORF amplicon). The two g11a offspring did not inherit the transgene and were not further analyzed. All offspring from g52 showed red pigmentation of the midrib and thus were derived from the fertilization of g52 sexual embryo sacs with Red-IA4X pollen. Nine offspring carried at least one copy of the ORF amplicon. The g3f offspring were a mix of both green and red pigmentation of the midrib, indicating both self- and cross-pollination of the ovaries and/or the production of haploid offspring for this line. Twenty-six g3f offspring carried at least one copy of the ORF amplicon.

All offspring from line g52 were assayed for parthenogenesis (Table 1, averaging all offspring; SI Appendix, Table S2, data for each individual; Fig. 2D and SI Appendix, Fig. S3 B and D). The percentage of structurally mature sexual embryo sacs for g52 offspring with or without inheriting the ORF amplicon averaged 77% and 78%, respectively. Only offspring inheriting the ORF amplicon showed parthenogenesis 2 d after anthesis. No embryo formation was identified in 791 structurally mature sexual embryo sacs from 21 g52 offspring not inheriting the ORF amplicon. Ninety total sexual embryo sacs displayed embryo development and polar nuclei from the nine g52 offspring inheriting the ORF amplicon (Table 1). The percentage of ovaries showing parthenogenesis ranged from 12% to 53% in the different g52 offspring (SI Appendix, Table S2).

A subset of offspring generated from line g3f were assayed for parthenogenesis (Table 1, averaging all offspring; SI Appendix, Table S3, data for each individual; SI Appendix, Fig. S3 A and C). The percentage of structurally mature sexual embryo sacs for g3f offspring with or without inheriting the ORF amplicon averaged 66% and 71%, respectively. No embryo formation was identified in a total of 951 structurally mature sexual embryo sacs from 28 g3f offspring not inheriting the ORF amplicon. Three of the 26 offspring inheriting the ORF amplicon could not be assayed for parthenogenesis, as they did not flower. Of the remaining 23 offspring inheriting the ORF amplicon, 19 displayed parthenogenesis, whereas 4 did not. The percentage of structurally mature embryo sacs in which parthenogenesis was observed ranged from 1% to 52% for the different g3f offspring. Additional ovaries over the initial screen of ∼50 ovaries were analyzed from the four lines, with a total of 79, 118, 82, and 77 structurally mature sexual embryo sacs screened for plants 123, 144, 159, and 183, respectively. Plants 123, 144, and 159 were assayed for transgene expression using RT-PCR analysis (SI Appendix, Fig. S5). In all three offspring, gPsASGR-BBML expression was detected in unpollinated ovaries at the day of anthesis. Two transgene-specific amplicons covering the gPsASGR-BBML cDNA transgene from the 5′ UTR through the 3′ UTR were sequenced from plants 123, 144, and 159 along with plant 105, which displayed parthenogenesis. All sequences were identical to PsASGR-BBML cDNA sequences derived from BC₇ (06-63) and BC₈ (06-A-58) apomictic plants.

As g3f offspring were a mix of both red and green pigmentation of the midrib, determination of the ploidy level of the green phenotypes was required (SI Appendix, Table S4). Six offspring were diploid/dihaploid in genome size based on flow cytometric analysis using sorghum as the genome size reference (Fig. 3 B and C). The diploid/dihaploid offspring were further confirmed by mixing predicted diploid/dihaploid and tetraploid offspring together to generate three peaks (Fig. 3D). All six dihaploid offspring carry the ORF amplicon, and of the four that flowered, all displayed parthenogenesis.

Embryo Ploidy of T₁ Seed.

Embryos dissected from mature seeds of untransformed IA4X, g3f offspring 104 (tetraploid with ORF amplicon/parthenogenesis) and 105 (diploid/dihaploid with ORF amplicon/parthenogenesis), and g3f offspring 325 (tetraploid with ORF amplicon/parthenogenesis) were analyzed in pools of five embryos using flow cytometry. No reduction of ploidy level was identified in nine embryo pools (45 total embryos) when untransformed IA4X embryos were analyzed. Diploid/dihaploid peaks were identified in three out of four embryo pools (20 total embryos) from embryos collected from the tetraploid 104 seed. Haploid peaks were identified in three out of six embryo pools (30 embryos total) from embryos collected from the diploid/dihaploid 105 seed (SI Appendix, Fig. S6 A–C). Diploid/dihaploid peaks were identified in three out of nine embryo pools (45 total embryos) from embryos collected from the tetraploid 325 seed.

Discussion

We initially attempted to analyze the cellular expression and function of the PsASGR-BBML gene in apomictic plants using RNA in situ hybridization and by creating an RNAi apomictic line where expression of the PsASGR-BBML was completely knocked down. Unfortunately, signal from RNA in situ hybridization of ovaries 1 d before natural anthesis was too weak to reliably ascertain cellular localization for this low-abundance transcript that we had shown by RT-PCR to be expressed at this time point. We could detect in situ hybridization signals for PsASGR-BBML in developing apomictic embryos at the globular stage. The lack of an F₁ aposporous transgenic line showing complete knockdown of the PsASGR-BBML in the RNAi experiment also precluded us from determining the role of PsASGR-BBML in apomixis. However, expression of PsASGR-BBML in ovaries before embryo development based on RT-PCR analysis and the significant reduction in number of ovaries showing parthenogenetic embryo development along with the degree of embryo development observed at 2 d postanthesis in the F₁ PsASGR-BBML reduced-expression lines suggested that additional work on the PsASGR-BBML gene was warranted, as some role in embryo development seemed likely.

We therefore generated tetraploid sexual transgenic pearl millet lines containing the PsASGR-BBMLpromoter-GUS construct to test for cellular expression of PsASGR-BBML. This experiment allowed us to determine that expression from the PsASGR-BBML promoter is restricted to the egg cell and possibly synergids of unfertilized sexual embryo sacs. Expression in the egg cell before fertilization would be a requirement of a parthenogenesis gene. Expression of the PsASGR-BBML promoter-GUS signal continues within cells of the developing embryo and confirms RT-PCR results that show expression of PsASGR-BBML in the developing seed from an apomict up to 5 d after pollination.

Knowing the ∼2.1-kb PsASGR-BBML promoter drove expression of the GUS gene in unfertilized egg cells, we next tested whether the PsASGR-BBML gene regulated by this endogenous promoter could induce parthenogenesis in sexual pearl millet plants expressing PsASGR-BBML. We identified three independent IA4X lines carrying the PsASGR-BBML transgene that clearly show parthenogenic embryo development in unfertilized ovaries 2 d after anthesis in sexual tetraploid pearl millet, plants that do not normally display this trait. Of the three original lines displaying parthenogenesis, two produced viable offspring inheriting the transgene. From lines g52 and g3f, 32% and 55% of rescued embryos survived to greenhouse planting and inherited the transgene ORF (29% and 24% transmission, respectively). The low level of embryo rescue/seedling survival precludes the use of Mendelian genetics to identify inheritance patterns of the transgene(s). However, the generation of offspring with the dominant Rp1 allele and the ORF amplicon demonstrates that the transgene can be transmitted through female meiosis. Although the three original transgenic plants displayed ∼35% parthenogenesis in unfertilized embryo sacs when assayed by ovule clearing 2 d after natural pollen release, parthenogenesis in offspring with the ORF amplicon ranged from 0% to ∼50%. There are many factors that could be implicated for the incomplete penetrance of parthenogenesis within the original T₀ lines and T₁ offspring of lines g3f and g52. Variation of transgene expression from independent lines produced using the same transgene construct has been widely reported and summarized (15). Factors affecting transgene expression levels include transgene copy number, the complexity of integration patterns, and transgene integration sites. In addition, transgene segregation within the sexual embryo sacs of both T₀ and T₁ lines; transgene modifications, such as methylation status, that could alter the expression level and/or timing of expression of the transgene; and/or unknown genetic factors interacting with PsASGR-BBML that enhance the protein’s ability to promote parthenogenesis would be variable in both the T₀ and the T₁ plants given that the starting transformation material was the heterozygous IA4X line. Transcription of the transgene was identified in the occasional offspring not demonstrating parthenogenesis. If penetrance of parthenogenesis was very low (<1.0%), the number of ovaries screened would not have been of sufficient number to detect parthenogenesis. Quantifying levels of transgene expression to the amount of parthenogenesis identified in the various lines is difficult due to variation in the percentage of structurally mature sexual embryo sacs found within each plant and the inability to verify that RNA is being extracted from ovaries that are at the same developmental stage. The generation of homozygous lines containing a single copy of the gPsASGR-BBML transgene or the expression of the gPsASGR-BBML transgene using egg-specific promoters with different expression levels would be advantageous to help address these questions.

We recovered six diploid/dihaploid plants from 170 seedlings of line g3f. No diploid/dihaploid seedlings were recovered from the original g52 T₀ line; however, embryo pools of g52 offspring 325 were found to produce diploid/dihaploid embryos. Several factors could be responsible for the low recovery of diploid/dihaploid seedlings from ovaries showing parthenogenesis 2 d after anthesis without fertilization in these plants. When the embryo develops precociously in the absence of fertilization, it likely impedes fertilization of the central cell and endosperm formation, thus leading to nonviable seed development. Most lines with the ORF amplicon showed some level of reduced mature seed set and produced both plump and shriveled seeds from the various heads. Health and morphology of the g3f diploid/dihaploid offspring varied greatly, with two plants failing to flower. However, ploidy level does not seem to be critical for PsASGR-BBML transgene-induced parthenogenesis, as the four chromosomally reduced diploid/dihaploid offspring derived from the g3f line that did flower showed a similar range of parthenogenesis levels as the unreduced tetraploid offspring also derived from the same line. In addition, using embryo flow cytometry, we were able to detect haploid embryo formation in seed set by g3f diploid/dihaploid offspring 105. Although the majority of natural apomicts are polyploid, natural and experimentally derived apomictic diploids/dihaploids from several species have been identified. Polyploid apomictic plants can produce diploid/dihaploid offspring either through the genetic recombination of the apomeiosis and parthenogenesis loci (16) or through the parthenogenetic development of a reduced egg carrying an apomixis locus (17).

Phylogenetically, the ASGR-BBML proteins cluster within a small group of other BBM-like proteins derived from monocot lineages including Setaria italica, Panicum virgatum, and Oryza sativa, but not with BBM-like proteins from Zea mays or Sorghum bicolor, which have genomes evolutionarily less diverged from Pennisetum than O. sativa (SI Appendix, Figs. S7 and S8). The function of the genes within this small group, except for PsASGR-BBML, is unknown. Expression of rice BBM1 is similar to that of PsASGR-BBML, except its expression in egg cells has not been tested to our knowledge (SI Appendix, Fig. S9). It will be interesting to discover if other members of this clade can promote parthenogenesis if regulated by the PsASGR-BBM promoter used in this experiment or if an uncharacterized domain in the ASGR-BBML protein is required to promote parthenogenesis. Although members of the BBM-like clade of AP2 transcription factors have noted roles in somatic embryogenesis and cell proliferation, our results have uncovered a role for PsASGR-BBML in parthenogenesis. This newly discovered role could have a major impact on the ability to genetically engineer apomixis into crop species and could be used as an alternative method for haploid induction to rapidly obtain homozygous lines for breeding.

Materials and Methods

P. squamulatum (PS26; PI 319196) and C. ciliaris (B12-9) are obligate apomict plants used in genetic studies of apomixis (3, 18). P. glaucum BC₇ (06-63) (13), P. glaucum BC₈ (06-A-58) (13), and P. glaucum BC₈-line 63 are facultative apomictic plants derived from double cross hybrids (19) and BCing with sexual tetraploid P. glaucum (IA4X). IA4X is a hybrid between an African and Indian line that was doubled with colchicine. Red-IA4X is derived from IA4X crossed with a tetraploid plant containing the Rp1 dominant marker (14).

Other methods such as DNA/RNA isolation, PCR primers and reactions (SI Appendix, Table S5), RT-PCR, RACE protocol for cDNA ends, transformation constructs and production of transgenic lines, sequence analysis of the PsASGR-BBML transgene, flow cytometry, chromosomal root tip counts, X-Gluc staining for β-Glucuronidase activity, semiquantitative expression and histological analysis of PsASGR-BBML RNAi lines, and phylogenetic analysis can be found in SI Appendix, SI Materials and Methods.