* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Download Polyploidy Enhances F Pollen Sterility Loci
Public health genomics wikipedia , lookup
Pathogenomics wikipedia , lookup
Essential gene wikipedia , lookup
Therapeutic gene modulation wikipedia , lookup
Site-specific recombinase technology wikipedia , lookup
Long non-coding RNA wikipedia , lookup
X-inactivation wikipedia , lookup
Genetically modified organism containment and escape wikipedia , lookup
Human–animal hybrid wikipedia , lookup
History of genetic engineering wikipedia , lookup
Polycomb Group Proteins and Cancer wikipedia , lookup
Quantitative trait locus wikipedia , lookup
Genome evolution wikipedia , lookup
Gene expression programming wikipedia , lookup
Nutriepigenomics wikipedia , lookup
Designer baby wikipedia , lookup
Biology and consumer behaviour wikipedia , lookup
Ridge (biology) wikipedia , lookup
Artificial gene synthesis wikipedia , lookup
Genomic imprinting wikipedia , lookup
Minimal genome wikipedia , lookup
Genome (book) wikipedia , lookup
Epigenetics of human development wikipedia , lookup
Hybrid (biology) wikipedia , lookup
Polyploidy Enhances F1 Pollen Sterility Loci Interactions That Increase Meiosis Abnormalities and Pollen Sterility in Autotetraploid Rice1[OPEN] Jinwen Wu 2, Muhammad Qasim Shahid 2, Lin Chen, Zhixiong Chen, Lan Wang, Xiangdong Liu*, and Yonggen Lu* State Key Laboratory for Conservation and Utilization of Subtropical Agro-Bioresources, South China Agricultural University, Guangzhou 510642, China ORCID IDs: 0000-0002-6767-4355 (M.Q.S.); 0000-0003-1568-1745 (X.L.). Intersubspecific autotetraploid rice (Oryza sativa ssp. indica 3 japonica) hybrids have greater biological and yield potentials than diploid rice. However, the low fertility of intersubspecific autotetraploid hybrids, which is largely caused by high pollen abortion rates, limits their commercial utility. To decipher the cytological and molecular mechanisms underlying allelic interactions in autotetraploid rice, we developed an autotetraploid rice hybrid that was heterozygous (SiSj) at F1 pollen sterility loci (Sa, Sb, and Sc) using near-isogenic lines. Cytological studies showed that the autotetraploid had higher percentages (.30%) of abnormal chromosome behavior and aberrant meiocytes (.50%) during meiosis than did the diploid rice hybrid control. Analysis of gene expression profiles revealed 1,888 genes that were differentially expressed between the autotetraploid and diploid hybrid lines at the meiotic stage, among which 889 and 999 were up- and down-regulated, respectively. Of the 999 down-regulated genes, 940 were associated with the combined effect of polyploidy and pollen sterility loci interactions (IPE). Gene Ontology enrichment analysis identified a prominent functional gene class consisting of seven genes related to photosystem I (Gene Ontology 0009522). Moreover, 55 meiosis-related or meiosis stage-specific genes were associated with IPE in autotetraploid rice, including Os02g0497500, which encodes a DNA repair-recombination protein, and Os02g0490000, which encodes a component of the ubiquitin-proteasome pathway. These results suggest that polyploidy enhances epistatic interactions between alleles of pollen sterility loci, thereby altering the expression profiles of important meiosis-related or meiosis stage-specific genes and resulting in high pollen sterility. Autotetraploid rice (Oryza sativa) is a useful germplasm developed from diploid rice through colchicine-mediated chromosome doubling. Intersubspecific autotetraploid rice (ssp. indica 3 ssp. japonica) hybrids have biological advantages over diploid hybrids, such as greater adaptability and yield potential (Tu et al., 2007; Shahid et al., 2011, 2012; Wu et al., 2013). However, because autotetraploid 1 This work was supported by The National Natural Science Foundation of China (grant nos. 31270352 and 31210103023 to X.L.), the Guangdong Provincial Science and Technique Project (grant no. 2014A030304055 to X.L.), and the Guangdong Provincial Key Platform of University and Major Research Project (Natural Science) Characteristic Innovation Project (grant no. Yue–JK2014 [65] to X.L.). 2 These authors contributed equally to the article. * Address correspondence to [email protected] and yglu@scau. edu.cn. The author responsible for distribution of materials integral to the findings presented in this article in accordance with the policy described in the Instructions for Authors (www.plantphysiol.org) is: Xiangdong Liu ([email protected]). J.W. performed all of the experiments, analyzed the data, and wrote the article; J.W. and M.Q.S. revised the article; M.Q.S. analyzed the data and wrote the article; L.C. performed the quantitative realtime PCR; Z.C. and L.W. contributed to the data analysis; X.L. and Y.L. conceived the project and revised the article. [OPEN] Articles can be viewed without a subscription. www.plantphysiol.org/cgi/doi/10.1104/pp.15.00791 2700 rice hybrids have low seed set, largely because of high pollen abortion rates, it is challenging to produce hybrids with commercially favorable characteristics (He et al., 2011a, 2011b). The ability to develop high-yielding intersubspecific autotetraploid rice hybrids depends on a detailed understanding of the cytological and molecular mechanisms underlying pollen abortion in these lines (He et al., 2011b). Autotetraploid rice hybrids contain four sets of homologous chromosomes that can undergo abnormal pairing during meiosis. Therefore, abnormal chromosome behavior was considered to be an important cause of pollen abortion in autotetraploid rice hybrids (Luan et al., 2007, 2009; He et al., 2011a, 2011b). However, we found that interactions between hybrid pollen sterility loci, including Sa, Sb, and Sc, contributed greatly to low pollen fertility in autotetraploid rice hybrids (Zhao et al., 2006; He et al., 2011b). Intersubspecific autotetraploid hybrids carry two alleles at each locus, an ssp. indica (Si) allele and a ssp. japonica (Sj) allele. Lines that are heterozygous (SiSj) at each pollen sterility locus produce partially sterile pollen, whereas those that are homozygous (SiSi or SjSj) produce normal pollen (He et al., 2011b). Allelic interactions between Sa, Sb, and Sc sterility loci also account for much of the sterility in diploid intersubspecific hybrids (Zhang and Lu, 1989, 1993; Zhang et al., 1994, 2006, Shahid et al., 2013a). Plant PhysiologyÒ, December 2015, Vol. 169, pp. 2700–2717, www.plantphysiol.org Ó 2015 American Society of Plant Biologists. All Rights Reserved. Downloaded from on June 14, 2017 - Published by www.plantphysiol.org Copyright © 2015 American Society of Plant Biologists. All rights reserved. Allelic Interaction and Gene Expression in Polyploid However, it was unclear how polyploidy affected allelic interactions between pollen sterility loci during pollen mother cell (PMC) meiosis in intersubspecific autotetraploid rice hybrids. Microarray and RNA sequencing (RNA-Seq)-based transcriptome profiling are helpful tools for characterizing molecular aspects of male and female gametophyte development in plants, including differentially expressed genes and stage-specific gene expression. The complexity of gene expression during anther and ovary development has been revealed using microarrays or RNA sequencing in diploid rice (Tang et al., 2010; Aya et al., 2011; Deveshwar et al., 2011; Jin et al., 2013; Pan et al., 2014), maize (Zea mays; DukowicSchulze et al., 2014), wheat (Triticum aestivum; Crismani et al., 2006), and Arabidopsis (Arabidopsis thaliana; Honys and Twell, 2003, 2004; Yang et al., 2011; Zhao et al., 2014). Microarrays have been used to identify differentially expressed genes in autotetraploid plants. For instance, microarray analysis was used to detect numerous subtle differences in gene expression between diploid (2x) and tetraploid (4x) Rangpur lime (Citrus limonia; Allario et al., 2011). Our laboratory carried out microarray analysis to assess genetic variation between autotetraploid (Taichung65-4x) and diploid (Taichung65) rice during pollen development. We identified a total of 1,251 differentially expressed genes, some of which are associated with meiosis and pollen development (Wu et al., 2014). Moreover, RNA-seq-based transcriptome profiling revealed that polyploidy is strongly associated with genome-wide disruption of gene expression and ultimately, resulted in aberrant phenotypes in allopolyploid rice hybrids (Xu et al., 2014). In this study, we used transcriptome profiling and cytological observations to examine the effect of polyploidy on three pollen sterility loci, namely Sa, Sb, and Sc, which resulted in high levels of pollen sterility in intersubspecific autotetraploid rice hybrids. Three factors are known to cause pollen sterility in autotetraploid rice hybrids: interactions between Sa, Sb, and Sc loci; polyploidy (tetraploidy); and the combined effect of pollen sterility loci and polyploidy. We sought to distinguish the effect of each factor on pollen sterility and gene expression profiles using two near-isogenic lines of diploid and autotetraploid rice and their F1 hybrids that are heterozygous at each of these loci (SaiSaj, SbiSbj, and SciScj). We examined chromosome behavior during meiosis and the process of microsporogenesis and microgametogenesis in diploid and autotetraploid hybrids. Furthermore, we identified genes associated with meiosis that were differentially expressed between autotetraploid and diploid rice hybrids, which is a key stage for Sa, Sb, and Sc loci expression and epistatic interactions. We found that autotetraploid hybrids had a distinctly higher percentage of cytological abnormalities than diploid rice hybrids and that polyploidy amplified epistatic interaction between F1 pollen sterility loci, leading to abnormalities in gene expression profiles and high levels of pollen sterility in intersubspecific autotetraploid rice hybrids. RESULTS Genotypes of Parental Lines and Hybrids at Sa, Sb, and Sc Loci and Their Effect on Pollen Fertility We determined the genotypes of the diploid and autotetraploid rice hybrids at the Sa, Sb, and Sc pollen sterility loci using closely linked molecular markers (Supplemental Figs. S1 and S2; Table I). Because the parental lines each have different alleles at Sa, Sb, and Sc, the hybrids showed allelic interactions at the three pollen sterility loci. We also determined the pollen fertility and seed setting of the two hybrids (Table I). Diploid hybrid (Taichung65 3 E245) F1 plants harbored Si and Sj alleles at these three loci and had low pollen fertility (30.51%) and seed setting (20.85%). However, autotetraploid hybrid (Taichung65-4x 3 E245-4x) F1 plants, which had epistatic interactions between alleles at the three pollen sterility loci, displayed very low pollen fertility (12.17%) and seed setting (3.56%; Table I). These results indicate that interactions between pollen sterility loci have a more dramatic effect on pollen fertility in the autotetraploid hybrid than in the diploid hybrid. Microsporogenesis and Microgametogenesis in Diploid and Autotetraploid Rice Hybrids We then used whole-mount eosin B-staining confocal laser-scanning microscopy (WE-CLSM) to observe microsporogenesis and microgametogenesis in diploid and autotetraploid rice hybrids. During microsporogenesis in the diploid hybrids, meiosis appeared similar to that in the parental lines of the diploid hybrids Table I. Frequency of abnormal pollen and seed setting in the diploid and autotetraploid hybrids and Sa, Sb, and Sc loci genotypes DF1 indicates the diploid hybrid (Taichung65 3 E245) that is heterozygous at the Sa, Sb, and Sc pollen sterility loci. TF1 indicates the autotetraploid hybrid (Taichung65-4x 3 E245-4x) that is heterozygous at the Sa, Sb, and Sc pollen sterility loci. Abnormal Pollen at the Mature Stage Name DF1 TF1 Genotype of Sa, Sb, and Sc Loci Sai Saj Sbi Sbj Sci Scj Sai Sai Saj Saj Sbi Sbi Sbj Sbj Sci Sci Scj Scj Seed Setting Pollen Fertility % 6 SD 20.85 6 0.23 30.51 6 1.63 3.56 6 0.02 12.17 6 1.03 Sample Size Asynchronous Pollen Abnormally Shaped Pollen Small Pollen 3,098 2,877 27.23 6 1.85 27.63 6 1.95 % 6 SD 32.93 6 3.27 56.52 6 1.03 1.88 6 0.46 4.22 6 0.16 Plant Physiol. Vol. 169, 2015 2701 Downloaded from on June 14, 2017 - Published by www.plantphysiol.org Copyright © 2015 American Society of Plant Biologists. All rights reserved. Wu et al. Figure 1. Cytological observation of pollen development in diploid and autotetraploid rice hybrids. A to D, Normal pollen development in diploid rice. A, Dyad stage. B, Tetrad stage. C, Late-stage microspore. D, Bicellular pollen stage. E to L, Abnormal pollen development in diploid hybrids. E, Abnormal dyad stage. F and G, Abnormal tetrad stage (arrows indicate abnormal tetrad cells). H to J, Examples of abnormal early-stage microspores. K, Abnormal midstage bicellular pollen (arrow indicates small pollen). L, Mature pollen stage (arrows indicate small pollen). M to X, Abnormal pollen development in autotetraploid rice hybrids. M, Abnormal dyad stage. N to P, Abnormal tetrad stage. Q, Abnormal early-stage microspore (arrows). R, Middle-stage abnormal microspore (arrow). S, Abnormal late-stage microspore (arrows indicate multiple apertures). T, Early-stage bicellular pollen (arrow indicates small pollen). U to X, Latestage bicellular pollen (arrows indicate small pollen). Bars = 40 mm. (Fig. 1, A and B). The diploid hybrids had few meiotic abnormalities, such as dyad and tetrad degeneration (Fig. 1, E–H). By contrast, the autotetraploid hybrids exhibited high percentages of degenerated dyads and tetrads (Fig. 1, M–O). After meiosis, microgametogenesis can be divided into seven stages, including early microspore stage, middle microspore stage, late microspore stage (Fig. 1C), early bicellular pollen stage, middle bicellular pollen stage (Fig. 1D), late bicellular pollen stage, and mature pollen stage. Some abnormal microspores and bicellular pollen were found in diploid hybrids during microgametogenesis (Fig. 1, I–L). However, nearly 40% of the microspores and bicellular pollen were abnormal in the autotetraploid hybrid during microgametogenesis. For instance, some microspores were surrounded by a thick callose wall (i.e. the callose remained after completion of meiosis; Fig. 1P), some exhibited cytoplasmic shrinkage (degeneration) at the early microspore stage (Fig. 1Q), some failed to form a large vacuole (Fig. 1R), and a high percentage of microspores had multiple apertures during the middle and late microspore stages (Fig. 1S; Table II). More than one-third (36.91%) of the microspores had two to four apertures at the late microspore stage, and most of the asynchronous development occurred during the bicellular pollen stage (Table II). The nuclei of some microspores underwent synchronous divisions during the late microspore stage, which resulted in 2702 Plant Physiol. Vol. 169, 2015 Downloaded from on June 14, 2017 - Published by www.plantphysiol.org Copyright © 2015 American Society of Plant Biologists. All rights reserved. Allelic Interaction and Gene Expression in Polyploid Table II. Frequency of pollen with different numbers of aperture DF1 indicates the diploid hybrid (Taichung65 3 E245) that is heterozygous at the Sa, Sb, and Sc pollen sterility loci. TF1 indicates the autotetraploid hybrid (Taichung65-4x 3 E245-4x) that is heterozygous at the Sa, Sb, and Sc pollen sterility loci. Name DF1 TF1 No. Single Aperture 3,797 3,670 94.04 6 0.14 63.09 6 2.17 Two Aperture %6 5.96 6 0.44 33.35 6 2.99 abnormal small pollen (Fig. 1, T–X). By the mature pollen stage, 88.37% of the autotetraploid rice pollen was abnormal (Table I). Meiosis in Diploid and Autotetraploid Hybrids Similar to normal diploid rice and consistent with descriptions presented in Wu et al., 2014, meiosis in the PMCs of the diploid rice hybrid could be divided into the following nine developmental stages: prophase I, consisting of leptotene, zygotene (Fig. 2A), pachytene (Fig. 2, B and C), diplotene (Fig. 2D), and diakinesis (Fig. 2E); metaphase I (Fig. 2F); anaphase I (Fig. 2G); telophase I (Fig. 2H); prophase II (Fig. 2I); metaphase II (Fig. 2J); anaphase II (Fig. 2K); telophase II (Fig. 2L); and tetrad stage (Fig. 2M). A low percentage of abnormal behavior, such as chromosome lagging (Fig. 2N) in metaphase II, straggling in anaphase II (Fig. 2O), and Three Aperture Four Aperture 0.00 3.08 6 1.03 0.00 0.48 6 0.16 SD asynchronous cell division (Fig. 2P), was observed in PMCs of the diploid hybrid. We also detected these nine developmental stages in the meiotic PMCs of autotetraploid rice hybrids (Fig. 3, A–N; leptotene is not shown). However, we observed many abnormalities, including chromosome lagging at metaphase I and metaphase II, chromosome straggling at anaphase I and anaphase II, micronuclei formation at telophase I and telophase II, and disordered spindles at metaphase II, in the autotetraploid hybrid (Table III). The main kinds of abnormalities were as follows. Prophase I: At diakinesis, we observed quadrivalents, trivalents, and bivalents simultaneously in the PMC (Fig. 3, D–F). The main chromosome configurations in the autotetraploid hybrid are shown in Table IV. Forty-eight univalent chromosomes were present in approximately 5% of PMCs at diakinesis (Fig. 3Q). Metaphase I: Chromosome lagging was frequent at this stage, with an average of 38.38% of cells having one Figure 2. Chromosome behavior during PMC meiosis in the diploid rice hybrid. A, Zygotene. B and C, Pachytene. D, Diplotene. E, Diakinesis. F, Metaphase I. G, Anaphase I. H, Telophase I. I, Prophase II. J, Metaphase II. K, Anaphase II. L, Telophase II. M, The tetrad stage. N, Abnormal metaphase I (arrow indicates lagging chromosome). O, Abnormal metaphase II (arrow indicates lagging chromosome). P, Abnormal anaphase II (asynchronous cell division). Bars = 10 mm. Plant Physiol. Vol. 169, 2015 2703 Downloaded from on June 14, 2017 - Published by www.plantphysiol.org Copyright © 2015 American Society of Plant Biologists. All rights reserved. Wu et al. Figure 3. Chromosome behavior during PMC meiosis in the autotetraploid rice hybrid. A, Zygotene. B, Pachytene. C, Diplotene. D to F, Diakinesis. G, Metaphase I. H, Anaphase I. I, Telophase I. J, Dyad. K, Prophase II. L, Metaphase II. M, Anaphase II. N, The tetrad stage. O, Abnormal metaphase I (arrow indicates lagging chromosome). P, Abnormal anaphase I (arrow indicates straggling chromosome). Q, Pollen mother cell containing 48 univalent chromosomes at diakinesis. R, Distorted spindle. S, Abnormal anaphase II showing asynchronous cell division. T, The triad stage. Bars = 10 mm. or more instance of chromosome lagging (Fig. 3O; Table III). Anaphase I to telophase I: We found that 30.49% of cells had one or more instance of chromosome straggling (i.e. some chromosomes remained at the center of the cell, whereas others had already reached the poles; Fig. 3P). Furthermore, 10.83% of cells contained micronuclei at telophase I (Table III). Metaphase II: We found that 54.81% of cells at this stage exhibited abnormal chromosome behavior, spindles, or division, including chromosome lagging (4.31%), V-shaped spindles (12.51%), T-shaped spindles (11.28%), misshapen spindles (1.38%), and asynchronous division (25.32%; Table V). The two spindles of the dyad were arranged in different orientations in PMCs at anaphase II (Fig. 3R). Anaphase II: We found that 11.63% of chromosomes exhibited straggling in anaphase II (Table VI). Moreover, we observed dyad asynchrony, in which one dyad was at metaphase II and the other was at anaphase II (Fig. 3S). Telophase II to the tetrad stage: We observed triads consisting of one large cell and two small cells (Fig. 3T). Furthermore, the tetrads contained different numbers of nucleoli, including one (14.65%), two (18.79%), three (4.14%), and four double nucleoli (2.55%; Table VII). Gene Expression Profile Changes and Gene Ontology Analysis in Autotetraploid and Diploid Rice Hybrids during Meiosis Our cytological analysis showed that abnormalities began to appear in PMC meiosis at the stage of dyad formation in autotetraploid rice hybrids. Therefore, we next performed transcriptome profiling using Affymetrix GeneChips to investigate the impact of polyploidy and interactions between pollen sterility loci on global patterns of gene expression in autotetraploid hybrids during PMC meiosis. Microarray data revealed significant correlations (correlation coefficient . 0.95; three 2704 Plant Physiol. Vol. 169, 2015 Downloaded from on June 14, 2017 - Published by www.plantphysiol.org Copyright © 2015 American Society of Plant Biologists. All rights reserved. Allelic Interaction and Gene Expression in Polyploid Table III. Frequency of cells exhibiting abnormal chromosome behavior during meiosis I in diploid and autotetraploid rice hybrids DF1 indicates the diploid hybrid (Taichung65 3 E245) that is heterozygous at the Sa, Sb, and Sc pollen sterility loci. TF1 indicates the autotetraploid hybrid (Taichung65-4x 3 E245-4x) that is heterozygous at the Sa, Sb, and Sc pollen sterility loci. Name Metaphase I No. Normal Lagging Anaphase I Multipolar Spindle No. Normal Straggling 93.52 63.41 % 6.48 30.49 % DF1 TF1 444 297 97.97 59.93 2.03 38.38 0.00 1.68 216 328 replicates) in the expression profiles of diploid and autotetraploid hybrids (Supplemental Fig. S3; Supplemental Table S1). Significant Analysis of Microarray software identified 1,888 genes that were differentially expressed (2-fold at P value , 0.05) between diploid and autotetraploid hybrids during meiosis, with 889 genes being up-regulated in the autotetraploid relative to the diploid hybrid and 999 genes being down-regulated (Fig. 4; Supplemental Table S2). We then categorized both the up- and down-regulated genes using Cluster 3.0 software and found that these genes could be clustered into subgroups based on expression levels (Supplemental Fig. S4). Next, we conducted Gene Ontology (GO) analysis to annotate the genes that were differentially expressed between the diploid and autotetraploid rice hybrids during meiosis. We found that many of the differentially expressed genes were associated with multiorganism processes and transcriptional regulatory functions, and there was significant variation between putatively up-regulated and down-regulated genes related to these two categories. Notably, we found that genes associated with the extracellular matrix and metallochaperone were down-regulated in the autotetraploid cells (Fig. 5). Among the 889 genes that were differentially upregulated in autotetraploid rice during early meiosis, three prominent functional gene groups, including response to stress (GO: 0006950; 41 genes), cell wall macromolecule catabolic process (GO: 0016998; 6 genes), and jasmonic acid stimulus (GO: 0009753; 6 genes), were common in the biological process category (Fig. 5; Supplemental Fig. S5). In the cellular component category, which contained a total of 180 genes, we identified one prominent functional gene group, which mainly involved cytoplasmic membrane-bound vesicles Telophase I Bridge No. Normal Micronucleus % 0.00 6.10 246 257 97.96 89.17 2.43 10.83 (GO: 0016023; Supplemental Fig. S6). In the molecular function category, we detected two prominent functional gene groups involved in Ser-type endopeptidase inhibitor activity (GO: 0004867; 7 genes) and transcription factor activity (GO: 0003700; 27 genes; Supplemental Fig. S7). Among the 999 genes that were differentially downregulated in autotetraploid rice during meiosis, we detected two prominent functional gene groups in the biological process category that were associated with starch metabolic processes (GO: 0005982) and fatty acid biosynthetic processes (GO: 0006633; Fig. 5; Supplemental Fig. S8) and consisted of 12 and 13 genes, respectively. In the cellular component category, we identified two prominent functional gene classes, namely chloroplast (GO: 0009507), consisting of 36 genes, and PSI (GO: 0009522), consisting of 7 genes (Supplemental Fig. S9). In the molecular function category, we detected one prominent functional gene class associated with catalytic activity (GO: 0003824), and 212 genes were included in this category (Supplemental Fig. S10). Furthermore, we then used the David data tool (National Institute of Allergy and Infectious Diseases, National Institutes of Health) to identify genes that were preferentially down-regulated in autotetraploid hybrids and found that 999 differentially down-regulated genes were mainly categorized into 17 clusters according to k score, Similarity, and GO enrichment analyses. Among these clusters of genes, clusters 1 and 2 consisted of photosynthetic-related genes, mainly associated with PSI and PSII. These results are consistent with the AgriGo data (Supplemental Fig. S9). In addition, clusters 10 and 11 were comprised of genes related to transcription regulation and DNA binding. In cluster 15, genes were mainly associated with transcription factors and pollen development (Supplemental Table S3). Table IV. Meiotic chromosome configurations in diploid and autotetraploid hybrids DF1 indicates the diploid hybrid (Taichung65 3 E245) that is heterozygous at the Sa, Sb, and Sc pollen sterility loci. TF1 indicates the autotetraploid hybrid (Taichung65-4x 3 E245-4x) that is heterozygous at the Sa, Sb, and Sc pollen sterility loci. I to IV indicate univalent, bivalent, trivalent, and quadrivalent chromosomes, respectively. Material Name/No. of Cells DF1 127 167 TF1 168 210 Stage Chromosome Configuration 6 SD Diakinesis Metaphase I (0.74 6 0.38) I + (11.26 6 0.19) II (0.23 6 0.04) I + (11.77 6 0.11) II Diakinesis Metaphase I (0.50 6 0.10) I + (9.24 6 0.35) II + (0.64 6 0.06) III + (6.25 6 0.21) IV (1.33 6 0.11) I + (15.79 6 0.36) II + (0.53 6 0.15) III + (3.74 6 0.19) IV Plant Physiol. Vol. 169, 2015 2705 Downloaded from on June 14, 2017 - Published by www.plantphysiol.org Copyright © 2015 American Society of Plant Biologists. All rights reserved. Wu et al. Table V. Frequency of abnormal chromosome behavior at metaphase II in diploid and autotetraploid rice hybrids DF1 indicates the diploid hybrid (Taichung65 3 E245) that is heterozygous at the Sa, Sb, and Sc pollen sterility loci. TF1 indicates the autotetraploid hybrid (Taichung65-4x 3 E245-4x) that is heterozygous at the Sa, Sb, and Sc pollen sterility loci. Name DF1 TF1 No. Normal 244 312 79.10 45.19 Lag 0.00 4.31 V Spindle T Spindle Chaos Asynchronous Meiocytes 6.47 12.51 % 4.09 11.28 0.50 1.38 9.84 25.32 To confirm the results of our microarray analysis, we validated the expression of nine genes (including four genes belonging to the pollen allergen Ole e I family, two belonging to the MYB transcription factor family, two encoding members of the ubiquitin-proteasome pathway, and one encoding a member of the WRKY transcription factor family) in diploid and autotetraploid hybrids during PMCs meiosis by quantitative reverse transcription (qRT)-PCR. The expression patterns of all of these genes as determined by qRT-PCR were consistent with the microarray data (Fig. 6). Genes Differentially Down-Regulated in Autotetraploid Rice Were Associated with IPE (the Combined Effect of Polyploidy and Pollen Sterility Loci Interactions) It was previously reported that interactions between pollen sterility loci (Sa, Sb, and Sc; IE) caused low pollen fertility in intersubspecific diploid hybrids (Shahid et al., 2013a). The following three possible factors could affect gene expression in autotetraploid rice hybrids: (1) polyploidy (PE), (2) interactions between Sa, Sb, and Sc loci (IE), and (3) the combined effect of the polyploidy and Sa, Sb and Sc loci interactions (IPE; Fig. 7). Reasoning that the down-regulation of important genes may affect pollen development in the autotetraploid hybrid, we evaluated the contribution of PE, IE, and IPE to the expression of the 999 genes that were downregulated in autotetraploid PMCs. To accurately determine the effect of each group, we made the following three types of comparisons. Type I Comparison (PE Factor) Comparison of the transcriptomic profiles of the diploid and autotetraploid parents (i.e. E1-4x versus E1-2x and E245-4x versus E245-2x) was mentioned as PE (1) and PE (2), respectively (Fig. 7, A and B). We previously showed that 125 genes were down-regulated during PMC meiosis in autotetraploid rice (E1-4x) versus diploid rice (E1-2x) and identified one prominent functional gene class with five genes associated with DNAdependent DNA replication (GO: 0006261) in the biological process category. However, we did not identify any significant GO terms in the cellular component category (Wu et al., 2014). In our study, we identified 231 genes that were down-regulated in PMC meiosis of autotetraploid rice (E245-4x) relative to the diploid line (E245-2x) during PMC meiosis (Fig. 7B). AgriGo analysis revealed that no significant GO terms were in the biological process, cellular component, or molecular function category. Type II Comparison (IE Factor) Comparison was made of transcriptomic profiles of diploid parents and their hybrids (i.e. DF1 versus E1 and E245). E245 is a pollen-sterile isogenic line (PSIL) of Taichung65 (E1, Sa j Sa j , Sb j Sb j , and ScjScj) at the Sa, Sb, and Sc loci (i.e. harboring homozygous genotypes, SiSi, at these loci), and therefore, their hybrids have the same genetic background as those of E1 and E245, except for at the Sa, Sb, and Sc loci. Differences in the transcriptomic profiles between the diploid hybrid (DF1) and parental lines (i.e. E1-2x and E245-2x) may be caused by IE. We identified a total of 591 and 735 genes that were up- and down-regulated in DF1 versus E1-2x and DF1 versus E245-2x, respectively (Fig. 7). AgriGo analysis of the 735 down-regulated genes indicated that there were no significant GO terms in the biological process category. One prominent functional gene class with 157 genes encoding cytoplasmic membranebounded vesicles (GO: 0016023) was found in the Table VI. Frequency of cells exhibiting abnormal chromosome behavior at anaphase II in diploid and autotetraploid rice hybrids DF1 indicates the diploid hybrid (Taichung65 3 E245) that is heterozygous at the Sa, Sb, and Sc pollen sterility loci. TF1 indicates the autotetraploid hybrid (Taichung65-4x 3 E245-4x) that is heterozygous at the Sa, Sb, and Sc pollen sterility loci. Materials No. Normal Stragglers V Spindle DF1 TF1 213 246 95.14 50.93 0.00 11.63 2.86 6.81 T Spindle Asynchronous Meiocytes 0.00 6.00 2.00 24.63 % 2706 Plant Physiol. Vol. 169, 2015 Downloaded from on June 14, 2017 - Published by www.plantphysiol.org Copyright © 2015 American Society of Plant Biologists. All rights reserved. Allelic Interaction and Gene Expression in Polyploid Table VII. Frequency of cells exhibiting abnormal chromosome behavior at telophase II and the tetrad stage in diploid and autotetraploid rice hybrids DF1 indicates the diploid hybrid (Taichung65 3 E245) that is heterozygous at the Sa, Sb, and Sc pollen sterility loci. TF1 indicates the autotetraploid hybrid (Taichung65-4x 3 E245-4x) that is heterozygous at the Sa, Sb, and Sc pollen sterility loci. Telophase II Tetrad Materials DF1 TF1 No. Normal Lag Abnormal Tetrad Micronucleus No. Normal (One Nucleolus) 284 258 98.60 71.85 0.00 0.78 % 0.70 21.98 0.70 5.39 250 314 96.80 59.55 cellular component (Supplemental Fig. S11). In the molecular function category, we identified two prominent functional gene classes, including 16 genes related to sequence-specific DNA binding protein (GO: 00043565) and 21 genes encoding transcription factors (GO: 0003700; Supplemental Fig. S12). Type III Comparison (IPE Factor) Comparison was made of transcriptomic profiles between the autotetraploid and corresponding diploid rice hybrids (i.e. TF1 versus DF1). We identified a total of 889 and 999 genes that were up- and down-regulated in TF1 versus DF1, respectively (Fig. 7, B and C). Furthermore, 940 of the down-regulated genes were specifically associated with IPE (Fig. 8; Supplemental Table S4). We performed predicted protein-protein interaction analysis using STRING software and identified interactions among these 940 genes (Supplemental Fig. S13). AgriGo analysis of the 940 down-regulated genes indicated one prominent functional gene class related to PSI (GO: 0009522) that contained 7 genes (i.e. Os03g0592500, Os04g0414700, Os08g0560900, Os12g0189400, Os03g0778100, One Double Nucleoli Two Double Nucleoli Three Double Nucleoli Four Double Nucleoli 3.20 14.65 % 0.00 18.79 0.00 4.14 0.00 2.55 Os06g0320500, and Os07g0435300) in the cellular component category that were not detected in the types I and II comparisons. However, several genes, including Os03g0592500, Os04g0414700, Os12g0189400, Os03g0778100, and Os06g0320500, were up-regulated in type II comparison. Overall, these results indicate that the effect of allelic interactions between the pollen sterility loci (IE) is strikingly different from that of the combined effect of polyploidy and pollen sterility loci interactions (IPE). Two (Os06g0320500 and Os03g0592500) of the seven genes affected the expression of Os05g0496100, which was annotated as encoding the translation initiation factor eukaryotic Initiation Factor-3 (eIF3) subunit. The above results suggest that IPE influences pollen development and the photosynthetic rate in the autotetraploid rice hybrid. We measured and compared the photosynthetic rates of the diploid and autotetraploid rice and found that the photosynthetic rate of autotetraploid hybrids was significantly lower than that of the diploid hybrids (Fig. 9A; Supplemental Table S5). Furthermore, we confirmed the differential gene expression of photosynthetic-related genes (i.e. Os03g0592500, Os04g0414700, Os08g0560900, Figure 4. Differentially expressed genes in diploid and autotetraploid rice hybrids. A, Scatter plot analysis of signal intensities for all normalized genes on the Affymetrix microarray in diploid and autotetraploid rice hybrids. Red and green dots indicate genes that are up- and down-regulated, respectively. DF 1 and TF 1 indicate the diploid (Taichung65 3 E245) and autotetraploid (Taichung65-4x 3 E245-4x) hybrids that are heterozygous at the Sa, Sb, and Sc pollen sterility loci, respectively. B, Total number of differentially expressed genes in diploid and autotetraploid rice hybrids. Genes were categorized as being down-regulated or up-regulated if their transcript abundance decreased or increased, respectively, and the changes were statistically significant (2-fold at P value , 0.05). Plant Physiol. Vol. 169, 2015 2707 Downloaded from on June 14, 2017 - Published by www.plantphysiol.org Copyright © 2015 American Society of Plant Biologists. All rights reserved. Wu et al. Figure 5. GO enrichment analysis of up- and down-regulated genes in diploid and autotetraploid rice hybrids during PMC meiosis. Genes were divided into three categories: biological process, molecular function, and cellular component. The percentage of genes in each category was calculated for every GO term. GO functional classification analysis was performed using a Plant GeneSet Enrichment Analysis Toolkit and AgriGO. Os12g0189400, Os03g0778100, and Os06g0320500) in diploid and autotetraploid hybrids using real-time qRT-PCR analysis (Fig. 9B). detected in the types II and III comparisons, we constructed a Venn diagram showing genes that were identified as being up-regulated in the IE and specifically Analysis of Meiotic-Related Down-Regulated Genes Associated with IPE in Autotetraploid Hybrid Because we used transcriptomic profiling to analyze the impact of polyploidy and interactions between pollen sterility loci on global patterns of gene expression in autotetraploid hybrids during PMC meiosis, we now focused on meiosis-related and meiosis stage-specific differentially down-regulated genes associated with IPE. We compared the expression of these genes in autotetraploid PMCs with microarray data reported for diploid rice anther meiosis stage-specific expression (Fujita et al., 2010; Deveshwar et al., 2011) and meiosisrelated expression (Fujita et al., 2010; Yant et al., 2013; Wright et al., 2015). We identified 54 meiosis stagespecific genes and 1 meiosis-related gene (i.e. Os02g0497500 [OsAffx.12275.1.S1_x_at] that encodes the DNA repair-recombination protein [RAD50]) that were strongly expressed in the anthers of diploid rice hybrids during meiosis but weakly expressed in autotetraploid rice hybrids. All of the genes were putatively down-regulated with at least a 2-fold change in expression and specifically associated with IPE. To compare the differentially expressed genes associated with the meiosis-related or stage-specific expression Figure 6. Confirmation of microarray data (gene expression profiles) by real-time qRT-PCR in diploid and autotetraploid rice hybrids during PMC meiosis. The bars denote the mean expression of nine genes (relative expression patterns). Gene identification numbers from left to right along the x axis are related to pollen allergen Ole e I family (four genes), MYB transcription factor family (two genes), ubiquitin-proteasome pathway (two genes), and WRKY transcription factor family (one gene). The error bars indicate the 6SD of the average of three independent replicates. 2708 Plant Physiol. Vol. 169, 2015 Downloaded from on June 14, 2017 - Published by www.plantphysiol.org Copyright © 2015 American Society of Plant Biologists. All rights reserved. Allelic Interaction and Gene Expression in Polyploid Figure 7. Relationship between differentially expressed genes, pollen sterility loci, and polyploidy. A, Expression levels of differentially expressed genes in four pairwise comparisons: (1) E1-4x versus E1-2x, (2) E245-4x versus E245-2x (polyploidy effect), (3) diploid hybrid (Taichung65 3 E245) that is heterozygous at the Sa, Sb, and Sc pollen sterility loci (DF1) versus E1-2x and E245-2x (pollen sterility loci interactions), and (4) autotetraploid hybrid (Taichung65-4x 3 E245-4x) that is heterozygous at the Sa, Sb, and Sc pollen sterility loci (TF1) versus DF1 (the effect of polyploidy, pollen sterility loci interactions, and the combined effect of polyploidy and pollen sterility loci interactions). B, Effect of allelic interactions between the Sa, Sb, and Sc pollen sterility loci (IE), polyploidy effect (PE), and the combined effect of polyploidy and pollen sterility loci interactions (IPE) on gene expression patterns during PMC meiosis. C, Venn diagram represents the differentially expressed genes in the PMCs meiosis of the parental lines, the diploid hybrid, and the autotetraploid hybrid that were putatively associated with PE (type I), IE (type II), and PE + IE + IPE (type III) factors. down-regulated in IPE (Supplemental Fig. S14). We found that 17 of 55 meiosis-related or stage-specific genes were putatively down-regulated because of IPE. These down-regulated genes displayed very low levels of expression, and the remaining 38 genes were upregulated in response to IE but down-regulated in response to IPE. Of the 17 IPE-specific genes, 6 genes were related to metabolism, encoding proteins, such as GDSL-like lipase and polygalacturonase (Supplemental Table S6). Os12g0578700 and Os10g0570700 encode a zinc finger protein and ribosome-inactivating protein, respectively. Two genes were associated with the ubiquitin-proteasome pathway (Os02g0490000: U-box domain-containing protein) and phenylpropanoid pathway (Os04g0310700: AMP-binding domain containing protein; Supplemental Figs. S15 and S16). Two genes (Os08g0496800 and Os09g0329000) encode a BURP domain-containing protein. Os07g0144500 and Os05g0496100 were associated with PLA IIIA/Patatinlike protein7 and the translation initiation factor eIF3 subunit, respectively. Three genes were annotated as encoding an expressed protein. Predicted Protein-Protein Interaction Analysis of MeiosisRelated or Stage-Specific Down-Regulated Genes Associated with IPE The products of 23 of the 55 genes found to be downregulated in response to IPE were predicted to undergo protein-protein interactions (Supplemental Fig. S13, A–E). We found that the DNA repair-recombination protein (RAD50) encoded by the key meiosis-related gene, Os02g0497500, interacted with BTB domain-containing Plant Physiol. Vol. 169, 2015 2709 Downloaded from on June 14, 2017 - Published by www.plantphysiol.org Copyright © 2015 American Society of Plant Biologists. All rights reserved. Wu et al. Figure 8. Venn diagram representing the overlap between downregulated genes associated with polyploidy effect (PE), effect of allelic interactions between the Sa, Sb, and Sc pollen sterility loci (IE), and the combined effect of polyploidy and pollen sterility loci interactions (IPE) factors (http://bioinfogp.cnb.csic.es/tools/venny/; Oliveros, 2007). protein (Os01g0893400) and the calmodulin-binding domains of retrotransposon protein (Os03g0423850; Fig. 10). Os05g0496100 encodes the translation initiation factor eIF3 subunit and interacted with genes encoding chlorophyll A-B binding proteins (i.e. Os01g0600900, Os08g0435900, Os07g0558400, Os06g0320500, Os07g0562700, and Os03g0592500; Fig. 10). Os08g0440800 was annotated as an aldehyde dehydrogenase and interacted with genes encoding protoporphyrin-IX (Os01g0286600), 6-phosphofructokinase (Os06g0151900), and Fru-bisphosphate aldolase isozyme (Os01g0905800; Fig. 10). Os01g0924933 encodes a transferase family protein that interacts with glycerol-3-P acyltransferase, 3-ketoacyl-CoA synthase, and cytochrome P450 (Fig. 10). Other important interactions were associated with DNA binding protein, ubiquitin fusion degradation protein, and HOTHEAD precursor. All of these interactions suggest that IPE has a significant impact on the expression of 23 meiosisrelated or stage-specific genes and thereby, PMC meiosis and pollen development. Coexpression and cis-Motif Analysis of Meiosis-Related or Stage-Specific Down-Regulated Genes Associated with IPE Coexpression analysis is a useful method to determine interactions between meiotic-related and stage-specific down-regulated genes associated with IPE. Coexpression analysis revealed that 8 of the 55 meiosisrelated genes were coexpressed with genes involved in metabolism (Fig. 11A). Of these eight putatively coexpressed genes, five encoded proteins related to metabolism, including an aspartyl protease (Os04g0595000), dihydroflavonol-4-reductase (Os01g0127500), stilbene synthase (Os10g0484800), polygalacturonase (Os03g0216800), and chalcone and stilbene synthase (Os07g0411300). Os07g0556800 and Os10g0570700 were annotated as encoding ribosome-inactivating proteins, and Os09g0329000 encodes a BURP domain containing protein. In addition, we detected another 11 genes that interact with the 8 specifically coexpressed genes (Supplemental Table S4A). Among these 11 genes, Os12g0242700 was annotated as encoding a 3-oxoacylreductase and chloroplast precursor; Os04g0573100 was annotated as encoding a HOTHEAD precursor, and five genes were annotated as encoding a zinc finger C-x8-C-x5-C-x3-H-type family protein, auxin response factor, aquaporin protein, polygalacturonase, and transporter. Furthermore, IPE prominently influenced 3 of the 11 genes that interacted with 8 specifically coexpressed genes and 12 of the 55 meiosis-related or stagespecific down-regulated genes (Fig. 11B). To investigate whether certain promoter motifs were involved in the transcriptional changes resulting from IPE, we surveyed the sequence 1,500 bp upstream of each meiosis-related and meiosis stage-specific gene for cis-motifs using the Plant CARE database. According to the microarray profiles of 55 meiosis-related or meiosis stage-specific genes, 19 types of cis-motifs were detected in the gene sequences. Interestingly, the Skn-1_motif, which is required for endosperm development (Lescot et al., 2002), was the main type of regulatory element present in the promoter regions of these 55 genes (Supplemental Table S7). DISCUSSION PSILs and Their Hybrids Revealed Epistatic Interactions between Sa, Sb, and Sc Loci in Autotetraploid Rice Hybrids Partial pollen sterility and embryo sac abortions affect seed setting of intersubspecific rice hybrids and limit the commercial production of hybrids (Zhang and Lu, 1989; Zhang et al., 2006; Ji et al., 2012; Shahid et al., 2013a). Six loci, including Sa, Sb, Sc, Sd, Se, and Sf, have been found to regulate intersubspecific rice hybrid pollen sterility (Zhang and Lu, 1989, 1993; Zhang et al., 1994). At all of these loci, the ssp. indica and ssp. japonica rice varieties carried the SiSi and SjSj alleles, respectively. Each locus interaction (SiSj) caused partial pollen sterility in intersubspecific hybrids (Zhang and Lu, 1989). Transmission electron microscopy revealed that Sa allelic interactions caused microspore abnormalities at the middle microspore stage and finally, produced empty abortive pollen; Sb allelic interactions resulted in asynchronous development of microspores at the middle microspore stage, producing stainable abortive 2710 Plant Physiol. Vol. 169, 2015 Downloaded from on June 14, 2017 - Published by www.plantphysiol.org Copyright © 2015 American Society of Plant Biologists. All rights reserved. Allelic Interaction and Gene Expression in Polyploid Figure 9. Comparison of photosynthetic data in diploid rice hybrids (DF1s) and autotetraploid rice hybrids (TF1s). A, Comparison of photosynthetic rates between the DF1s and TF1s. B, Confirmation of the gene expression profiles of photosynthetic-related genes by real-time qRT-PCR in DF1s and TF1s; y axis denotes the mean expression of six genes (relative expression patterns). The error bars indicate the 6SD of the average of three independent replicates. pollen, and Sc allelic interactions mainly led to the nondissolution of the generative cell wall and finally, produced stainable abortive pollen (Zhang et al., 2006). In this study, we used Taichung65 and its PSIL to generate a diploid hybrid with Sa, Sb, and Sc loci interactions. We observed abnormal chromosome behavior (e.g. chromosome lagging at metaphases I and II) in a low percentage of PMCs during meiosis, suggesting that interactions between alleles at the Sa, Sb, and Sc loci (SaiSaj SbiSbj SciScj) may have an effect on the stage of meiosis I in diploid hybrids. Intersubspecific autotetraploid rice hybrids are robust plants with the potential to increase rice yield but suffer from low seed set (He et al., 2011b; Shahid et al., 2013b; Wu et al., 2013). We developed over 1,000 combinations of autotetraploid rice in our laboratory during 2002 to 2010 and found that partial pollen sterility is a major cause of low seed setting in intersubspecific autotetraploid rice hybrids. Furthermore, we found that interactions between different alleles of hybrid pollen sterility loci (Sa, Sb, and Sc) caused low pollen fertility in autotetraploid rice hybrids (Zhao et al., 2006; He et al., 2011b; Wu et al., 2013). It took 4 years to generate the autotetraploid spp. japonica rice, Taichung65-4x, and its PSILs (PSILs-4x) in our laboratory. We self-crossed PSILs-4x for more than 20 generations in our laboratory and confirmed that their genotypes are stable and the same as those of diploid rice at the Sa, Sb, and Sc loci. In this study, we used Taichung65-4x and PSILs-4x to develop an autotetraploid rice hybrid that was heterozygous at Sa, Sb, and Sc loci. More than 30% of PMCs exhibited abnormal chromosome behavior, such as chromosome lagging in metaphase I and metaphase II, chromosome straggling in anaphase I and anaphase II, and micronuclei formation in telophase II and during meiosis in the autotetraploid rice hybrids. These results suggest that interactions at the Sa, Sb, and Sc loci (i.e. SaiSaj SbiSbj Plant Physiol. Vol. 169, 2015 2711 Downloaded from on June 14, 2017 - Published by www.plantphysiol.org Copyright © 2015 American Society of Plant Biologists. All rights reserved. Wu et al. Figure 10. The protein-protein interaction network of meiosis-related and stage-specific genes associated with the IPE factor. The protein-protein interaction subnetwork was constructed using four important meiosis-related and stage-specific genes (red) and other genes that interact with these genes (blue). SciScj) may affect the early stage of prophase I (leptotene) in autotetraploid rice hybrids and that epistatic effects between alleles at these loci were more pronounced in autotetraploid than diploid rice hybrids. The epistatic interactions increased pollen sterility, resulting in much lower fertility in the autotetraploid rice hybrid than in the diploid hybrid. Although numerous studies have used a variety of cytological techniques to examine pollen development in rice, no live cytological observations of microsporocytes and developing microspores have been reported for autotetraploid rice hybrids using WE-CLSM (Feng et al., 2001; Lu et al., 2002; Zhang et al., 2005, 2006). To observe microsporocyte meiosis and microspore development, we used WE-CLSM and found that interactions between three pairs of Si and Sj alleles in autotetraploid hybrids resulted in abnormalities in microsporocytes, such as abnormal dyads and a high percentage of abnormal tetrads. These results were consistent with our observations of aberrant chromosomes in PMCs. Polyploidy Enhanced Epistatic Interactions between Pollen Sterility Loci That Led to Distinct Alterations in Transcriptome Profiles and Resulted in High Pollen Sterility in Autotetraploid Rice Hybrids Hybridization and whole-genome duplication alter specific locus or genome-wide gene expression patterns (transcriptomic shock) in many allopolyploids, although the extent of the impact varies in different species (Hegarty et al., 2006; Flagel et al., 2008; Buggs et al., 2011). Xu et al. (2014) used RNA-seq-based transcriptome profiling to analyze the impact of hybridization and whole-genome duplication in reciprocal hybrids. Xu et al. (2014) developed allopolyploids from rice japonica and indica ssp. and showed genome-wide changes in gene expression, including biased parental expression in the hybrids. Xu et al. (2014) speculated that the emerging property of whole-genome doubling had repercussions that reverberate throughout the transcriptome, ultimately generating altered phenotypes. Here, we focused on the impact of polyploidy on interactions between hybrid pollen sterility loci (Sa, Sb, and Sc) and conducted global transcriptomic profiling of Taichung65, its PSILs, and their reciprocal F1 diploid and tetraploid hybrids. We explored genome-wide alterations in gene expression caused by polyploidy and Sa, Sb, and Sc loci interactions during meiosis, a key stage in pollen development. It used to be technically challenging to isolate meiotic cells at a specific stage for RNA-seq analysis. Bright-field microscopy, 4’,6diamidino-2-phenylindole (DAPI) fluorescence staining, and laser capture of individual cells have made it possible to dissect these cells (Fujita et al., 2010; Tang et al., 2010; Yang et al., 2011). Using DAPI fluorescence staining and WE-CLSM techniques, we were able to obtain PMCs at precise stages of meiosis and identify genes that were differentially expressed during these stages in rice (Feng et al., 2001; Lu et al., 2002; Zhang et al., 2005, 2006; He et al., 2011a, 2011b; Wu et al., 2014). Theoretically, the fact that the PSILs have the same genetic background, except at the Sa, Sb, and Sc loci, eliminates the effect of different parental genomes, thus providing insight into the contribution of IE, PE, and IPE but not hybridization of the whole genome on genome-wide expression patterns. We detected 1,888 genes that were differentially expressed during meiosis, which included 889 upregulated and 999 down-regulated genes. We found that 940 of the genes were down-regulated specifically because of IPE. The 940 IPE genes included many genes associated with pollen development and 7 genes that were putatively associated with PSI and displayed significant down-regulation during PMC meiosis. These seven genes also showed a low level of expression in leaves, which resulted in a reduction in photosynthetic rate in autotetraploid rice versus the diploid hybrid. However, we were unable to detect such a phenomenon in genes associated with IE and PE. By contrast, Fu et al. (1999) found that photosynthetic rate 2712 Plant Physiol. Vol. 169, 2015 Downloaded from on June 14, 2017 - Published by www.plantphysiol.org Copyright © 2015 American Society of Plant Biologists. All rights reserved. Allelic Interaction and Gene Expression in Polyploid Figure 11. The coexpression and protein-protein interaction network of meiosis-related and stagespecific genes associated with the IPE factor. A, The coexpression subnetwork was constructed using eight meiosis-related and stage-specific genes (blue circles and red labels). B, Proteinprotein interaction network of 3 genes (green) that interact with 8 specifically coexpressed genes and 12 of 55 meiosis-related and stage-specific genes (red labels) that were found to be down-regulated and associated with the IPE factor. was significantly higher in autotetraploid rice than in diploid rice. Net photosynthetic rate decreased significantly, especially under severe drought stress conditions, in diploid rice, whereas drought did not have a significant effect on net photosynthetic rate in autotetraploid lines (Yang et al., 2014). The results of our study suggest that polyploidy enhances the interaction between Sa, Sb, and Sc loci, which reduces the photosynthetic rate in autotetraploid rice hybrids compared with the corresponding diploid rice hybrids. Meiosis is a crucial process in plant reproduction. Twenty-eight rice meiotic genes have been characterized to date (Luo et al., 2014), and more than 300 meiosis stage-specific genes have been identified in rice anthers (Fujita et al., 2010; Jin et al., 2013). Here, our main focus was on differentially down-regulated meiosis-related or stage-specific genes associated with the IPE factor. We found that 54 meiosis stage-specific genes were putatively down-regulated during various stages of meiosis. Among the meiosis stage-specific genes, 16 down-regulated genes were specifically related to IPE. Therefore, IPE might be a major reason for the low pollen fertility of autotetraploid rice hybrids. Among the down-regulated genes associated with the IPE factor, we detected two genes, Os05g0496100 and Os02g0490000, related to pollen development. Os05g0496100 regulates the translation of proteins that function in the cell cycle (Pyronnet and Sonenberg, Plant Physiol. Vol. 169, 2015 2713 Downloaded from on June 14, 2017 - Published by www.plantphysiol.org Copyright © 2015 American Society of Plant Biologists. All rights reserved. Wu et al. 2001) and forms the 48S complex through high-affinity binding to eIF4G in mammalian eIF3 (LeFebvre et al., 2006). Protein-protein interaction analysis indicated significant interactions between the translation initiation factor eIF3 subunit (encoded by Os05g0496100) and chlorophyll A-B binding protein encoded by six genes (Os01g0600900, Os08g0435900, Os07g0558400, Os06g0320500, Os07g0562700, and Os03g0592500). This interaction suggests that down-regulation of Os05g0496100 might have a strong effect on the regulation of starch biosynthesis genes in response to levels of photoassimilates (Morita et al., 2015). Here, Os05g0496100 may cause a significant reduction in the photosynthetic rate of the autotetraploid in comparison with the corresponding diploid rice. Os02g0490000 was annotated as a U-box domain-containing protein, which possessed E3 ubiquitin-protein ligase activity and was involved in the ubiquitination pathway. The ubiquitination pathway is known to function in signaling, transcription, DNA repair, cell viability, and endosomal trafficking (Bader and Steller, 2009; Li et al., 2012). Os02g0490000 is required for normal pollen development in kiwifruit (Actinidia deliciosa; Scoccianti et al., 1999). Thus, we speculated that down-regulation of Os02g0490000 may affect pollen development of autotetraploid rice hybrids. Moreover, another important down-regulated gene, Os02g0497500, was found to encode a DNA repairrecombination protein (RAD50; Gherbi et al., 2001). RAD50 plays important roles in meiosis, chromosomal recombination, mitosis, telomere maintenance, and cellular DNA damage responses and forms a protein complex with Mre11 and Xrs2/Nbs1 (Mre11/RAD50/ Xrs2; Gallego and White, 2001; Gherbi et al., 2001; Daoudal-Cotterell et al., 2002; Vannier et al., 2006). Absence of RAD50 stimulates homologous recombination, and disruption of RAD50 leads to male sterility in Arabidopsis (Gallego et al., 2001; Gherbi et al., 2001; Bleuyard et al., 2004). The down-regulation of RAD50 observed in our study may increase homologous recombination and thereby, promote abnormal chromosome pairing, resulting in the formation of quadrivalents and trivalents during meiosis in autotetraploid rice hybrids. All of the above putatively down-regulated genes were associated with anther development and the photosystem. Therefore, down-regulation of their expression would affect PMC meiosis and result in low pollen fertility in the autotetraploid rice hybrid. We have revealed the different expression patterns of some key genes related to pollen development in autotetraploid rice. Such a combined study is important for understanding locus/gene function(s). These results provide a genome-wide view of the complexity of multigene interactions during early meiosis and insight into the intricate molecular mechanism underlying the effect of polyploidy on allelic interactions between Sa, Sb, and Sc loci that result in pollen sterility of autotetraploid hybrids. Future work should aim to develop new autotetraploid rice lines harboring neutral or widely compatible pollen fertility genes, which do no interact with each other, to overcome pollen sterility and develop fertile autotetraploid rice hybrids. MATERIALS AND METHODS Plant Materials Two hybrids, an autotetraploid F1 hybrid (Taichung65-4x 3 E245-4x) and a diploid F1 hybrid (Taichung65 3 E245), that are heterozygous, SiSj, at the Sa, Sb, and Sc loci were used in this study. Taichung65-4x and E245-4x were developed from Taichung65 (E1) and E245 by colchicine-mediated chromosome doubling and self-crossed for more than 20 generations in our laboratory. E245 is a PSIL of Taichung65 (E1), which has the same genetic background as E1, except at the Sa, Sb, and Sc loci. The genotypes of all hybrids are shown in Table I. Diploid F1 hybrids (Taichung65 3 E245) were used as the control for comparisons with autotetraploid F1 hybrids (Taichung65-4x 3 E245-4x). Taichung65 (E1), E245, Taichung65-4x, and E245-4x were also used as controls to evaluate the effect of allelic interactions between the Sa, Sb, and Sc pollen sterility loci (IE), the effect of polyploidy (PE), and the combined effect of polyploidy and pollen sterility loci interactions (IPE) on pollen fertility. We evaluated PE for two groups of parental lines (i.e. Taichung65-4x [E1-4x] versus Taichung65 [E1] and E245-4x versus E245; Fig. 7, A and B). All plants were grown under natural conditions at the experimental farm of South China Agricultural University in 2013, and standard practices were followed. Validation of Parental and Hybrid Genotypes at the Sa, Sb, and Sc Loci Polymorphic pairs of primers were selected from published markers at the Sa, Sb, and Sc loci for genotype confirmation (Shahid et al., 2013a). The Sa gene has been cloned, and Sb and Sc have been fine mapped; therefore, closely linked markers to these genes were used. Two single-nucleotide polymorphism markers (G02-148 and G02-69) at the Sa locus, two markers (A07-55 and A07-130) at the Sb locus, and two Simple Sequence Repeat markers (P24-85.7 and P24-100.7) at the Sc locus were selected (Shahid et al., 2013a). Classification of Chromosome Behavior Spikelets were collected from the shoots of rice (Oryza sativa) plants with 22 to 2 cm between their flag leaf cushion and the second to last leaf cushion and fixed in Carnoy solution (ethanol:acetic acid [3:1, v/v]) for at least 24 h. The samples were washed three times using 90% (v/v) ethanol and then stored in 70% (v/v) ethanol at 4°C. Anthers were removed from the floret using forceps and a dissecting needle and placed in a drop of 1% (w/v) acetocarmine on a glass slide. After 3 to 5 min, the glass slide was covered with a coverslip and examined under a microscope (Motic BA200). Meiotic stages were classified and explained according to the works by He et al. (2011b) and Wu et al. (2014). Characterization of Microsporogenesis and Microgametogenesis WE-CLSM was used to observe microsporogenesis and microgametogenesis in diploid and autotetraploid rice hybrids. Spikelets were collected from the shoots of rice plants with 24 to 20 cm between their flag leaf cushion and the second to last leaf cushion and kept in a petri dish with moist filter paper. Anthers were removed from the floret using forceps and a dissecting needle and placed in a drop of 10 mg L21 eosin B (C20H6N2O9Br2Na2; Mr 624.1; a tissue stain for cell granules and nucleoli) solution (dissolved in 4% [w/v] Suc) on a glass slide. After 10 min, the glass slide was covered with a coverslip and scanned under a Leica SPE Laser-Scanning Confocal Microscope (Leica Microsystems). The excitation wavelength was 543 nm, and emission light was detected between 550 and 630 nm (Shahid et al., 2010; Wu et al., 2014). Pollen fertility was evaluated according to the method by Shahid et al. (2013a). Tissue Collection and RNA Extraction Anthers of the two hybrids and their parental lines were collected during the meiotic stage of pollen development, and DAPI fluorescence staining was used to confirm the developmental stage. After collection, samples were frozen 2714 Plant Physiol. Vol. 169, 2015 Downloaded from on June 14, 2017 - Published by www.plantphysiol.org Copyright © 2015 American Society of Plant Biologists. All rights reserved. Allelic Interaction and Gene Expression in Polyploid immediately in liquid nitrogen and stored at 280°C until used for RNA extraction. All samples were collected in biological triplicate. Total RNA was isolated following the RNA isolation protocol from Invitrogen Trizol Reagent. RNA quality was assessed by formaldehyde agarose gel electrophoresis and quantitated by spectrophotometry (Wu et al., 2014). Affymetrix Gene Chip Microarrays and Data Analysis To identify differences in gene expression between the diploid and autotetraploid F1 hybrids at the meiotic stage, Affymetrix microarrays were used. Two micrograms of total RNA isolated from anthers was used to generate biotin-labeled complementary RNA targets, and then fragmented complementary RNA targets were hybridized to GeneChip arrays. The GeneChip arrays were washed, stained, and scanned according to the Affymetrix gene chip protocol. The quantified data were analyzed using GeneChip Operating software (GCOS 1.4). The scanned images were first assessed by visual inspection and then analyzed to generate raw data files saved as cell intensity values (CEL) files using default settings of GCOS 1.4. An invariant set normalization procedure was used to normalize the different arrays using a DNA chip analyzer (Wu et al., 2014). GO Annotation and Pathway Analysis All microarray data were analyzed using Significant Analysis of Microarray software. Genes with a fold change of $2 or #0.5 were subjected to Student’s t test analysis, and genes with P values of #0.05 were chosen for further analysis. Cluster analysis was performed using Cluster 3.0 software. GO analysis was performed for the functional categorization of differentially expressed genes using the Plant GeneSet Enrichment Analysis Toolkit (http:// structuralbiology.cau.edu.cn/PlantGSEA/) and AgriGO tool (http://bioinfo. cau.edu.cn/agriGO/). Pathway analysis was performed using MapMan software and David analysis tools (http://david.abcc.ncifcrf.gov/home.jsp; Huang et al., 2009a, 2009b). Predicted protein-protein interactions were analyzed using STRING software (http://www.string-db.org/). Annotations for differentially expressed genes were retrieved from the Rice Genome Annotation Project (http://rice.plantbiology.msu.edu/). Gene identifications were converted using the identification converter (http://rapdb.dna.affrc.go.jp/tools/ converter/run) and bioDBnet (http://biodbnet.abcc.ncifcrf.gov/). Coexpression and cis-Motif Analysis of Genes Associated with IPE To determine predicted gene functions and gene regulatory networks of differentially down-regulated genes that were putatively associated with IPE, coexpression analysis was performed using the RiceFREND database (http:// ricefrend.dna.affrc.go.jp/; Stuart et al., 2003; Sato et al., 2013). To determine the potential functional relevance of the down-regulated genes that are involved in meiosis or expressed at specific stages of meiosis, the PlantCARE database (http://bioinformatics.psb.ugent.be/webtools/plantcare/html/) was used to search for transcriptional cis-elements. qRT-PCR To verify the microarray results, 15 candidate genes from the microarray analysis were validated by qRT-PCR using the same RNA samples as used in the microarray analysis. Fifteen gene-specific primer pairs were designed based on the target gene sequences using Primer Premier5.0 and Oligo6.0 software. Primers were designed to obtain a product size of between 80 and 200 bp (Supplemental Table S8). Complementary DNA (cDNA) was synthesized from 1 mg of total RNA using oligo(dT)18 primers and a Transcriptor First Strand cDNA Synthesis Kit (Roche). qRT-PCR was performed using SYBR Green (BioRad) in a LightCycler480 System (Roche). The final reaction volume for qRTPCR was 20 mL, and each reaction contained 2 mL of cDNA, 10 mM forward and reverse primer, and 10 mL (23) Advanced SYBR Green Supermix (Bio-RAD). The following amplification program was used: denaturation at 95°C for 30 s and 40 cycles of amplification (95°C for 5 s followed by 58°C for 20 s). Melting curve analysis was performed from 65°C to 95°C, with 5-s increments of 0.5°C. The rice Ubiquitin gene was used as an internal control. The relative expression levels were calculated as 22(DCt of treatment 2 DCt) of control (Livak and Schmittgen, 2001), where Ct represents threshold cycle. Each PCR, including the control reaction, was performed in triplicate (Wu et al., 2014). Measurement of Photosynthetic Rate An Li-6400XT Portable Photosynthesis System (LI-COR) was used to measure the photosynthetic rate of autotetraploid and diploid rice. All measurements were performed at 26°C (air temperature) with an ambient CO2 concentration of about 350 mmol mol21 and a relative air humidity of about 75%. Each calculation was the mean of 20 replicates (Yang et al., 2014). Supplemental Data The following supplemental materials are available. Supplemental Figure S1. PCR amplification of genomic DNA of diploid and autotetraploid rice hybrids using the P24-85.7 marker. Supplemental Figure S2. PCR amplification of genomic DNA of diploid and autotetraploid rice hybrids using the A07-55 marker. Supplemental Figure S3. Clustering of microarray data of diploid and autotetraploid hybrids during meiosis. Supplemental Figure S4. Expression patterns of differentially expressed genes in diploid and autotetraploid rice hybrids during meiosis. Supplemental Figure S5. GO enrichment analysis of differentially upregulated genes of the biological process category in autotetraploid rice hybrid. Supplemental Figure S6. GO enrichment analysis of differentially upregulated genes of the cellular component category in autotetraploid rice hybrid. Supplemental Figure S7. GO enrichment analysis of differentially upregulated genes of the molecular function category in autotetraploid rice hybrid. Supplemental Figure S8. GO enrichment analysis of putatively downregulated genes of the biological process category in autotetraploid rice hybrid. Supplemental Figure S9. GO enrichment analysis of differentially downregulated genes of the cellular component category in autotetraploid rice hybrid. Supplemental Figure S10. GO enrichment analysis of differentially downregulated genes of the molecular function category in autotetraploid rice hybrid. Supplemental Figure S11. GO enrichment analysis of differentially downregulated genes of the cellular component category associated with Sa, Sb, and Sc loci interaction (IE). Supplemental Figure S12. GO enrichment analysis of differentially downregulated genes of the molecular function category associated with Sa, Sb, and Sc loci interaction (IE). Supplemental Figure S13. Predicted protein-protein interaction network of differentially down-regulated genes (940) associated with IPE (combined effect of polyploidy and pollen sterility loci interaction) in the autotetraploid rice hybrid during meiosis. Supplemental Figure S14. Venn diagram showing differentially expressed up- and down-regulated genes associated with IE and IPE. Supplemental Figure S15. Ubiquitin-proteasome pathway analysis of 55 down-regulated genes. Supplemental Figure S16. Phenylpropanoid pathway analysis of 55 downregulated genes. Supplemental Table S1. The correlation coefficients of microarray data of diploid and autotetraploid rice hybrids during meiosis. Supplemental Table S2. Differentially up- and down-regulated genes in diploid and autotetraploid rice hybrids during meiosis. Supplemental Table S3. Groups of 940 genes classified by the David analysis tool. Supplemental Table S4. Product(s) or function(s) of genes associated with IPE. Plant Physiol. Vol. 169, 2015 2715 Downloaded from on June 14, 2017 - Published by www.plantphysiol.org Copyright © 2015 American Society of Plant Biologists. All rights reserved. Wu et al. Supplemental Table S5. Photosynthetic rate and variance analysis between autotetraploid and diploid rice hybrids. Supplemental Table S6. Product(s) or function(s) of meiosis-related and stage-specific genes associated with IPE. Supplemental Table S7. The number of cis-motifs found upstream of the genomic sequences corresponding to the meiosis-related and stagespecific genes. Supplemental Table S8. List of primers used for qRT-PCR analysis. ACKNOWLEDGMENTS We thank Guiquan Zhang for donating Taichung65 and E245; Hongbin Wang for helpful advice and discussions; and Dr. Meiyang Duan, Dunzhou Mo, Minyi Chen, and Shuhong Yu for technical assistance. Received May 26, 2015; accepted October 27, 2015; published October 28, 2015. LITERATURE CITED Allario T, Brumos J, Colmenero-Flores JM, Tadeo F, Froelicher Y, Talon M, Navarro L, Ollitrault P, Morillon R (2011) Large changes in anatomy and physiology between diploid Rangpur lime (Citrus limonia) and its autotetraploid are not associated with large changes in leaf gene expression. J Exp Bot 62: 2507–2519 Aya K, Suzuki G, Suwabe K, Hobo T, Takahashi H, Shiono K, Yano K, Tsutsumi N, Nakazono M, Nagamura Y, et al (2011) Comprehensive network analysis of anther-expressed genes in rice by the combination of 33 laser microdissection and 143 spatiotemporal microarrays. PLoS One 6: e26162 Bader M, Steller H (2009) Regulation of cell death by the ubiquitinproteasome system. Curr Opin Cell Biol 21: 878–884 Bleuyard JY, Gallego ME, White CI (2004) Meiotic defects in the Arabidopsis rad50 mutant point to conservation of the MRX complex function in early stages of meiotic recombination. Chromosoma 113: 197–203 Buggs RJ, Zhang L, Miles N, Tate JA, Gao L, Wei W, Schnable PS, Barbazuk WB, Soltis PS, Soltis DE (2011) Transcriptomic shock generates evolutionary novelty in a newly formed, natural allopolyploid plant. Curr Biol 21: 551–556 Crismani W, Baumann U, Sutton T, Shirley N, Webster T, Spangenberg G, Langridge P, Able JA (2006) Microarray expression analysis of meiosis and microsporogenesis in hexaploid bread wheat. BMC Genomics 7: 267 Daoudal-Cotterell S, Gallego ME, White CI (2002) The plant Rad50-Mre11 protein complex. FEBS Lett 516: 164–166 Deveshwar P, Bovill WD, Sharma R, Able JA, Kapoor S (2011) Analysis of anther transcriptomes to identify genes contributing to meiosis and male gametophyte development in rice. BMC Plant Biol 11: 78 Dukowic-Schulze S, Sundararajan A, Mudge J, Ramaraj T, Farmer AD, Wang M, Sun Q, Pillardy J, Kianian S, Retzel EF, et al (2014) The transcriptome landscape of early maize meiosis. BMC Plant Biol 14: 118 Feng JH, Lu YG, Liu XD, Xu XB (2001) Pollen development and its stages in rice (Oryza sativa L.). Chinese J Rice Sci 15: 21–28 Flagel L, Udall J, Nettleton D, Wendel J (2008) Duplicate gene expression in allopolyploid Gossypium reveals two temporally distinct phases of expression evolution. BMC Biol 6: 16 Fu Y, Yan H, Li L, Yu Y, Si H, Hu G, Xiao H, Sun Z (1999) Photosynthesis-related characteristics of different ploidy rice plants. Chinese J Rice Sci 13: 157–160 Fujita M, Horiuchi Y, Ueda Y, Mizuta Y, Kubo T, Yano K, Yamaki S, Tsuda K, Nagata T, Niihama M, et al (2010) Rice expression atlas in reproductive development. Plant Cell Physiol 51: 2060–2081 Gallego ME, Jeanneau M, Granier F, Bouchez D, Bechtold N, White CI (2001) Disruption of the Arabidopsis RAD50 gene leads to plant sterility and MMS sensitivity. Plant J 25: 31–41 Gallego ME, White CI (2001) RAD50 function is essential for telomere maintenance in Arabidopsis. Proc Natl Acad Sci USA 98: 1711–1716 Gherbi H, Gallego ME, Jalut N, Lucht JM, Hohn B, White CI (2001) Homologous recombination in planta is stimulated in the absence of Rad50. EMBO Rep 2: 287–291 He JH, Shahid MQ, Chen ZX, Chen XA, Liu XD, Lu YG (2011a) Abnormal PMC microtubule distribution pattern and chromosome behavior resulted in low pollen fertility of an intersubspecific autotetraploid rice hybrid. Plant Syst Evol 291: 257–265 He JH, Shahid MQ, Li YJ, Guo HB, Cheng XA, Liu XD, Lu YG (2011b) Allelic interaction of F1 pollen sterility loci and abnormal chromosome behaviour caused pollen sterility in intersubspecific autotetraploid rice hybrids. J Exp Bot 62: 4433–4445 Hegarty MJ, Barker GL, Wilson ID, Abbott RJ, Edwards KJ, Hiscock SJ (2006) Transcriptome shock after interspecific hybridization in senecio is ameliorated by genome duplication. Curr Biol 16: 1652–1659 Honys D, Twell D (2003) Comparative analysis of the Arabidopsis pollen transcriptome. Plant Physiol 132: 640–652 Honys D, Twell D (2004) Transcriptome analysis of haploid male gametophyte development in Arabidopsis. Genome Biol 5: R85 Huang W, Sherman BT, Lempicki RA (2009a) Bioinformatics enrichment tools: paths toward the comprehensive functional analysis of large gene lists. Nucleic Acids Res 37: 1–13 Huang W, Sherman BT, Lempicki RA (2009b) Systematic and integrative analysis of large gene lists using DAVID bioinformatics resources. Nat Protoc 4: 44–57 Ji Q, Zhang M, Lu J, Wang H, Lin B, Liu Q, Chao Q, Zhang Y, Liu C, Gu M, et al (2012) Molecular basis underlying the S5-dependent reproductive isolation and compatibility of indica/japonica rice hybrids. Plant Physiol 158: 1319–1328 Jin Y, Yang H, Wei Z, Ma H, Ge X (2013) Rice male development under drought stress: phenotypic changes and stage-dependent transcriptomic reprogramming. Mol Plant 6: 1630–1645 LeFebvre AK, Korneeva NL, Trutschl M, Cvek U, Duzan RD, Bradley CA, Hershey JWB, Rhoads RE (2006) Translation initiation factor eIF4G1 binds to eIF3 through the eIF3e subunit. J Biol Chem 281: 22917–22932 Lescot M, Déhais P, Thijs G, Marchal K, Moreau Y, Van de Peer Y, Rouzé P, Rombauts S (2002) PlantCARE, a database of plant cis-acting regulatory elements and a portal to tools for in silico analysis of promoter sequences. Nucleic Acids Res 30: 325–327 Li W, Dai L, Wang GL (2012) PUB13, a U-box/ARM E3 ligase, regulates plant defense, cell death, and flowering time. Plant Signal Behav 7: 898–900 Livak KJ, Schmittgen TD (2001) Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta C(T)) method. Methods 25: 402–408 Lu YG, Feng JH, Liu XD, Xu XB (2002) Ultrastructural studies on the developmental process of pollen and anther in rice (Oryza sativa L.). Chinese J Rice Sci 16: 29–37 Luan L, Tu SB, Long WB, Wang X, Liu YH, Kong FL, He T, Yan WG, Yu MQ (2007) Cytogenetic studies on two F1 hybrids of autotetraploid rice varieties showing extremely high level of heterosis. Plant Syst Evol 267: 205–213 Luan L, Wang X, Long WB, Liu YH, Tu SB, Xiao XY, Kong FL (2009) A comparative cytogenetic study of the rice (Oryza sativa L.) autotetraploid restorers and hybrids. Genetika 45: 1225–1233 Luo Q, Li Y, Shen Y, Cheng Z (2014) Ten years of gene discovery for meiotic event control in rice. J Genet Genomics 41: 125–137 Morita R, Sugino M, Hatanaka T, Misoo S, Fukayama H (2015) CO2responsive CONSTANS, CONSTANS-like, and time of chlorophyll a/b binding protein Expression1 protein is a positive regulator of starch synthesis in vegetative organs of rice. Plant Physiol 167: 1321–1331 Oliveros JC (2007) VENNY: an interactive tool for comparing lists with Venn Diagrams. http://bioinfogp.cnb.csic.es/tools/venny/index.html (March 29, 2015) Pan Y, Li Q, Wang Z, Wang Y, Ma R, Zhu L, He G, Chen R (2014) Genes associated with thermosensitive genic male sterility in rice identified by comparative expression profiling. BMC Genomics 15: 1114 Pyronnet S, Sonenberg N (2001) Cell-cycle-dependent translational control. Curr Opin Genet Dev 11: 13–18 Sato Y, Namiki N, Takehisa H, Kamatsuki K, Minami H, Ikawa H, Ohyanagi H, Sugimoto K, Itoh J, Antonio BA, et al (2013) RiceFREND: a platform for retrieving coexpressed gene networks in rice. Nucleic Acids Res 41: D1214–D1221 Scoccianti V, Speranza A, Crinelli R, Calzoni GL, Biasi R, Altamura MM, Bagni N (1999) Development-related changes of protein ubiquitination in pollen from male and female kiwifruit (Actinidia deliciosa). Physiol Plantarum 107: 128–135 Shahid MQ, Chen FY, Li HY, Wang SZ, Chen PF, Lin SQ, Liu XD, Lu YG (2013a) Double-neutral genes, Sa-n and Sb-n, for pollen fertility in rice to overcome indica 3 japonica hybrid sterility. Crop Sci 53: 164–176 Shahid MQ, Li YJ, Saleem MF, Naeem M, Wei CM, Liu XD (2013b) Yield and yield components in autotetraploid and diploid rice genotypes 2716 Plant Physiol. Vol. 169, 2015 Downloaded from on June 14, 2017 - Published by www.plantphysiol.org Copyright © 2015 American Society of Plant Biologists. All rights reserved. Allelic Interaction and Gene Expression in Polyploid (indica and japonica) sown in early and late seasons. Aust J Crop Sci 7: 632–641 Shahid MQ, Liu GF, Li JQ, Naeem M, Liu XD (2011) Heterosis and gene action study of agronomic traits in diploid and autotetraploid rice. Acta Agr Scand Sect B-S P 61: 23–32 Shahid MQ, Sun JF, Wei CM, Zhang P, Liu XD (2010) Studies on the abnormality of embryo sac and pollen fertility in autotetraploid rice during different growing seasons. Pak J Bot 42: 7–19 Shahid MQ, Xu HM, Lin SQ, Chen ZX, Naeem M, Li YJ, Liu XD (2012) Genetic analysis and hybrid vigor study of grain yield and other quantitative traits in autotetraploid rice. Pak J Bot 44: 237–246 Stuart JM, Segal E, Koller D, Kim SK (2003) A gene-coexpression network for global discovery of conserved genetic modules. Science 302: 249–255 Tang X, Zhang ZY, Zhang WJ, Zhao XM, Li X, Zhang D, Liu QQ, Tang WH (2010) Global gene profiling of laser-captured pollen mother cells indicates molecular pathways and gene subfamilies involved in rice meiosis. Plant Physiol 154: 1855–1870 Tu SB, Luan L, Liu YH, Long W, Kong FL, He T, Xu QF, Yan WG, Yu MQ (2007) Production and heterosis analysis of rice autotetraploid hybrids. Crop Sci 47: 2356–2363 Vannier JB, Depeiges A, White C, Gallego ME (2006) Two roles for Rad50 in telomere maintenance. EMBO J 25: 4577–4585 Wright KM, Arnold B, Xue K, Šurinová M, O’Connell J, Bomblies K (2015) Selection on meiosis genes in diploid and tetraploid Arabidopsis arenosa. Mol Biol Evol 32: 944–955 Wu J, Shahid MQ, Guo H, Yin W, Chen Z, Wang L, Liu X, Lu Y (2014) Comparative cytological and transcriptomic analysis of pollen development in autotetraploid and diploid rice. Plant Reprod 27: 181–196 Wu JW, Hu CY, Shahid MQ, Guo HB, Zeng YX, Liu XD, Lu YG (2013) Analysis on genetic diversification and heterosis in autotetraploid rice. Springerplus 2: 439 Xu C, Bai Y, Lin X, Zhao N, Hu L, Gong Z, Wendel JF, Liu B (2014) Genome-wide disruption of gene expression in allopolyploids but not hybrids of rice subspecies. Mol Biol Evol 31: 1066–1076 Yang H, Lu P, Wang Y, Ma H (2011) The transcriptome landscape of Arabidopsis male meiocytes from high-throughput sequencing: the complexity and evolution of the meiotic process. Plant J 65: 503–516 Yang PM, Huang QC, Qin GY, Zhao SP, Zhou JG (2014) Different droughtstress responses in photosynthesis and reactive oxygen metabolism between autotetraploid and diploid rice. Photosynthetica 52: 193–202 Yant L, Hollister JD, Wright KM, Arnold BJ, Higgins JD, Franklin FCH, Bomblies K (2013) Meiotic adaptation to genome duplication in Arabidopsis arenosa. Curr Biol 23: 2151–2156 Zhang GQ, Lu YG (1989) Genetic studies of the hybrid sterility in cultivated rice (Oryza sativa). I. Diallel analysis of the hybrid sterility among isogenic F1 sterile lines. Chinese J Rice Sci 3: 97–101 Zhang GQ, Lu YG (1993) Genetic studies of the hybrid sterility in cultivated rice (Oryza sativa). II. A genic model for F1 pollen sterility. Acta Genet Sin 20: 222–228 Zhang GQ, Lu YG, Zhang H, Yang JC, Liu GF (1994) Genetic studies on the hybrid sterility in cultivated rice (Oryza sativa). IV. Genotypes for F1 pollen sterility. Acta Genet Sin 21: 34–41 Zhang ZS, Lu YG, Liu XD, Feng JH (2005) Nuclear and cell migration during pollen development in rice (Oryza sativa L.). Sex Plant Reprod 17: 297–302 Zhang ZS, Lu YG, Liu XD, Feng JH, Zhang GQ (2006) Cytological mechanism of pollen abortion resulting from allelic interaction of F1 pollen sterility locus in rice (Oryza sativa L.). Genetica 127: 295–302 Zhao L, He J, Cai H, Lin H, Li Y, Liu R, Yang Z, Qin Y (2014) Comparative expression profiling reveals gene functions in female meiosis and gametophyte development in Arabidopsis. Plant J 80: 615–628 Zhao MH, Liu XD, Lu YG, Li JQ, Guo HB (2006) Chromosome pairing behavior and reproduction in the hybrid developed by the interaction of different pollen sterile genes in autotetraploid rice. Acta Agron Sin 32: 1472–1478 Plant Physiol. Vol. 169, 2015 2717 Downloaded from on June 14, 2017 - Published by www.plantphysiol.org Copyright © 2015 American Society of Plant Biologists. All rights reserved.