Plant male sterility is a condition where the plant fails to produce functional pollen. This failure could be due to defects in anther specification and differentiation, abnormal microsporogenesis and pollen development, non-dehiscent anthers, or the inability of pollen to germinate and fertilize the female gametes (Pei et al., 2024). Male-sterile female lines form the foundation of large-scale hybrid seed production in various crop species. Such hybrids with superior agronomic traits such as higher yield, enhanced disease resistance and stress tolerance have greatly contributed to food, feed and nutritional security (Kim and Zhang, 2018). To date, male sterility has been reported in ~617 plant species, including crops like rice, wheat, Indian mustard, cotton, soybean, and Sorghum (Chen and Liu, 2014; Vasupalli et al., 2021; Ren et al., 2022) and tree species like bamboo (De Souza et al., 2020), olive (Besnard et al., 2000), rubber (Shearman et al., 2014). Besides its utility in F1 seed production, male sterility might be an essential feature of plant evolution and adaptation (Budar and Pelletier, 2001).

Various kinds of male sterile lines have been identified/developed to date, such as cytoplasmic male sterility (CMS), genic male sterility (GMS), Environment-sensitive GMS (EGMS), and biotechnology-based male sterile (BBMS) lines. CMS is a maternally inherited trait governed by the mitochondrial genome and rescued by the nuclear genes (Vasupalli et al., 2016). Although CMS has been used commercially in many crops, lack of stable restorer genes and cytoplasmic penalty have severely restricted the expansion of CMS to new crops (Kumar et al., 2012; Chamola et al., 2013). GMS, which is controlled by only nuclear genes, can overcome these disadvantages (Wan et al., 2019). However, the primary disadvantage is that pure male sterile stocks cannot be perpetuated (Anjani, 2005). EGMS is a conditional male sterile system where male sterility/fertility is influenced by environmental conditions such as photoperiod, temperature and humidity. Thus, by growing plants under appropriate environmental conditions, the same stock can be used as male sterile or fertile to produce F1 hybrid seeds or for maintenance, respectively (Peng et al., 2023) ( Figure 1 ). Therefore, EGMS systems are preferred for hybrid seed production and widely adopted especially in China. Further, the discovery of GMS genes and advancements in genetic transformation/gene editing technologies have led to the development of BBMS lines (Wu et al., 2016). In this review, we present a brief introduction to GMS and provide in-depth details of molecular mechanisms of EGMS. Further, we also provide recent developments in BBMS systems for hybrid seed production in crop plants.

GMS controlled by nuclear genes is widely reported in plants, which is not surprising considering that anther/pollen differentiation and development involve the interplay of many genes (Cheng et al., 2020). Anther/pollen development comprises different phases viz., archesporial cell specification, somatic differentiation, pollen mother cell meiosis, tapetum development and mature pollen development, and involves cell division, differentiation, development, degradation, and maturation ( Figure 2 ). Anther/pollen development by far needs more genes than any other plant organ, and mutations of these genes often lead to complete or partial male sterility (Marchant and Walbot, 2022). Archesporial cells are the premeiotic cells that develop into the plant germline cells, which ultimately develop into gametes. Therefore, defects in archesporial cell development lead to genic male sterility (Zhou et al., 2017). The genes reported to cause GMS in archesporial cell development are OsMIL1 (Hong et al., 2012b), ZmMSCA1 (Chaubal et al., 2003), AtROXY1/2 (Xing and Zachgo, 2008) and AtTGA9/10 (Murmu et al., 2010). After the archesporial cells specification, somatic cell differentiation takes place. During somatic cell division, many genes are reported to be involved in causing GMS. For example, AtEMS1/EXS (Zhao et al., 2002), AtTPD1 (Yang et al., 2005), OsMSP1 (Nonomura et al., 2003), OsMIL2 (Hong et al., 2012a), OsTIP2/bHLH142 (Fu et al., 2014), ZmMAC1 (Wang et al., 2012), ZmOCL4 (Vernoud et al., 2009), and ZmMs23 (Nan et al., 2017). The four somatic layers, epidermis, endothecium, middle layer and tapetum, are formed from the periclinal somatic cell division. The tapetum is the innermost layer that encircles the developing pollen mother cell and provides the nutrients required for pollen development and also supplies important proteins that coat the surface of pollen grains. Thus, any defect in the development of the tapetum layer causes GMS. Further, tapetum differentiation and meiosis occur simultaneously during anther development. Some of the genes reported that cause defects in tapetum development and pollen mother cell meiosis are OsTDF1 (Cai et al., 2015), OsUDT1 (Jung et al., 2005), ZmMS32 (Moon et al., 2013a), AtDYT1 (Zhang et al., 2006), AtMYB33/65 (Millar and Gubler, 2005), OsGAMYB (Liu et al., 2010b), AtNEF1 (Ariizumi et al., 2004) etc. Besides, callose deposition on the newly formed tetrad microspores and timely callose degradation to release the tetrad microspores are crucial in pollen development. Mutations in genes associated with callose metabolism such as OsDMD1 (Ren et al., 2020), OsGSL1, OsGSL5 (Enns et al., 2005), OsLecRK5 (Wang et al., 2020), OsG1 (Wan et al., 2011), AtCDM1 (Lu et al., 2014), AtCalS5 (Dong et al., 2005) have been shown to cause GMS. Further, genes acting at later stages of pollen development, namely, AtMS1/2 (Ito and Shinozaki, 2002; Chen et al., 2011a), AtMYB26 (Steiner-Lange et al., 2003), AtMGP1 (Li et al., 2010), AtSK32 (Dong et al., 2015), OsGT1 (Moon et al., 2013b), OsDTC1 (Yi et al., 2016), and ZmIPE1 (Chen et al., 2017) have been identified whose loss-of-function lead to male sterility. Post anthesis, successful pollen germination on the stigma involves crucial steps such as adhesion, hydration, germination and tube growth. A few genes, such as AtMSL8 (Hamilton and Haswell, 2017), AtAPY1/2 (Wolf et al., 2007), AtSEC8 (Cole et al., 2005), and OsHXK5 (Lee et al., 2020), are reported to be involved in mature pollen interaction with stigma and germination, and mutations in these genes result in male sterility.

Unlike CMS, GMS requires only two lines for hybrid seed production ( Figure 1 ). Since all normal lines can serve as restorers, GMS permits the testing of a vast pool of germplasm for the exploitation of heterosis. Besides, there is no cytoplasmic effect. The main hindrance, however, is that a complete male sterile population cannot be obtained, and male sterile plants have to be isolated from a 1:1 mixture of sterile and fertile progeny. Further, as in CMS, two separate isolated fields are needed one each for hybrid seed production and multiplication of male steriles. Nevertheless, GMS has been used for hybrid seed production in some crops such as cotton, cowpea, and vegetable Brassica. To circumvent the problem of isolating male steriles from mixture, conditional male sterile lines were identified which could produce fertile pollen under permissible conditions thereby giving rise to all male sterile offspring. Such conditional male sterile systems form the basis of Environment-sensitive GMS (EGMS) systems, and their molecular mechanisms are discussed below.

Genes regulating anther development which respond to environmental stimuli will express male sterility under certain environment and are therefore referred to as EGMS (Fan and Zhang, 2018). EGMS was first identified in rice Nongkeng58S (NK58S) by Shi in 1973 (Peng et al., 2023). So far, four types of EGMS responding to photoperiod, temperature, nitrogen status or humidity have been described (Peng et al., 2023). Generally, photosensitive GMS (PGMS) is characterized by male sterility under long days and male fertility under short days. In contrast, ‘reverse PGMS’ is male sterile during short days and male fertile during long days. Thermo-sensitive GMS (TGMS) is characterized by male sterility under high temperature (exceeding a threshold temperature) and male fertility under low temperatures (below a critical sterility-inducing temperature). On the other hand, ‘reverse TGMS’ exhibits the opposite phenotype. Male sterility caused by nitrogen deficiency is known as nitrogen-sensitive genic male sterility (NGMS), where male fertility is observed under conditions of ample nitrogen supply. Additionally, male sterility under low humidity conditions and male fertility under high humidity conditions is referred to as humidity-sensitive genic male sterility (HGMS). Thus, EGMS behaves as male sterile or male fertile under different conditions and, therefore, could be easily maintained by selfing and used for hybrid seed production under appropriate conditions. Hence, EGMS-based pollination control systems are called two-line systems.

EGMS two-line hybrid system ( Figure 1 ) offers distinct advantages over CMS or GMS, which include ease of maintenance of male steriles, ready availability of restorers, and male sterility being recessive, the F1s will be fertile under all conditions (Chen et al., 2023; Peng et al., 2023). These features allow rapid improvement of parental lines for quality, stress tolerance etc. However, two-line systems have limitations, such as the need for different locations with suitable environmental conditions, and vulnerability to sudden fluctuations in weather patterns. Further, different environmental conditions may also have adverse effects, such as reduced plant vigor, low yield of hybrid or parental seeds (Chen et al., 2023). Despite these limitations, two-line hybrids have been particularly exploited in China for breeding hybrid rice and contributed significantly to enhance rice production and productivity.

EGMS has been reported in numerous plant species, such as Arabidopsis (P/TGMS Acos5-2, TGMS Res1 Rpg1, HGMS Cer1-1, Cer1-m) (Aarts et al., 1995; Zhang et al., 2020a; Zhu et al., 2020), rice (PGMS NK58S, TGMS AnS-1, Zhu1S, HGMS hms1) (Zhou et al., 2014; Fan et al., 2016; Chen et al., 2020; Wu et al., 2022), maize (TGMS TMS5, MAGO^KD) (Li et al., 2017; Lee et al., 2021), soybean (rPGMS MS3‐KO) (Hou et al., 2022), wheat (TGMS YanZhan 4110S, P/TGMS K78S and K456S) (Li et al., 2020; Yang et al., 2021), millet (P/TGMS A2) (Wei et al., 2021), tomato (TGMS Da107) (Yu et al., 2015), sorghum (TGMS Ji 130A) (Ma et al., 2012), rapeseed (TGMS TE5A) (Yan et al., 2016), cotton (PGMS CCRI9106) (Zhang et al., 2020b) and barley (rTGMS HvMS1OEx) (Fernandez-Gomez et al., 2020) ( Table 1 ). The EGMS genes were identified by fine mapping, map-based cloning, or transcriptome analysis. At least 18 TGMS, five rTGMS, six PGMS, three rPGMS, three P/TGMS, one NGMS and four HGMS genes have been identified in rice ( Table 1 ). Similarly, in other plant species, such as Arabidopsis, maize, wheat, soybean, barley, millet and rapeseed, numerous EGMS genes have been identified. The characterization of EGMS genes not only helps devise two-line hybrid systems but also provides a better understanding of plant-environment interactions influencing male reproductive development (Liu et al., 2023). Molecular analyses of various EGMS lines have revealed differential gene regulation operating at different levels such as transcription, translation or post translation, and involving non-coding RNA as responsible for male sterility/fertility transition. Here we focus on the molecular mechanisms of various EGMS systems.

EGMS systems have been mostly restricted to rice. Although many GMS mutants have been identified in several crop species, their utility in commercial seed production is restricted due to issues such as unstable sterility or unintended side effects, the complexity of producing pure male sterile progenies on a large scale, labor-intensive and time-consuming plant breeding methods to transfer recessive male sterility traits into elite lines. Persistent challenges, like false selection and linkage drag appear unavoidable (Zhou et al., 2023). The advent of genetic engineering offered alternative ways to quickly generate male steriles in elite varieties in a single generation avoiding linkage drag (Gao, 2021).

The first transgenic male sterility/fertility system was developed in tobacco and rape seed through tapetum-specific expression of barnase and barstar genes sourced from Bacillus amyloliquefaciens strain H (IAM1521) (Mariani et al., 1990, Mariani et al., 1992). barnase gene codes for an extracellular ribonuclease enzyme with two cysteine residues, which is explicitly inhibited by a small 89 amino acids intracellular protein coded by the barstar gene. Male sterile plants were obtained by anther specific expression of the barnase gene (Mariani et al., 1990) whereas fertility restoration was achieved by anther-specific expression of the barstar gene (Mariani et al., 1992). A bar gene conferring tolerance to glufosinate herbicide was linked to barnase and barstar genes. Barnase-barstar system has been demonstrated to work in several crop species like rapeseed, Indian mustard, soybean, rice, vegetable species (Jagannath et al., 2002; Colombo and Galmarini, 2017). This system is comparable to CMS system and involves three lines for hybrid seed production. Kriete et al. (1996) engineered conditional male sterility system by anther-specific expression of a bacterial gene argE coding for N-acetyl L-ornithine deacetylase. When transgenic tobacco plants are sprayed with non-phytotoxic herbicide conjugate N-acetyl phosphinothricin, expression of argE gene in anthers produces toxic phosphinothricin in anthers causing their destruction. This conditional male sterility system is a two-line system. Ruiz and Daniell (2005) engineered cytoplasmic male sterility through plastid transformation. Transplastomic tobacco plants carrying phA gene coding for ß-ketothiolase driven by psbA promoter showed complete male sterility. Male fertility restoration was achieved when such plants were grown under continuous illumination. However, a practical male fertility restoration system for transplastomic male sterility has not been devised so far. Subsequently, several transgenic male sterility-fertility restoration systems have been reported in model plants (Shukla et al., 2017; Gautam et al., 2023). So far, only the Barnase-barstar system has been commercialized in a few countries such as USA, Canada and Australia. Objections to genetically modified crops in Europe and many other countries have prevented commercialization of transgenic male sterility systems for hybrid seed production.

To bypass the use of transgenic lines in hybrid seed production, Perez-Prat and Van Lookeren Campagne (2002) proposed generation of transgenic male fertile maintainer lines for recessive GMS lines. Using this concept, DuPont Pioneer developed the Seed Production Technology (SPT) system in maize utilising ms45 male sterility gene (Wu et al., 2016). This system consists of a transgenic hemizygous SPT maintainer line (in ms45/ms45 background) and carries wild-type Ms45 transgene linked to a chimeric α-amylase gene containing amyloplast targeting signal peptide to disrupt the pollen germination, and a fluorescent seed color marker gene (DsRed2) ( Figure 4A ). The wild-type Ms45 allele in transgenic plant rescues male sterility, producing 50% non-transgenic viable pollen. The rest 50% transgenic pollen are rendered non-viable due to breakdown of starch by the action of α-amylase transgene. Therefore, the cross between male-sterile line and hemizygous SPT maintainer line yields only non-transgenic male-sterile seeds for use in hybrid seed production. Further, the SPT maintainer line could be propagated through self-pollination where 50% seeds containing the SPT construct can be separated from non-transgenic seeds by fluorescent seed color sorting (Wu et al., 2016).

The SPT system has many advantages, such as the availability of a broad range of germplasm for parental selection, an increase in hybrid seed purity and yield, and the elimination of public concerns about GM crops. However, the SPT constructs show pollen transmission ranging from 0.002% to 0.518% depending on the SPT construct, genetic background and transformation event (Wu et al., 2016). To reduce transgene escape, multicontrol sterility (MCS) system carrying two pollen disruption genes (α-amylase and Dam) has been developed (Zhang et al., 2018). Further, the MCS system also includes a herbicide-resistant gene (Bar) for selecting transgenic seeds (Zhang et al., 2018).

Similar to SPT and MCS systems, a nuclear male sterile rice maintainer line was developed by cloning OsNP1, α-amylase and DsRed2 genes into the EMS mutant osnp1 (Chang et al., 2016). Likewise, Du et al. (2020) created transgenic maintainer line in tomato by introducing the wild type fertility allele linked to seedling color marker in male sterile mutant. The offspring of homozygous male sterile plants and hemizygous maintainers produce 50% non-transgenic male sterile and 50% transgenic fertile seedlings, which can be sorted out through the seedling color before transplantation and obtain a large quantity of recessive genic male-sterile plants.

The above SPT systems are based on recessive male sterility and require prior knowledge of male sterility causing genes and male sterile mutants. If the mutants are not available, two different stocks have to be generated through transformation; the recessive male sterile line through genome editing (followed by selection of transgene-free lines) and transgenic male fertile maintainer line. To circumvent this major bottleneck, An et al. (2020) developed a dominant male sterility system (DMS) in maize, which would require development of only one transgenic line for hybrid seed production. They demonstrated that premature expression of a tapetum specific gene ZmMs7 causes dominant male sterility through delayed PCD of tapetum leading to defects in pollen exine deposition. By linking male sterility construct (p5126::ZmMs7) with the fluorescent marker gene mcherry (expressed in aleurone layer of seed) and constitutively expressed bar gene (conferring Basta herbicide tolerance), they developed a system to track the male sterile individuals at the seed stage. The dominant male sterility-based hybrid seed production system is depicted in Figure 4B . The male sterile line could be multiplied by crossing with its non-transgenic counterpart, which will yield 50% transgenic (male sterile) seeds that could be sorted by fluorescence screening. The presence of bar gene in male sterile lines provides additional opportunity to eliminate any non-transgenic plants in hybrid seed production field. Similarly, hybrid seeds could be produced by crossing with non-transgenic combiner line where non-transgenic hybrid seeds could be separated for commercial use by fluorescence screening. The Ms7 gene is highly conserved among diverse plant species such as rice, barley, Arabidopsis, Brassica etc. Further, transformation of Arabidopsis with the above construct gave dominant male sterility as observed in maize (An et al., 2020) thereby demonstrating that this is a versatile hybrid seed production system. Thus, DMS system provides a robust, widely applicable approach to engineer pollination control system across crop species.

In a further attempt to avoid the use of transgenics altogether, Zhou et al. (2023) identified a male sterility gene (TM6) in tomato which is closely linked to a seedling marker gene Anthocyanin Without (AW) coding for DFR enzyme. Male sterile TM6 mutants (ms-15, ms-26 and ms-47) can thus be identified at the seedling stage by hypocotyl color. They employed CRISPR/Cas9 system to knock out these two genes to obtain male sterile line lacking anthocyanin pigment in the seedling hypocotyls. This system exhibited male sterility along with a distinctive marker (green hypocotyls), facilitating the identification of sterile plants at the seedling stage. However, it still gives rise to a few recombinants between these linked genes which need physical rouging ( Figure 5 ). Likewise, mutant alleles of tomato MS-10 gene, ms-10, ms-24, ms-35 and ms-36, cause male sterility and MS-10 is linked to anthocyanin absent (aa) gene (Zhang et al., 2024). Thus ms-10 and aa gene pair provides another opportunity to use linked seedling marker for selecting male sterile progenies. However, finding a closely linked morphological marker to the male sterility gene in each crop species will be a challenge.

EGMS has demonstrated significant value in hybrid seed production due to its ability to exploit environmental conditions to induce male sterility and fertility phenotype. However, the molecular mechanisms underlying EGMS systems have been extensively studied only in rice, leaving a major gap in understanding their mode of function in other major crops. Investigations into EGMS systems of other major crops are crucial to maximize their utility. Additionally, studies exploring the impact of climate change on EGMS systems are also needed to ensure their stability and adaptability in hybrid seed production. Meanwhile, SPT, MCS and DMS systems offer innovative solutions to overcome the challenges of conventional seed production methods such as manual emasculation and dependence on environmental conditions. Despite their advantages, these systems have limitations such as the need for sophisticated seed sorting machinery and increased cost of seed production associated with 50% unused seeds. Addressing these challenges through technological advancements and cost-effective solutions will be crucial for optimizing their utilization in hybrid seed production. Seedling marker linked male sterility as demonstrated in tomato (Zhou et al., 2023) is a very good approach. However, detecting and removing low frequency recombinant male fertile plants is a problem in seed production. In recent years gene editing technologies have been employed to create chromosomal inversions at target locations (Zhang et al., 2017; Schmidt et al., 2019; Ronspies et al., 2021; Hu et al., 2024). Thus, creating a small inversion between the seedling marker gene and male sterility gene can overcome the problem of recombination and thereby make this a perfect pollination control system.

Heterosis benefits have largely been restricted to a few major global crops. In particular, lack of reliable male sterility/fertility control systems has discouraged heterosis breeding in many crops such as pulses, crops bearing small, hermaphrodite flowers such as millets (e.g. finger millet, foxtail millet etc.), seed spices, amaranth. The accumulated knowledge of genes causing male sterility in crops, availability of whole genome sequences, conserved genes and genetic pathways of male gamete development offer great scope for developing male sterility systems in these crops. Although the third-generation DMS system presents the best promise, CMS and EGMS systems will continue to serve the hybrid seed industry, especially in countries where transgenics are not accepted.

Pollen competition is a phenomenon where pollen of different genetic constitutions display variation in the rate of pollen tube growth, leading to one type of pollen showing preferential transmission through the male side. For example, Arabidopsis Cals5 gene mutants produce pollen whose tubes lack callose, a major component in the pollen wall. The cals5-3 mutant produces seeds upon selfing (no pollen competition) but mutant pollen competes poorly against wild type pollen (Nishikawa et al., 2005). Hence, in heterozygous plants mutant allele is poorly transmitted from the male side. Similarly, in rice, loss-of-function mutants of pollen expressed hexokinase gene Hxk5, produce viable pollen and give 5-10% seed set upon selfing. However, hkx5 mutant pollen grow slowly and fail to compete with wild type pollen. Hence, in heterozygous (+/hkx5) plants, the mutant allele is not transmitted from the male side (Lee et al., 2020). A similar slow pollen germination mutant has been identified in ragi (Eleucine coracana), which shows promise for heterosis breeding (Manjappa et al., 2015). Such slow pollen germination mutants offer unique opportunity for hybrid seed production because homozygous mutant stocks can serve as male sterile line for hybrid seed production and can also be multiplied through selfing. Considering that starch metabolism pathway in pollen grains is highly conserved, slow pollen germination feature is particularly attractive. In crops where seedlings are transplanted, genic male sterility combined with seedling/herbicide marker can be effectively used for heterosis breeding.