• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of plntcellLink to Publisher's site
Plant Cell. Dec 2005; 17(12): 3350–3361.
PMCID: PMC1315374

Arabidopsis SOMATIC EMBRYOGENESIS RECEPTOR KINASES1 and 2 Are Essential for Tapetum Development and Microspore MaturationW in Box


Among the >200 members of the leucine-rich repeat receptor kinase family in Arabidopsis thaliana, only a few have been functionally characterized. Here, we report a critical function in anther development for the SOMATIC EMBRYOGENESIS RECEPTOR KINASE1 (SERK1) and SERK2 genes. Both SERK1 and SERK2 are expressed widely in locules until stage 6 anthers and are more concentrated in the tapetal cell layer later. Whereas serk1 and serk2 single insertion mutants did not show developmental phenotypes, serk1 serk2 double mutants were not able to produce seeds because of a lack of pollen development in mutant anthers. In young buds, double mutant anthers developed normally, but serk1 serk2 microsporangia produced more sporogenous cells that were unable to develop beyond meiosis. Furthermore, serk1 serk2 double mutants developed only three cell layers surrounding the sporogenous cell mass, whereas wild-type anthers developed four cell layers. Further confocal microscopic and molecular analyses showed that serk1 serk2 double mutant anthers lack development of the tapetal cell layer, which accounts for the microspore abortion and male sterility. Taken together, these findings demonstrate that the SERK1 and SERK2 receptor kinases function redundantly as an important control point for sporophytic development controlling male gametophyte production.


Plant life alternates between the sporophytic phase and the gametophytic phase represented by the female embryo sac and the male pollen grain. In flowers, special organs are dedicated to the transition from the sporophytic to the gametophytic phase and the maturation of female and male gametophytes. Anthers in Arabidopsis thaliana have a four-lobed structure, each containing a sporangium in which, early in development, two cell lines differentiate: (1) the germ line is a mass of cells that through sporogenesis and gametogenesis produces the male gametophytes and (2) the surrounding sporophytic tissues differentiate into four cell layers named from outside to inside, the epidermis, the endothecium, the middle cell layer, and the tapetum (Sanders et al., 1999; Ma, 2005). Several studies have shown that the surrounding four cell layers are crucial for the development of microsporocytes into mature pollen (Mariani et al., 1990; Burgess et al., 2002). Thus, mature anthers are highly organized structures. However, the early steps in development that determine cell fate remain largely unknown (Ma, 2005).

Plant organ development requires communication between cell layers to coordinate tissue differentiation (Fletcher, 2002; Larkin et al., 2003; Shpak et al., 2004, 2005; Chevalier et al., 2005). Receptor-like kinases (RLKs) (Shiu and Bleecker, 2001; Fritz-Laylin et al., 2005) are good candidates to coordinate these processes during anther development. Indeed, recent genetic studies showed that the plasma membrane–localized leucine-rich repeat receptor kinase (LRR-K) of the BRASSINOSTEROID-INSENSITIVE1 (BRI1) subfamily, EXCESS MICROSPOROCYTES1/EXTRA SPOROGENOUS CELLS (EMS1/EXS), functions in the control of cell fate during anther development (Canales et al., 2002; Zhao et al., 2002). EMS1/EXS1 and BRI1 belong to subfamily X of the LRR-Ks (Shiu and Bleecker, 2001). Mutations in the EMS1/EXS gene cause an increased differentiation of anther cells in microsporocytes, affect anther wall organization, and lack the tapetal cell layer. Moreover, the same developmental phenotypes were observed for mutations in the TAPETUM DETERMINENT1 (TPD1) gene, which encodes a predicted small secreted protein (Yang et al., 2003). TPD1 could regulate cell fate in coordination with EMS1/EXS (Yang et al., 2005).

The SOMATIC EMBRYOGENESIS RECEPTOR KINASE (SERK) genes encode LRR-Ks belonging to the 14-member subfamily II of the LRR-Ks (Shiu and Bleecker, 2001). They share the canonical structure of LRR-Ks but have a limited number of LRR motifs. Arabidopsis SERK3, also named BAK1 (for BRI1-Associated Kinase1), was characterized as a component of brassinosteroid perception/signaling (Li et al., 2002; Nam and Li, 2002). Initial studies have shown that overexpression of SERK1 enhances the ability of suspension cells to undergo somatic embryogenesis (Schmidt et al., 1997; Hecht et al., 2001). To date, nothing is known about the functions of the remaining members of the Arabidopsis LRR II subfamily.

In this study, we investigated the function of a functionally uncharacterized member of the LRR II family, SERK2, which is the closest homolog to SERK1 and related to SERK3/BAK1. Phenotypic analyses of serk1 serk2 double mutants show that SERK1 and SERK2 have a crucial and redundant function in anther development and male gametophyte maturation.


Identification of Single and Double Mutants in the SERK1 and SERK2 Genes

Overexpression of SERK1 in Arabidopsis has been shown to enhance the efficiency of initiation of somatic embryogenesis of embryogenic cultures (Hecht et al., 2001). To identify loss-of-function–associated phenotypes in SERK1 and SERK2 genes, we obtained two SERK1 and one SERK2 insertional T-DNA mutant alleles from the Signal Collection at the Salk Institute (La Jolla, CA). They were named serk1-1, serk1-2, and serk2-1 (Figure 1A). Sequencing of T-DNA insertion flanking borders showed that serk1-1 has a tandem T-DNA insertion located in the 11th exon and an inframe stop codon 151 bp into the T-DNA insertion, and serk1-2 has an insertion in the 11th exon and an inframe stop codon 55 bp into the T-DNA insertion. The predicted proteins SERK1-1 and SERK1-2 lack 103 and 74 C-terminal amino acids, respectively, corresponding to deletions in the kinase domain of 55 and 26 amino acids, respectively. In serk2-1, a tandem T-DNA insertion is located in the third intron (Figure 1A).

Figure 1.
Characterization of SERK1 and SERK2 Loci.

In order to assess the expression level of SERK1 and SERK2 in homozygous T-DNA lines, RT-PCR experiments were performed on bud cDNAs from the wild type and the homozygous double mutants serk1-1 serk2-1 and serk1-2 serk2-1 (Figure 1B). With SERK1-specific primers upstream of the T-DNA insertions, a PCR product was amplified for the double mutants as for the wild type. However, with primers surrounding the insertions, we were not able to amplify any product from serk1-1 and serk1-2 alleles (Figure 1B). Thus, in the double mutants, truncated transcripts were detected at low levels, but mature SERK1 transcripts were undetectable. With SERK2-specific primers, two PCR products were amplified with primers designed either upstream or downstream of the insertion. With primers surrounding the insertion, two serk2-1 products were amplified that showed low levels of expression compared with the wild type (Figure 1B). Sequencing of those products showed that the larger one was a wild-type transcript, whereas the shorter one was due to aberrant splicing causing the loss of exon 4 (Figures 1A and 1B). To better quantify the expression level of SERK2 transcript in both serk1-1 serk2-1 and serk1-2 serk2-1, we performed RNA gel blot experiments on bud mRNA. We were not able to detect expression of full-length or truncated SERK2 transcript in double mutants, whereas full-length SERK2 mRNA was detected in the wild type (Figure 1C). Thus, serk2-1 mutation strongly reduces the SERK2 transcript level and causes aberrant splicing, but the serk2-1 mutant is likely not a null allele. Despite further efforts, we were unable to identify null alleles in publicly unrestricted accessible populations of insertional lines.

Despite multiple analyses, we were not able to unveil a dramatic phenotype for the serk1 and serk2 single mutants at the whole plant developmental level. This suggested that either the single mutants have subtle phenotypes or that SERK genes are functionally redundant to some extent. Indeed, SERK1 and SERK2 are closely related genes from the LRR-K encoding gene family and share 78.9 and 89.3% identity at the nucleotide and amino acid levels, respectively. We therefore analyzed the two double mutants serk1-1 serk2-1 and serk1-2 serk2-1.

T-DNA Insertions in SERK Loci Trigger Sterility

During vegetative growth, serk1-1 serk2-1 and serk1-2 serk2-1 were indistinguishable from wild-type plants, and flowering times were not affected (data not shown). However, during the reproductive phase, both double mutants exhibited short fruits (Figure 2A) and did not produce any seeds.

Figure 2.
Siliques of serk1 serk2 Double Mutants Do Not Produce Seeds.

In order to evaluate the genetic linkage of the two insertion loci to the observed sterility, we examined the progeny of the two different genotypes (serk1-1 serk2-1 and serk1-2 serk2-1), which were either heterozygous for the serk1 alleles or heterozygous for the serk2-1 allele (Table 1). These parental genotypes were fertile (n > 30 for each). Among the progeny, 16.7 to 27.9% of the plants were sterile (Table 1). All plants were PCR genotyped, and all of the sterile plants were homozygous for insertions at both the SERK1 and SERK2 loci (Table 1). All fertile plants carried at least one wild-type allele at the SERK1 or SERK2 locus with a ratio of ~1:2 (wild type:heterozygote), as expected for the segregation of a unique locus within the fertile and nonhomozygous plants (χ2 < 1.89). These results indicate that the phenotype is genetically linked to both the SERK1 and SERK2 loci and that the two mutations that trigger sterility are recessive. We identified two independent serk1 mutant alleles that showed a similar sterile phenotype when combined with the serk2-1 mutation. This result suggests that the disruption of SERK1 was responsible for the sterile phenotype observed for double mutants. As serk1 single mutants produced seeds and the sterile phenotype was observed only when serk1 lines were crossed with the serk2-1 line, serk1 mutations need to be combined with a mutation that is genetically linked to SERK2 to trigger sterility.

Table 1.
Sterility Phenotype and Mutant Genotypes Are Linked

To determine whether mutation of the SERK2 gene, when combined with serk1 mutations, causes the sterile phenotype, we analyzed complementation of the serk1 serk2 sterile phenotype. No complementation was obtained when the SERK2 ORF was expressed under the control of the ectopic cauliflower mosaic virus 35S promoter, indicating that the noncoding sequences of the SERK2 locus may contribute to SERK2 function. Therefore, complementation with a genomic clone of the SERK2 locus was pursued. We introduced a PCR-amplified genomic fragment containing the SERK2 locus, which included 1442 bp upstream of the start codon and 1000 bp downstream from the stop codon in serk1-1/SERK1 serk2-1/serk2-1 plants. Based on EST and cDNA database analyses, this genomic fragment contained no further ORFs apart from SERK2. We genotyped the transformed progeny to identify plants homozygous for serk1-1 and serk2-1 and found that they were able to produce seeds (Figure 2B). Complemented plants expressed full-length SERK2 transcript (Figure 1C). Taken together, these findings demonstrate that the combination of serk1 and serk2 receptor kinase mutations triggers the sterile phenotype observed in the serk1 serk2 double mutants.

serk1 serk2 Double Mutants Do Not Produce Pollen

To explore the cause of the serk1 serk2 sterility, serk1-1 serk2-1 double mutant pistils were fertilized with wild-type pollen. The sterile phenotype was rescued, allowing fruit development and the production of seeds (Figure 2C), and the resulting seedlings were all heterozygous for both insertion loci (n = 16; data not shown). These findings, together with the above genetic analyses, show that the serk1 serk2 mutations trigger male sterility.

We further investigated anther development. Double mutant serk1 serk2 nondehiscent open flowers had six anthers that were smaller than wild-type anthers (data not shown). Alexander's solution, which stains pollen cytosol red to monitor pollen viability (Alexander, 1969), was applied to open flowers. Viable pollen grains clearly appeared in the wild type, in serk1 and serk2 single mutants, and in the complemented serk1-1 serk2-1 double mutant lines (Figure 3A). However, we did not detect any staining in serk1 serk2 double mutant anthers, indicating that serk1 serk2 anthers did not produce pollen (Figure 3A). Moreover, sections in mature anthers reveal the collapsing of the four microsporangia in serk1-1 serk2-1 double mutants (Figure 3B). To examine possible differences in the development of wild-type and mutant pollen, young floral buds were cleared using a protocol that allows examination and optical sectioning of whole anthers (see Methods). In wild-type flowers, premeiotic, tetrad, and mature pollen stages were all identifiable (stages 6, 7, 9, and 10, respectively; Figures 3C, 3E, and 3G). In serk1-1 serk2-1 double mutant anthers at the same stage (based on the size of microsporangia), we observed the same premeiotic structure (Figure 3D). But, unlike wild-type plants, these premeiotic structures appeared to be arrested during development. serk1-1 serk2-1 double mutant anthers did not show the formation of the characteristic imbricated microspores organized as clear multiple distinguishable tetrads at stages 7 (Figure 3F), 10 (Figure 3H), or later (data not shown). However, at this stage, it remains unclear whether initial tetrad-like structures may form or not and whether meiosis does occur or not in double mutant anthers (see below). In addition, some alterations of the anther wall were observed in serk1 serk2 and are described later.

Figure 3.
serk1 serk2 Double Mutants Do Not Produce Pollen.

serk1 serk2 Double Mutants Produce an Increased Number of Microsporocytes That Undergo Meiosis

We investigated the number of sporogenous cells in young locules (stages 4 to 6) by optical sectioning of propidium iodide–stained anthers. We found that the wild type had 17.7 ± 1.3 (±se, n = 26) microsporocytes per microsporangium. By contrast, serk1-1 serk2-1 and serk1-2 serk2-1 showed a significantly increased number of microsporocytes: 26.0 ± 2.9 (n = 23) and 26.6 ± 2.2 (n = 22), respectively. Both wild-type and double mutants were characterized by a large variability in microsporocyte numbers among microsporangia. Nevertheless, whereas we did not find any microsporangia with ≥31 microsporocytes in wild-type stamens, 43.5% in serk1-1 serk2-1 and 40.9% in serk1-2 serk2-1 showed between 30 and 50 nuclei (Figure 4A).

Figure 4.
serk1 serk2 Microsporangia Contain More Microsporocytes Than the Wild Type.

Optical sectioning of propidium iodide–stained anthers showed that both wild-type and double mutant microsporocytes were able to undergo meiosis. Cells with one, two, and four nuclei were observed in both wild-type and serk1-1 serk2-1 mutants corresponding to the product of the first and second meiotic divisions (Figure 4B). At later stages, the wild type showed microspores organized in well-defined tetrad structures (Figure 4C, top), which then developed into pollen grains (Figure 4C, bottom). Comparable well-developed multiple tetrad structures were not observed in serk1 serk2 at this stage, but instead separated cells were observed (Figure 4D, top) that degenerated after meiosis (Figure 4D, bottom). These data indicate that SERK1 and SERK2 function in the control of male germ line cell numbers during anther morphogenesis but perhaps not directly in male germ line fate.

Tapetum Development Is Impaired in serk1 serk2

High-resolution confocal microscopy revealed another impairment of serk1 serk2 double mutant anthers. Whereas in young anthers (stage 5) four cell layers surrounded the sporogenous cell mass in wild-type stamens (Figure 5A), interestingly only three cell layers surrounding serk1-1 serk2-1 double mutant microsporangia could be identified (Figure 5B).

Figure 5.
Absence of Tapetal Cell Layer in serk1 serk2 Double Mutants.

The four cell layers were morphologically distinguishable during development of wild-type anthers. Interestingly, confocal observation of more mature wild-type microsporangia (stage 6) (Sanders et al., 1999) showed that cell layer 3 is flattened with the development of the neighboring cell layers and its nuclei appear flat as well (Figures 5A and 5D). The flat cells remain until stage 10 (Figure 5F). The tapetum develops as a thick and autofluorescent cell layer (Figures 5C to 5F) and undergoes a last nuclear division leading to cells with two separated nuclei (Figures 5C to 5E) (Yang et al., 1999; Weiss and Maluszynska, 2001).

In early stages of serk1 serk2 double mutant anther development, we did not observe the flattened middle cell layer surrounding the pretapetal cells found in the wild-type anthers (3 and 4 in Figure 5A), but instead, we observed a single cell layer with globular nuclei (3 in Figure 5B). At later stages, this cell layer persists (Figure 5G). Moreover, whereas in wild-type anthers an accumulation of autofluorescent material was visible between the middle cell layer and the tapetum (Figures 5D and 5E, arrowheads), in the serk1 serk2 mutants, accumulation of autofluorescence was facing the inside of the microsporangia (Figure 5G, arrowheads). These observations led us to hypothesize that the tapetal cell layer is missing in anthers of the serk1-1 serk2-1 mutant. Moreover, the third cell layer of the double mutant (3 in Figure 5B) does not morphologically develop as the middle cell layer.

Expression of Tapetum-Specific Genes Is Strongly Impaired in serk1 serk2 Double Mutants

To test the hypothesis that the tapetum is absent in serk1 serk2 double mutants, we investigated a set of genes reported to be specifically transcribed in the tapetal cell layer in inflorescences of wild-type and both double mutant plants by semiquantitative RT-PCR. The following tapetum-specific mRNA levels were analyzed: ARABIDOPSIS THALIANA ANTHER7 (ATA7; Rubinelli et al., 1998), MALE STERILITY1 (MS1; Wilson et al., 2001; Ito and Shinozaki, 2002), LYSINE HISTIDINE TRANSPORTER1 (LHT1; Chen and Bush, 1997), and QUARTET3 (QRT3; Rhee et al., 2003). As a control, we made use of ACTIN7, which is expressed throughout plant tissues. Whereas ACTIN7 showed similar expression in wild-type and both serk1 serk2 mutants (Figure 6A), the expression levels of ATA7, MS1, QRT3, and, to a lesser extent, LHT1 were either virtually absent or reduced in the double mutant inflorescences (Figures 6A and 6B). These expression data are in agreement with the confocal microscopy analyses (Figures 4 and and5)5) and further support the notion that the tapetum is absent from anthers in serk1 serk2 double mutants.

Figure 6.
Expression of Tapetum-Specific Genes Is Strongly Reduced in serk1 serk2.

SERK1 and SERK2 Are Expressed in Anther Locules

RT-PCR and microarray analyses showed that SERK1 and SERK2 are expressed in all aerial organs and particularly in flowers and siliques (Figure 7A). serk1 serk2 double mutants showed a tissue-restricted phenotype, indicating that SERK1 and SERK2 transcripts may be abundant in specific cells within the anther. We generated transgenic lines expressing the β-glucuronidase gene (GUS) under the control of a 1.5-kb fragment of the SERK2 promoter. Only few lines (4 of 70) exhibited a weak blue staining, suggesting, together with the difficulty of SERK1 and SERK2 mRNA detection by RNA gel blots, that SERK1 and SERK2 are low abundance transcripts. Cross sections of SERK2 promoter-GUS lines showed that the SERK2 promoter triggers broad gene expression in stage 6 anthers (Figure 7B) but is restricted to the tapetal cell layer at later stages (stage 9; Figure 7C).

Figure 7.
SERK1 and SERK2 Expression in Anthers.

The low expression level of SERK2 was also confirmed by analyzing data from digital Northerns, which have integrated 1401 (in February, 2005) Affymetrix microarray experiments (Zimmermann et al., 2004). These 1401 microarray experiments were analyzed for correlations of SERK1 or SERK2 expression levels with expression levels of the tapetum-specific ATA7 gene. These analyses showed a degree of correlation of SERK1 and SERK2 expression to high ATA7 expression values and further indicate that SERK genes are expressed more widely than the tapetum-specific gene ATA7 (see Supplemental Text 1 and Supplemental Figure 1 online; Rubinelli et al., 1998; Shiu and Bleecker, 2001; Canales et al., 2002; Zhao et al., 2002; Zimmermann et al., 2004) in agreement with RT-PCR experiments (Figure 7A). Furthermore, expression analyses of the same 1401 experiments show that elevated expression levels of SERK1 and SERK2 correlate with high expression levels of the EMS1/EXS receptor kinase gene in a subset of experiments (see Supplemental Figure 1 online).

We further investigated SERK1 and SERK2 tissue expression using RNA in situ hybridizations. SERK1 RNA (Figure 7D) but not SERK2 (Figure 7E) was detected in stamen primordia (stage 2). In older buds, both transcripts are present in locules of stage 4 to 5 anthers, including expression in sporogenous cells and tapetum and also more external cell layers, including the endothecium and middle cell layer (Figures 7F and 7G). At stage 5 to 6, both transcripts were more concentrated in the tapetal cell layer and apparently also the middle cell layer (Figures 7H and 7I). Then, the hybridization signals faded once meiosis occurred (stages 8 to 9; data not shown). As a positive control, the expression of a gynoecium-specific gene was analyzed in parallel by in situ hybridizations (B. Crawford and M.F. Yanofsky, unpublished data). As expected, the gynoecium-specific transcript was not detected in anther locules (data not shown) and therefore differed from the expression pattern of SERK1 and SERK2 transcripts (Figures 7F to 7I). In addition, like SERK1 (Shah et al., 2001), SERK2 is expressed in ovules (data not shown).

This pattern of SERK1and SERK2 expression is in accordance with SERK-yellow fluorescent protein fusion data presented in Albrecht et al. (2005) and Kwaaitaal et al. (2005). The authors show a wide expression of SERK genes in locules at early stages that faded by the end of stage 5 in the microsporocytes. Small differences, such as expression at late stages that included the epidermal cell layer and endotheticum observed using protein-YFP fusions but not in in situ hybridizations, may reflect either technical limitations of both approaches or a divergence between the abundances of SERK transcripts and SERK proteins. Nevertheless, both studies gave a similar general expression pattern for SERK1 and SERK2 in anthers; namely, at first, a wide expression in the locules at early stages, including microsporocytes, and later an expression restricted to surrounding cell layers.


In this study, we report the characterization of redundant SERK1 and SERK2 functions during organ development. The combination of serk1 and serk2 mutations triggers male sterility characterized by a lack of pollen production, showing functional redundancy among these two LRR-K encoding SERK genes. Confocal analyses of propidium iodide–stained microsporangia showed that serk1 serk2 double mutant microsporangia contained more microsporocytes than the wild type. However, serk1 serk2 double mutants contained only three cell layers surrounding the microsporangia instead of the characteristic four cell layers. High-resolution microscopic and molecular investigations showed that the missing cell layer is the tapetum. The lack in tapetum development can explain the degeneration at later stages of the sporogenous cells and the ensuing male sterility.

serk1 serk2 Show Phenotypes Related to tpd1 and ems1/exs

serk1 serk2 double mutants exhibit a lack of development of the tapetal cell layer. The lack of the tapetal cell layer is similar to other mutants described as male sterile, which do not produce any pollen because of either a tapetum function failure or its absence. Previous tapetum ablation studies demonstrated that the presence of the tapetum is essential for pollen production (Koltunow et al., 1990; Mariani et al., 1990). The sporocyteless/nozzle (spl/nzz) mutants are mutated in a nuclear factor necessary for male and female sporangium development (Schiefthaler et al., 1999; Yang et al., 1999). In spl/nzz, at a very early stage, the anther wall fails to develop and microspores do not differentiate. MS1 encodes a tapetum-expressed plant homeodomain-finger transcription factor (Wilson et al., 2001; Ito and Shinozaki, 2002), and NO EXINE FORMATION1 encodes an unknown protein likely involved in lipid metabolism (Ariizumi et al., 2004). These two genes seem to function in pollen feeding and maturation once differentiation of the tapetal cell layer has occurred.

In contrast with the above mutations, ems1/exs and tpd1 are phenotypically similar to serk1 serk2 double mutants: both fail to produce pollen because of the lack of tapetal cell layer development, and both produce more sporogenous cells that degenerate after meiosis (Canales et al., 2002; Zhao et al., 2002; Yang et al., 2003; Ma, 2005). In addition, ems1/exs exhibits smaller embryos in mature seeds, a phenotype not observed in serk1 serk2 double mutants (data not shown) and not reported for tpd1 (Yang et al., 2003).

The model describing early anther development in Arabidopsis proposes that an early cell division separates sporogenous cells from cells that differentiate into the three inner cell layers of the anther wall (Canales et al., 2002; Zhao et al., 2002; Yang et al., 2003, 2005; Ma, 2005). In ems1/exs and tpd1 anthers, the precursors of tapetal cells develop into microsporocytes instead of the tapetum (Zhao et al., 2002; Yang et al., 2003). This suggests that EMS1/EXS and TPD1 proteins function in the developmental fates of the tapetal precursors and likely not in the mitotic processes leading to the physical separation of cell layers. Our results on the serk1 serk2 double mutant support this published model of position-dependent differentiation of tapetum cells and sporogenous cells (Ma, 2005). Nevertheless, an alternative explanation could be that the developing tapetum produces a signal to negatively regulate microsporocyte proliferation. In this later model, serk1 serk2 mutants that lack the tapetal cell layer would have as a secondary phenotypic trait an altered number of microsporocytes triggered by a cell-to-cell communication failure. Time-dependent cell lineage studies using cell type–specific markers and confocal microscopy may help to further analyze these two hypotheses.

Confocal analyses of wild-type anthers in this study showed that the middle cell layer persisted throughout stage 10 as a flat cell layer surrounding the tapetum (Figure 5). Previous findings show middle cell layer degeneration at stages 6 and 7 (Sanders et al., 1999). Confocal microscopy may enhance resolution or growth, and other experimental conditions may slightly affect the stage of degeneration. The inner cell layer in serk1 serk2 double mutants (3 in Figure 5B) never became flat, indicating either that the flattening may be a mechanical consequence of tapetum development or that the inner cell layer of the mutant is not comparable to cell layer 3 of wild-type anthers. Further analyses of the flattening of the middle cell layer identified here may lead to new insights into mechanisms during anther development and functions of SERK1 and SERK2.

Whereas several genes have been identified showing specific expression in the tapetal cell layer, no genes have been identified that are specifically expressed in the middle cell layer or the endothecium, two cell layers with still unknown functions. Identification of cell layer–specific expression markers would greatly help to characterize the serk1 serk2 phenotypic class of mutants. Expression analyses of the tapetum-specific genes ATA7, MS1, QRT3, and LHT1 provide molecular evidence that the missing cell layer in serk1 serk2 is the tapetum (Figure 6; see Supplemental Figure 1 online).

SERK Genes and Functional Redundancy

We show a largely overlapping pattern of expression of SERK1 and SERK2 genes within the locules of developing anthers (Figure 7). In addition, using a SERK2-green fluorescent protein fusion, we detected a large fraction of SERK2 protein in the plasma membrane and additional expression in endomembranes (data not shown). Nevertheless, expression analyses show that both SERK genes are expressed more widely in Arabidopsis, for example, in ovules (data not shown) and in leaves (Figure 7). Whereas serk1 serk2 double mutants showed an anther-specific phenotype and did not clearly affect other aspects of growth or development analyzed in this study, the SERK1 and SERK2 genes are more broadly expressed (Figure 7A; see Supplemental Figure 1 online). The male sterile phenotype of serk1 serk2 double mutants may also limit the resolution of further developmental phenotypes. Overexpression of SERK1 has previously been shown to enable somatic embryogenesis in culture (Schmidt et al., 1997; Hecht et al., 2001). Both SERK1 and SERK2 transcripts are expressed in guard cells (Leonhardt et al., 2004), but double mutants showed no stomatal development or signal response phenotypes (data not shown). Thus, the SERK proteins may have other functions in planta, but either the corresponding phenotypes are weak or are masked by genetic redundancies with other LRR-Ks of the same subfamily II that has 14 members or by functional redundancies with other proteins and network pathways.

Do LRR-Ks of the Families II and X Function as Complexes?

The phenotypic similarities of tpd1, ems1/exs, and serk1 serk2 mutants suggest that SERK1 and SERK2 could function in coordination with the ems1/exs and TPD1 gene products during anther development. LRR-Ks are thought to interact with each other in a specific manner (Dievart et al., 2003; Shpak et al., 2003; Wang et al., 2005b). The involvement of two LRR-K classes in the same developmental process is intriguing, and interaction between those two LRR-K proteins could occur during anther formation to constitute a functional receptor. Independent findings support this hypothesis. Like BAK1/SERK3, SERK2 and SERK1 belong to the LRR II subfamily (Hecht et al., 2001). BAK1/SERK3 was shown to interact with BRI1, an LRR-RLK belonging to the LRR X subfamily, to constitute a functional brassinosteroid receptor complex (Li et al., 2002; Nam and Li, 2002; Wang et al., 2005a) with the extracellular domain of BRI1 functioning as a brassinolide binding protein (Kinoshita et al., 2005). Interestingly, like BRI1, the EMS1/EXS protein is a member of the LRR-K X subfamily. Consequently, it is tempting to suggest a model in which RLKs of the LRR-X (BRI1/EMS1/EXS) subfamily can interact with specific members of the LRR II (SERK/BAK1) subfamily, giving rise to functional perception systems in plants.

Albrecht et al. (2005) have independently identified serk1 serk2 double mutants as lacking the tapetal cell layer and also show SERK1 and SERK2 functions during anther development.


Plant Growth Conditions and Mutant Line Genotyping

Arabidopsis thaliana plants (ecotype Columbia 0) were grown in a Conviron growth chamber in plastic pots filled with ready-to-use soil (Professional Blend). After sewing, pots were kept at 4°C for 4 to 7 d. Growing conditions were 22°C and 75% humidity with a 16-h-light/8-h-dark photoperiod regime at ~75 μE·m2·s−1.

serk2-1 (Salk_058020), serk1-1 (Salk_044330), and serk1-2 (Salk_053021) alleles of the At1g34210 and At1g71830 genes were obtained from the Signal Collection at the Salk Institute. Genotyping PCR reactions for single and double mutants were performed using 5′-GTCTGCTTAATCTCACTGCTTCTTCTG-3′/5′-GTCAATGACATGGGAATTGGTCC-3′ and 5′-GAGCTACAAGTGGCGAGTGATGG-3′/5′-CGACGCTGTTTCGCTTTTGTG-3′ primer pairs for SERK2 and SERK1, respectively, which were mixed with LBa1 (5′-TGGTTCACGTAGTGGGCCATCG-3′) primers of the T-DNA to genotype plants in one reaction. Because both homozygous mutations in the double mutants led to a sterile phenotype, the segregating progenies of homozygous plants for one locus and heterozygous for the other locus were analyzed.

Plant Expression and Transformation

For the SERK2 gene promoter reporter fusion, a 3.3-kb fragment of the SERK2 locus was PCR amplified using 5′-CCAGATCCGAATCGGACTCTTAAGCTGCAC-3′/5′-CCATGGACCAAAAAAAAGCAAATTTCTCCTCCCAG-3′ primers, introducing a NcoI site (underlined) and cloned in pGEM-T easy (Promega). A 1.5-kb fragment was subcloned in pCAMBIA1303 (CAMBIA) using EcoRI and NcoI sites. For GUS activity assays, inflorescences were incubated in 80% acetone for 30 min at room temperature, washed three times with 50 mM phosphate buffer/0.1% Triton X-100, and incubated in the same buffer supplemented with 1 mg·mL−1 5-bromo-4-chloro-3-indolyl-β-d-glucuronic acid overnight at 37°C. For cross sections, inflorescences were then fixed in FBA for 4 h then dehydrated with increasing ethanol solutions. Ethanol was progressively replaced by Histoclear II (National Diagnostics) and Histoclear II by Paraplast Embedding Media (Sigma-Aldrich). Eight-micrometer sections were stained with Toluidine blue (0.5% for 30 s) and mounted in Permount (Fisher Scientific).

For genomic complementation, a 6129-bp fragment was PCR amplified from the F23M19 BAC clone using the 5′-CTCGAGAATATTGGTATGTGTTGTGTTCACGTGAG-3′/5′-CTCGAGCCG TCTGAGAACCCTTTCCATGCTCCCTC-3′ pair of primers (the introduced XhoI site is underlined), cloned in pCR-Blunt (Invitrogen), and sequenced. This fragment was then subcloned using XhoI in pGREENII-0229 (Hellens et al., 2000). After confirmation by restriction digest, Arabidopsis transformation was performed by floral dipping with Agrobacterium tumefaciens carrying pSOUP and the SERK2 genomic construct as previously described (Clough and Bent, 1998). Because serk1-1 serk2-1 plants were unable to produce seeds, we transformed the progeny of a serk1-1/SERK1 serk2-1/serk2-1 plant. Transformants were identified by Basta selection and genotyped by PCR as described above in order to identify the homozygous serk1-1 serk2-1 plants.

Expression Analyses

Total RNAs were extracted using Trizol and quantified by absorption and migration of an aliquot on a gel. For RNA gel blots, total RNA was separated in a denaturing formaldehyde-agarose gel and blotted to a Hybond N+ membrane (Amersham Pharmacia). Blots were hybridized with random-priming 32P-labeled probes (Megaprime DNA labeling system; Amersham Pharmacia). Full-length SERK2 ORF probe was amplified by PCR from a plasmid containing SERK2 ORF using 5′-ATGGGGAGAAAAAAGTTTGAAGCTTTTGGTTTT-3′/5′-TTATCTTGGACCAGACAACTCCATAGC-3′ primers. ACTIN7 probe was amplified by PCR from cDNA using 5′-GGCCGATGGTGAGGATATTCAGCCACTTG-3′/5′-TCGATGGACCTGACTCATCGTACTCACTC-3′. PCR fragments were purified using the QIAEX II kit (Qiagen).

For RT-PCR, 10 μg were treated with RNase-free DNase using the DNA-free kit (Ambion) and used to produce cDNA. Thirteen microliters of DNA-free RNA (0.77 μg·μL−1) were incubated for 5 min at 70°C with oligo(dT) and chilled on ice before adding 0.5 μL RNase inhibitors (RNAsin Plus; Promega), deoxynucleotide triphosphates to a final concentration of 0.5 mM each, M-MLV reverse transcriptase buffer, and 200 units of M-MLV reverse transcriptase (Promega) in a final volume of 25 μL. The cDNA mixes were incubated for 2 h at 37°C, diluted five times, and kept at −20°C prior to use. For each experiment, 5 μL of the cDNAs were used per reaction, and PCR premixes were carefully done to avoid bias between samples for a final volume of 50 μL per reaction. Five-microliter samples were amplified for 20, 25, 30, and 35 cycles. DNA quantifications were performed for nonsaturating conditions using Adobe Photoshop software. Primers used for RT-PCR are described in Table 2.

Table 2.
Primers Used for RT-PCR

In situ hybridization was performed as described previously (Dinneny et al., 2004). The SERK1 and SERK2 antisense probes were transcribed using SP6 RNA polymerase (Promega) on M13F/M13R PCR fragments amplified from pGEM-T easy containing the SERK ORFs in the proper orientation.

Microscopic Phenotyping of Anthers

Flowers of wild-type and double mutants were incubated for 1 h in Alexander's solution (Alexander, 1969) and then cleared in Herr's solution (Herr, 1971) for 2 h. Cross sections were prepared as described for GUS lines. Differential interference contrast pictures were taken from Herr-cleared young buds using an inverted microscope (Nikon TE300) coupled to a digital camera (Nikon D100).

For confocal observation of nuclei, plants were fixed and stained with propidium iodide as described previously (Laufs et al., 1998). Development stages of anthers were defined in accordance with Sanders et al. (1999). Data were acquired using an inverted microscope (Nikon TE2000-U) set up with a QLC100 Dual Nipkow spinning disk (Solamere Technologies), an argon laser (LaserPhysics), and a CCD camera (Photometrics CoolSNAP HQ) controlled by Metamorph software (Universal Imaging). Laser excitation was 488 nm and emission 600 or 510 nm for propidium iodide and autofluorescence background, respectively. The number of microsporocytes was determined using confocal microscopy on propidium iodide–stained stamens at premeiotic stages. Z-stacks with 0.7- to 1-μm intervals were recorded, including whole microsporangia, and used to count the typical microsporocyte nuclei. To avoid a positional artifact between external and internal microsporangia of the same stamen, both were systematically measured.

Accession Numbers

Arabidopsis Genome Initiative numbers for SERK1, 2, and 3 are AT1G71830.1, AT1G34210.1, and AT4G33430.1, respectively.

Supplemental Data

The following materials are available in the online version of this article.

  • Supplemental Text 1. Expression Correlation of SERK Genes.
  • Supplemental Figure 1. SERK1 and SERK2 Expression Levels Correlate with Tapetum Markers.


We thank Marty Yanofsky (University of California, San Diego, CA), Sheila McCormick (University of California, Berkeley, CA), and Hong Ma (Pennsylvania State University, University Park, PA) for comments on the manuscript, José Dinneny and Brian Crawford (University of California, San Diego, CA) for advice on in situ hybridizations, and Daphné Autran (Institut de Recherche pour le Développement, Montpellier, France) for propidium iodide staining protocols. This work was supported by National Institutes of Health Grant R01GM060396, National Science Foundation Grant MCB 0417118, and in part by Department of Energy Grant DE-FG02-03ER15449 to J.I.S. J.C. was partially supported by a Lavoisier Fellowship from the French Ministry for Sciences and Technologies, and R.R.-P. was supported by a Marie Curie International Fellowship.


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.plantcell.org) is: Julian I. Schroeder (ude.dscu.liamoib@nailuj).

W in BoxOnline version contains Web-only data.

Open Access articles can be viewed online without a subscription.

Article, publication date, and citation information can be found at www.plantcell.org/cgi/doi/10.1105/tpc.105.036731.


  • Albrecht, C., Kwaaitaal, M., Hecht, V., Baaijens, E., Kulikova, O., Russinova, E., and de Vries, S.C. (2005). The Arabidopsis thaliana SOMATIC EMBRYOGENESIS RECEPTOR-LIKE KINASES1 and 2 control male gametogenesis. Plant Cell 17, 3337–3349. [PMC free article] [PubMed]
  • Alexander, M.P. (1969). Differential staining of aborted and nonaborted pollen. Stain Technol. 44, 117–122. [PubMed]
  • Ariizumi, T., Hatakeyama, K., Hinata, K., Inatsugi, R., Nishida, I., Sato, S., Kato, T., Tabata, S., and Toriyama, K. (2004). Disruption of the novel plant protein NEF1 affects lipid accumulation in the plastids of the tapetum and exine formation of pollen, resulting in male sterility in Arabidopsis thaliana. Plant J. 39, 170–181. [PubMed]
  • Burgess, D.G., Ralston, E.J., Hanson, W.G., Heckert, M., Ho, M., Jenq, T., Palys, J.M., Tang, K., and Gutterson, N. (2002). A novel, two-component system for cell lethality and its use in engineering nuclear male-sterility in plants. Plant J. 31, 113–125. [PubMed]
  • Canales, C., Bhatt, A.M., Scott, R., and Dickinson, H. (2002). EXS, a putative LRR receptor kinase, regulates male germline cell number and tapetal identity and promotes seed development in Arabidopsis. Curr. Biol. 12, 1718–1727. [PubMed]
  • Chen, L., and Bush, D.R. (1997). LHT1, a lysine- and histidine-specific amino acid transporter in Arabidopsis. Plant Physiol. 115, 1127–1134. [PMC free article] [PubMed]
  • Chevalier, D., Batoux, M., Fulton, L., Pfister, K., Yadav, R.K., Schellenberg, M., and Schneitz, K. (2005). STRUBBELIG defines a receptor kinase-mediated signaling pathway regulating organ development in Arabidopsis. Proc. Natl. Acad. Sci. USA 102, 9074–9079. [PMC free article] [PubMed]
  • Clough, S.J., and Bent, A.F. (1998). Floral dip: A simplified method for Agrobacterium-mediated transformation of Arabidopsis thaliana. Plant J. 16, 735–743. [PubMed]
  • Dievart, A., Dalal, M., Tax, F.E., Lacey, A.D., Huttly, A., Li, J., and Clark, S.E. (2003). CLAVATA1 dominant-negative alleles reveal functional overlap between multiple receptor kinases that regulate meristem and organ development. Plant Cell 15, 1198–1211. [PMC free article] [PubMed]
  • Dinneny, J.R., Yadegari, R., Fischer, R.L., Yanofsky, M.F., and Weigel, D. (2004). The role of JAGGED in shaping lateral organs. Development 131, 1101–1110. [PubMed]
  • Fletcher, J.C. (2002). Shoot and floral meristem maintenance in Arabidopsis. Annu. Rev. Plant Biol. 53, 45–66. [PubMed]
  • Fritz-Laylin, L.K., Krishnamurthy, N., Tor, M., Sjolander, K.V., and Jones, J.D. (2005). Phylogenomic analysis of the receptor-like proteins of rice and Arabidopsis. Plant Physiol. 138, 611–623. [PMC free article] [PubMed]
  • Hecht, V., Vielle-Calzada, J.P., Hartog, M.V., Schmidt, E.D., Boutilier, K., Grossniklaus, U., and de Vries, S.C. (2001). The Arabidopsis SOMATIC EMBRYOGENESIS RECEPTOR KINASE 1 gene is expressed in developing ovules and embryos and enhances embryogenic competence in culture. Plant Physiol. 127, 803–816. [PMC free article] [PubMed]
  • Hellens, R.P., Edwards, E.A., Leyland, N.R., Bean, S., and Mullineaux, P.M. (2000). pGreen: A versatile and flexible binary Ti vector for Agrobacterium-mediated plant transformation. Plant Mol. Biol. 42, 819–832. [PubMed]
  • Herr, J.M. (1971). A new clearing-squash technique for the study of ovule development in angiosperms. Am. J. Bot. 58, 785–790.
  • Ito, T., and Shinozaki, K. (2002). The MALE STERILITY1 gene of Arabidopsis, encoding a nuclear protein with a PHD-finger motif, is expressed in tapetal cells and is required for pollen maturation. Plant Cell Physiol. 43, 1285–1292. [PubMed]
  • Kinoshita, T., Cano-Delgado, A., Seto, H., Hiranuma, S., Fujioka, S., Yoshida, S., and Chory, J. (2005). Binding of brassinosteroids to the extracellular domain of plant receptor kinase BRI1. Nature 433, 167–171. [PubMed]
  • Koltunow, A.M., Truettner, J., Cox, K.H., Wallroth, M., and Goldberg, R.B. (1990). Different temporal and spatial gene expression patterns occur during anther development. Plant Cell 2, 1201–1224. [PMC free article] [PubMed]
  • Kwaaitaal, M., de Vries, S.C., and Russinova, E. (2005). Arabidopsis thaliana Somatic Embryogenesis Receptor Kinase 1 protein is present in sporophytic and gametophytic cells and undergoes endocytosis. Protoplasma 226, 55–65. [PubMed]
  • Larkin, J.C., Brown, M.L., and Schiefelbein, J. (2003). How do cells know what they want to be when they grow up? Lessons from epidermal patterning in Arabidopsis. Annu. Rev. Plant Biol. 54, 403–430. [PubMed]
  • Laufs, P., Grandjean, O., Jonak, C., Kieu, K., and Traas, J. (1998). Cellular parameters of the shoot apical meristem in Arabidopsis. Plant Cell 10, 1375–1390. [PMC free article] [PubMed]
  • Leonhardt, N., Kwak, J.M., Waner, D., Robert, N., Leonhardt, G., and Schroeder, J.I. (2004). Microarray expression analyses of Arabidopsis guard cells and isolation of a recessive abscisic acid hypersensitive protein phosphatase 2C mutant. Plant Cell 16, 596–615. [PMC free article] [PubMed]
  • Li, J., Wen, J., Lease, K.A., Doke, J.T., Tax, F.E., and Walker, J.C. (2002). BAK1, an Arabidopsis LRR receptor-like protein kinase, interacts with BRI1 and modulates brassinosteroid signaling. Cell 110, 213–222. [PubMed]
  • Ma, H. (2005). Molecular genetic analyses of microsporogenesis and microgametogenesis in flowering plants. Annu. Rev. Plant Biol. 56, 393–434. [PubMed]
  • Mariani, C., Beuckeleer, M.D., Truettner, J., Leemans, J., and Goldberg, R.B. (1990). Induction of male sterility in plants by a chimaeric ribonuclease gene. Nature 347, 384–387.
  • Nam, K.H., and Li, J. (2002). BRI1/BAK1, a receptor kinase pair mediating brassinosteroid signaling. Cell 110, 203–212. [PubMed]
  • Rhee, S.Y., Osborne, E., Poindexter, P.D., and Somerville, C.R. (2003). Microspore separation in the quartet 3 mutants of Arabidopsis is impaired by a defect in a developmentally regulated polygalacturonase required for pollen mother cell wall degradation. Plant Physiol. 133, 1170–1180. [PMC free article] [PubMed]
  • Rubinelli, P., Hu, Y., and Ma, H. (1998). Identification, sequence analysis and expression studies of novel anther-specific genes of Arabidopsis thaliana. Plant Mol. Biol. 37, 607–619. [PubMed]
  • Sanders, P.M., Bui, A.Q., Weterings, K., McIntire, K.N., Hsu, Y.C., Lee, P.Y., Truong, M.T., Beals, T.P., and Goldberg, R.B. (1999). Anther developmental defects in Arabidopsis thaliana male-sterile mutants. Sex. Plant Reprod. 11, 297–322.
  • Schiefthaler, U., Balasubramanian, S., Sieber, P., Chevalier, D., Wisman, E., and Schneitz, K. (1999). Molecular analysis of NOZZLE, a gene involved in pattern formation and early sporogenesis during sex organ development in Arabidopsis thaliana. Proc. Natl. Acad. Sci. USA 96, 11664–11669. [PMC free article] [PubMed]
  • Schmidt, E.D., Guzzo, F., Toonen, M.A., and de Vries, S.C. (1997). A leucine-rich repeat containing receptor-like kinase marks somatic plant cells competent to form embryos. Development 124, 2049–2062. [PubMed]
  • Shah, K., Gadella, T.W., Jr., van Erp, H., Hecht, V., and de Vries, S.C. (2001). Subcellular localization and oligomerization of the Arabidopsis thaliana Somatic Embryogenesis Receptor Kinase 1 protein. J. Mol. Biol. 309, 641–655. [PubMed]
  • Shiu, S.H., and Bleecker, A.B. (2001). Receptor-like kinases from Arabidopsis form a monophyletic gene family related to animal receptor kinases. Proc. Natl. Acad. Sci. USA 98, 10763–10768. [PMC free article] [PubMed]
  • Shpak, E.D., Berthiaume, C.T., Hill, E.J., and Torii, K.U. (2004). Synergistic interaction of three ERECTA-family receptor-like kinases controls Arabidopsis organ growth and flower development by promoting cell proliferation. Development 131, 1491–1501. [PubMed]
  • Shpak, E.D., Lakeman, M.B., and Torii, K.U. (2003). Dominant-negative receptor uncovers redundancy in the Arabidopsis ERECTA leucine-rich repeat receptor-like kinase signaling pathway that regulates organ shape. Plant Cell 15, 1095–1110. [PMC free article] [PubMed]
  • Shpak, E.D., McAbee, J.M., Pillitteri, L.J., and Torii, K.U. (2005). Stomatal patterning and differentiation by synergistic interactions of receptor kinases. Science 309, 290–293. [PubMed]
  • Wang, X., Goshe, M.B., Soderblom, E.J., Phinney, B.S., Kuchar, J.A., Li, J., Asami, T., Yoshida, S., Huber, S.C., and Clouse, S.D. (2005. a). Identification and functional analysis of in vivo phosphorylation sites of the Arabidopsis BRASSINOSTEROID-INSENSITIVE1 receptor kinase. Plant Cell 17, 1685–1703. [PMC free article] [PubMed]
  • Wang, X., Li, X., Meisenhelder, J., Hunter, T., Yoshida, S., Asami, T., and Chory, J. (2005. b). Autoregulation and homodimerization are involved in the activation of the plant steroid receptor BRI1. Dev. Cell 8, 855–865. [PubMed]
  • Weiss, H., and Maluszynska, J. (2001). Molecular cytogenetic analysis of polyploidization in the anther tapetum of diploid and autotetraploid Arabidopsis thaliana plants. Ann. Bot. (Lond.) 87, 729–735.
  • Wilson, Z.A., Morroll, S.M., Dawson, J., Swarup, R., and Tighe, P.J. (2001). The Arabidopsis MALE STERILITY1 (MS1) gene is a transcriptional regulator of male gametogenesis, with homology to the PHD-finger family of transcription factors. Plant J. 28, 27–39. [PubMed]
  • Yang, S.L., Jiang, L., Puah, C.S., Xie, L.F., Zhang, X.Q., Chen, L.Q., Yang, W.C., and Ye, D. (2005). Overexpression of TAPETUM DETERMINANT1 alters the cell fates in the Arabidopsis carpel and tapetum via genetic interaction with EXCESS MICROSPOROCYTES1/EXTRA SPOROGENOUS CELLS. Plant Physiol. 139, 186–191. [PMC free article] [PubMed]
  • Yang, S.L., Xie, L.F., Mao, H.Z., Puah, C.S., Yang, W.C., Jiang, L., Sundaresan, V., and Ye, D. (2003). TAPETUM DETERMINANT1 is required for cell specialization in the Arabidopsis anther. Plant Cell 15, 2792–2804. [PMC free article] [PubMed]
  • Yang, W.C., Ye, D., Xu, J., and Sundaresan, V. (1999). The SPOROCYTELESS gene of Arabidopsis is required for initiation of sporogenesis and encodes a novel nuclear protein. Genes Dev. 13, 2108–2117. [PMC free article] [PubMed]
  • Zhao, D.Z., Wang, G.F., Speal, B., and Ma, H. (2002). The EXCESS MICROSPOROCYTES1 gene encodes a putative leucine-rich repeat receptor protein kinase that controls somatic and reproductive cell fates in the Arabidopsis anther. Genes Dev. 16, 2021–2031. [PMC free article] [PubMed]
  • Zimmermann, P., Hirsch-Hoffmann, M., Hennig, L., and Gruissem, W. (2004). GENEVESTIGATOR. Arabidopsis microarray database and analysis toolbox. Plant Physiol. 136, 2621–2632. [PMC free article] [PubMed]

Articles from The Plant Cell are provided here courtesy of American Society of Plant Biologists
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


  • Gene
    Gene links
  • GEO Profiles
    GEO Profiles
    Related GEO records
  • HomoloGene
    HomoloGene links
  • MedGen
    Related information in MedGen
  • Pathways + GO
    Pathways + GO
    Pathways, annotations and biological systems (BioSystems) that cite the current article.
  • Protein
    Published protein sequences
  • PubMed
    PubMed citations for these articles
  • Taxonomy
    Related taxonomy entry
  • Taxonomy Tree
    Taxonomy Tree