AtMPK4 is required for malespecific meiotic ... - Wiley Online Library

The Plant Journal (2011) 67, 895–906

doi: 10.1111/j.1365-313X.2011.04642.x

AtMPK4 is required for male-specific meiotic cytokinesis in Arabidopsis Qingning Zeng1,2, Jin-Gui Chen1 and Brian E. Ellis2,* Department of Botany, University of British Columbia, Vancouver, BC V6T 1Z4, Canada, and 2 Michael Smith Laboratories, University of British Columbia, Vancouver, BC V6T 1Z4, Canada

1

Received 7 April 2011; revised 3 May 2011; accepted 13 May 2011; published online 29 June 2011. * For correspondence (fax 604 822 2114; e-mail [email protected]).

SUMMARY Mitogen-activated protein kinase (MAPK) cascades have been implicated in regulating various aspects of plant development, including somatic cytokinesis. The evolution of expanded plant MAPK gene families has enabled the diversification of potential MAPK cascades, but functionally overlapping components are also well documented. Here we report that Arabidopsis MPK4, an MAPK that was previously described as a regulator of disease resistance, can interact with and be phosphorylated by the cytokinesis-related MAP kinase kinase, AtMKK6. In mpk4 mutant plants, anthers can develop normal microspore mother cells (MMCs) and peripheral supporting tissues, but the MMCs fail to form a normal intersporal callose wall after male meiosis, and thus cannot complete meiotic cytokinesis. Nevertheless, the multinucleate mpk4 microspores subsequently proceed through mitotic cytokinesis, resulting in enlarged mature pollen grains that possess increased sets of the tricellular structure. This pollen development phenotype is reminiscent of those observed in both atnack2/ tes/stud and anq1/mkk6 mutants, and protein–protein interaction analysis defines a putative signalling module linking AtNACK2/TES/STUD, AtANP3, AtMKK6 and AtMPK4 together as a cascade that facilitates male-specific meiotic cytokinesis in Arabidopsis. Keywords: Arabidopsis, MAP kinase cascade, pollen formation, meiotic cytokinesis, AtMPK4, AtMKK6.

INTRODUCTION Cell division, a fundamental process for all living organisms, requires the faithful replication of the nuclear genome followed by the precise partitioning of the replication products between the daughter cells. Coordination of this complex process involves signalling networks that can integrate environmental and developmental cues to ensure that each division occurs at the correct time, and with the appropriate spatial orientation. Cytokinesis forms the final step of the cell division programme, and engages elaborately regulated cellular machinery responsible for the physical division of the mother cell cytoplasm containing the replicated genomes. Somatic cytokinesis in higher plants differs from the process used in animal and yeast cells, where an actomyosin ring contracts centripetally to separate the daughter cells (Balasubramanian et al., 2004). In dividing plant cells the new cell plate expands centrifugally along a plane marked earlier by a cortical microtubule array, the preprophase band (PPB) (Ju¨rgens, 2005). Golgi-derived vesicles deliver the cell wall and membrane components to the edge of the expanding cell plate, which ultimately fuses with the ª 2011 The Authors The Plant Journal ª 2011 Blackwell Publishing Ltd

parental plasma membrane and cell walls. The delivery of these vesicles is directed by the phragmoplast, a plantspecific structure composed of antiparallel sets of microtubules and actin filaments (Nishihama and Machida, 2001). In contrast to somatic cytokinesis, the process of meiotic cytokinesis that gives rise to gamete formation is unusual in that neither a preprophase band nor a classical phragmoplast is formed as part of the cell separation process (Otegui and Staehelin, 2000). Nevertheless, the microtubule cytoskeleton still plays a critical role in the meiotic cytokinesis process. During microsporogenesis in Arabidopsis, four haploid nuclei are generated within the microspore mother cells, and radial microtubule arrays are observed originating from each of the nuclear envelopes. These delineate the nuclear cytoplasmic domain (NCD), a zone along which cell division is executed with the participation of a special cell plate and a centripetally-closing ‘mini-phragmoplast’ (Otegui and Staehelin, 2004). A series of pioneering studies of the regulatory circuits involved in plant cytokinesis have established that signalling through MAPK cascades is required for the successful 895

896 Qingning Zeng et al. execution of the somatic cell division process (Nakashima et al., 1998; Bo¨gre et al., 1999; Calderini et al., 2001; Nishihama et al., 2001; Ishikawa et al., 2002; Soyano et al., 2003; Takahashi et al., 2004). In Nicotiana tabacum (tobacco), NtNACK1/2 kinesins are thought to activate the upper tier of an MAPK cascade consisting of NtNPK1 (MAPKKK), NtNQK1/NtMEK1 (MAPKK) and NtNRK1/Ntf6 (MAPK). MAPK-mediated phosphorylation of the microtubule-associated protein, NtMAP65-1, then alters microtubule dynamics, and thereby modulates the expansion of the phragmoplast (Sasabe et al., 2006). Homologues of each of the tobacco NtNACK-NPQRK cascade components can be identified in the Arabidopsis thaliana genome, and genetic analysis has shown that the AtNACK1/HINKEL kinesin (Strompen et al., 2002), AtANPs (mitogen-activated protein kinase kinase kinase, MAPKKK) (Krysan et al., 2002) and AtMKK6 (MAPKK) (Soyano et al., 2003) are all involved in somatic cytokinesis. Loss of function in any of these loci results in cytokinesis defects such as the formation of incomplete cell walls and multinucleate cells, as well as an overall dwarf growth pattern. However, there is neither genetic nor biochemical data demonstrating whether they indeed form a module and function in a sequential pathway. Although the putative Arabidopsis orthologue of NtNRK/Ntf6 is AtMPK13, there are conflicting results concerning the ability of AtMPK13 to be phosphorylated by AtMKK6 (Melikant et al., 2004; Lee et al., 2008; Lin et al., 2010; Takahashi et al., 2010), and no described cytokinesis defects have been reported in AtMPK13 loss-offunction mutants. Meanwhile, AtMKK6 was recently found to physically interact with not only AtMPK13, but also AtMPK4, AtMPK11 and AtMPK6 in a directed yeast twohybrid screen (Lee et al., 2008). This observation, together with the role that AtMPK4 has recently been shown to play in regulating cortical microtubule bundling (Beck et al., 2010), led us to speculate that AtMPK4 and/or AtMPK11, rather than AtMPK13, might be the biological target(s) of AtMKK6 in regulating cytokinesis. To address this question, we initially confirmed the interactions between AtMKK6 and its target MPKs in vivo, and tested its ability to phosphorylate these targets in vitro. We then explored the functional relationship between AtMPK4 and its relatively unknown paralogue, AtMPK11. The results indicate that these two genes are not functionally equivalent, and that loss of AtMPK11 function has no apparent fertility consequences, whereas mpk4 mutant plants have reduced male fertility. Detailed examination of male reproduction in the mpk4 mutant shows that mpk4 microspore mother cells specifically fail to complete meiotic cytokinesis, although they remain competent to execute mitotic cytokinesis during pollen formation. Protein–protein interaction analysis allows the definition of a signalling module consisting of AtNACK2/TES/STUD, AtANP3 and AtMKK6, acting upstream of AtMPK4 in meiotic cytokinesis.

Taken together, our research reveals an MAP kinase cascade that is indispensible for the successful completion of malespecific meiotic cytokinesis during pollen formation. RESULTS mpk4 plants are pollen defective but mpk11 plants are not AtMKK6 has been shown to be required for cytokinesis (Soyano et al., 2003), and it can interact with AtMPK6, AtMPK13, AtMPK4 and AtMPK11 (Lee et al., 2008). We re-examined this pattern of interactions in yeast, and quantified their strength by using a b-galactosidase (gal) assay, which confirmed that AtMKK6 displays a preferential interaction with AtMPK4 and AtMPK11 (Figure 1a). We further confirmed the interactions of AtMKK6 and the four MPKs in vivo by using the BiFC assay (Figure 1b), where the fluorescent signals were detected in both the nuclear and cytosolic compartments. To test the biochemical relevance of these interactions, we conducted in vitro kinase assays by using recombinant proteins. As shown in Figure 1c, purified constitutively activated (CA)-MKK6 can phosphorylate AtMPK4 strongly, whereas AtMPK6, AtMPK11 and AtMPK13 were much less efficient substrates. In addition, the phosphorylated AtMPK4 displayed a significantly increased activity in phosphorylating myelin basic protein (MBP), whereas the other three MAPKs showed a much weaker increased activity after co-incubation with CA-KK6. AtMPK6 and AtMPK13 display autophosphorylation activity, which has also been observed for other MAPKs (Zhou et al., 2009). AtMPK4 and AtMPK11 form a paralogous set of kinases that share 88% amino acid sequence identity. However, whereas loss of function at the AtMPK4 locus is known to have severe phenotypic consequences (Petersen et al., 2000), we found that the homozygous mpk11 knock-out mutant (SALK_049352; Figure S1a) is indistinguishable from wild type (WT). To explore this apparent functional distinction, we made reciprocal crosses between mpk4 and mpk11 plants, and found that crosses using mpk11 pollen on mpk4 stigmas generated normal seed set (Figure 2a, left panel), whereas siliques resulting from pollination with mpk4 pollen on mpk11 stigmas contained no seeds (data not shown). Siliques originating from mpk4 self-pollinated flowers were short and contained only a small number of seeds (Figure 2a, left panel and Figure S2a). To exclude the possibility that the observed pattern is caused by a genetic interaction between the two mutant loci, we conducted reciprocal crosses between mpk4 and WT Ler plants. The results showed that WT pollen grains could fertilize mpk4 mutant flowers, but that most mpk4 pollen grains could not fertilize WT flowers (Figure 2a, right panel). These observations suggested that the mpk4 mutant has specific functional defects in the male, but not in the female, reproductive organs. Nevertheless, we failed to recover the mpk4 mpk11 double mutant from 179 individual F2 progeny of

ª 2011 The Authors The Plant Journal ª 2011 Blackwell Publishing Ltd, The Plant Journal, (2011), 67, 895–906

AtMPK4 functions in male meiotic cytokinesis 897

Figure 1. AtMKK6 and its potential targets. (a) Quantitative b-galactosidase activity, indicating the intensity of protein– protein interactions in the yeast two-hybrid assay. The relative b-gal units and the standard deviation (SD) for each value are indicated below the graph. (b) Interaction between AtMKK6 and AtMPK4, AtMPK6, AtMPK11 or AtMPK13 in the Arabidopsis protoplast BiFC assay. BF, bright field image; EV, empty vector; YFP, yellow fluorescent protein. (c) In vitro phosphorylation of glutathione-S-transferase (GST)-AtMPK4, -AtMPK6, -AtMPK11 and -AtMPK13 by GST-CA-KK6, and of Myelin basic protein (MBP) by native AtMPKs; CA, constitutively active. p-MPK and p-MBP appear as radioactive spots in the autoradiogram. The corresponding MPK and MBP protein bands are shown in the Coomassie Brilliant Blue stained membrane. Stars indicate the GST-CA-KK6 bands.

an MPK4+/)MPK11+/) plant, probably because of a T-DNA translocation event in the mpk11 mutant that links the MPK4 and MPK11 loci (Clark and Krysan, 2010). The distinct morphological differences between mpk4 and mpk11 mutants argue for little functional redundancy between these two genes, despite their sequence relatedness. However, AtMPK4 is expressed throughout the plant, whereas AtMPK11 transcripts have a much more restricted

Figure 2. Morphological phenotype of mpk4 anther and pollen grains. (a) Pollination of mpk4 flowers with mpk11 pollen (left panel) and reciprocal crosses between Ler and mpk4 (k4) (right panel). The silique resulting from a cross is indicated by the red arrow, and the siliques resulting from mpk4 selfpollination are indicated by the blue arrows. In all the labels, the female genotype is listed before the male genotype. (b) Ler and mpk4 intact flowers. (c) Alexander staining of Ler and mpk4 mature pollen grains. (d) Scanning electron microscopic views of Ler and mpk4 anther and pollen grains.

distribution (Figure S1b), leaving open the possibility that the phenotypic difference between mpk4 and mpk11 plants is a reflection of their different expression patterns, and that


898 Qingning Zeng et al. the absence of AtMPK11 in many tissues therefore does not allow restoration of the WT phenotype in an mpk4 background. To test this, we used the AtMPK4 promoter to drive AtMPK11 expression in the mpk4 mutant background, and looked for complementation of the mpk4 phenotype (Appendix S1). Examination of multiple PromoterMPK4::MPK11 T2 transgenic lines did not detect any evidence for restoration of the WT phenotype, even in lines in which ectopic AtMPK11 expression was higher than expression of the native AtMPK4 gene (Figure S1c). This confirms that AtMPK11 is not functionally equivalent to AtMPK4. Whereas mpk4 floral organs are generally smaller than in the WT (Figure 2b), the overall floral morphology of mpk4 mutants resembles the WT. Alexander staining (Alexander, 1969) of the protoplasm of mature pollen grains suggested that the mpk4 mutant pollen grains are largely nonviable (Figure 2c). Scanning electron microscopic examination of the morphology of mpk4 stamens and pollen demonstrated that mpk4 anthers contained far fewer pollen grains compared with the WT anthers (Figure 2d). In addition, although WT pollen grains were oval with typical midline germination apertures, the mpk4 mutant pollen grains were generally much larger and round, and frequently possessed distorted external morphology, including an abnormal germination aperture. Transgenic expression of the AtMPK4 coding sequence or HA-MPK4 (Petersen et al., 2000) under the control of the AtMPK4 native promoter in the mpk4 mutant can restore the normal size and shape of mature pollen grains (Figure 3), demonstrating that the malformed and enlarged pollen phenotype is caused by the loss of function of AtMPK4 alone. Developing mpk4 pollen cannot undergo normal male meiotic cytokinesis To further examine the impact of the loss of AtMPK4 function on the process of microsporogenesis, we sectioned mpk4 anthers at different developmental stages. At stage 5, the WT anther primordium forms four locules, and the microspore mother cells (MMCs) can typically be visualized within the locules (Figure 4a; Sanders et al., 1999). The MMCs enter meiosis at stage 6, followed by meiotic cytokinesis at stage 7, at which point the resulting haploid microspores are bounded by a callose wall to form tetrads (Figure 4a). This callose wall is enzymatically degraded at development stage 8, which allows the microspores to be released from the tetrads. Individual microspores then become polarized, and continue through two rounds of mitosis to form mature pollen grains. In the mpk4 mutant, anthers were found to develop normally to form locules and MMCs at stage 5 (Figure 4a), but as the MMCs finish meiosis, no callose wall appears between the microspores. At later developmental stages, most of the mpk4 microspores are enlarged and contain multiple nuclei (Figure 4a and Figure S2b). Empty locules,

Figure 3. Complementation of the mpk4 mutant. (a) Pollen phenotype of the wild type (WT; Ler), mpk4 and complementation lines. PromoterMPK4::MPK4 (CDS) was transformed into the MPK4+/) plant and the desired PromoterMPK4::MPK4/mpk4 plants were selected by antibiotic resistance and genotyping. PromoterMPK4::HA-MPK4/mpk4 plants were kindly provided by Morten Petersen (Petersen et al., 2000). All the images were acquired at the same magnification. (b) Confirmation of the genotypes by genomic PCR. Endogenous AtMPK4 (edMPK4) and exogenous AtMPK4 CDS (exMPK4) were amplified by using AtMPK4 forward and reverse primers. The Ds insertion was confirmed by amplifying the GUS gene that is incorporated in the Ds element. (c) Statistical analysis of pollen size for the above genotypes. The relative pollen grain size is defined as the area of pollen on two-dimensional images, and is measured by using IMAGEJ.

or locules containing apparently aborted pollen grains, are also observed in the mpk4 mutant anthers at later stages of flower development (Figure S2b). To further confirm the absence of callose wall formation during microspore development in the mpk4 mutant, we stained whole mpk4 flower buds with Aniline Blue, which binds to callose and emits fluorescence under ultraviolet


AtMPK4 functions in male meiotic cytokinesis 899 light (Spielman et al., 1997; Enns et al., 2005). In WT anthers, the callose wall is clearly present around and between microspores at the tetrad stage (Figure 4b), whereas in the mpk4 mutant, intersporal callose staining is greatly reduced, and faint furrows can be observed that may mark the attempted division of the MMC cytoplasm (Figure 4b). Taken together, these data indicate that mpk4 anther primordia can develop normally to the stage of producing MMCs, but fail to form complete intersporal callose walls during meiotic cytokinesis. mpk4 male gametes can proceed through mitosis and mitotic cytokinesis

Figure 4. Callose wall in pollen tetrads is defective in the mpk4 mutant during meiotic cytokinesis. (a) Toluidine Blue staining of transverse sections of anther locules. Anther developmental stages are indicated at the left side of the pictures, according to Sanders et al. (1999); E, epidermis; En, endothecium; MC, meiotic cell; ML, middle layer; MMC, microspore mother cell; MSp, microspores; T, tapetum; Tds, tetrads. All the images were acquired at the same magnification. (b) Aniline Blue staining of callose in Ler and mpk4 tetrads. Scale bar: 11 lm.

In WT anther locules, when microspores are freed from the tetrads, the nucleus of each microspore moves to one side of the cell and undergoes asymmetric pollen mitosis I (PMI) to form a vegetative nucleus and a generative nucleus. The newly formed generative cell is separated from the vegetative cytoplasm by a transient callose wall (Figure 5a), which is fused at its margins with the vegetative cell wall. In the mpk4 mutant, although the haploid microspores have not separated completely, they can still execute PMI normally. As a result of the prior failure of meiotic cytokinesis, however, PMI yields various shapes of microgamete that contain multiple generative cells (Figure 5a and Figure S3), all of which are properly walled-off from the vegetative cytoplasm. After PMI, the generative cell undergoes a second round of mitosis (PMII) to produce two identical germ cells, resulting in the classic tricellular structure in Arabidopsis WT pollen. The three nuclei can be visualized by 4¢,6diamidino-2-phenylindole (DAPI) staining, which shows the vegetative nucleus as a more diffuse organelle, whereas the two germ cell nuclei appear bright and compact (Figure 6a). When we quantified the number of nuclei in WT pollen by microscopy, 88% of the pollen grains were found to possess the typical tricellular structure, whereas approximately 10% appeared to contain one germ cell and one vegetative cell (Figure 6a), although the latter class may consist of pollen grains in which the two germ cell nuclei were aligned such that one of them could not be detected. When grown under the same conditions, less than 40% of the pollen grains in the mpk4 mutant contained the expected tricellular structure (Figure 6b). Instead, mpk4 pollen grains often contained larger numbers of nuclei (Figure 6a), the frequency distribution of which displayed peaks attributed to (2S + 1V) and (4S + 2V) nuclei, in addition to a few (6S + 3V) nuclei (Figure 6b). The observation that mpk4 pollen grains often contain additional sets of nuclei at varying ploidy levels suggests either that some of the vegetative nuclei in the mpk4 mutant can proceed through postmeiotic development, whereas others cannot, or that the attempts to complete meiotic cytokinesis in the mpk4 background might sometimes be successful at later developmental stages. Consistent with the latter possibility, we observe within


900 Qingning Zeng et al.

Figure 6. Numbers of nuclei in mature pollen grains. (a) 4¢,6-Diamidino-2-phenylindole (DAPI) staining of Ler and mpk4 mature tricellular pollen grains. (b) Quantitative analysis of nuclear abundance in Ler and mpk4 pollen.

Figure 5. Transmission electron microscopy (TEM) of mpk4 pollen. (a) TEM of Ler (i, ii) and mpk4 (iii, iv) microspores at the bicellular stage after PMI; (ii) close-up image of (i). Note that mpk4 microspores contain more than one generative cell, but that these are properly walled-off from the vegetative cell cytoplasm. (b) TEM of mature mpk4 pollen containing internal pollen walls. White arrowheads indicate internal exine-like wall and black arrowheads indicate internal intine-like wall; GC, generative cells; n, nuclei of the vegetative cells; VC, vegetative cells. Scale bar: 10 lm in ai, aiii, aiv, bi and biii; 2 lm in aii, bii and biv.

near-mature mpk4 pollen grains portions of internal exine-like cell walls, which resemble the bacula of exine in terms of staining intensity and shape, as well as internal intine-like cell walls, which appear to have originated from the peripheral intine (Figure 5b). Mature angiosperm pollen do not possess such internal cell walls. Upstream activators of the AtMKK6–AtMPK4 cascade in meiotic cytokinesis It has been reported that anq1/mkk6 mutant Arabidopsis plants form enlarged round pollen grains (Soyano et al.,

2003), implying that AtMKK6 could be the upstream MAP kinase kinase of AtMPK4 in pollen formation. We identified another T-DNA knock-out allele of AtMKK6 (SALK_117230, designated mkk6-2; Figure S4a). The mkk6-2 mutant plants also produce malformed pollen with obvious cell division defects (Figure 7a). AtMKK6-rescued plants display normal vegetative growth as well as pollen formation (Figure S4b,c), indicating that the pollen phenotypes we observed are caused by the loss of function of AtMKK6. Consistent with this conclusion, GUS activity was detected primarily in pollen grains in floral organs from both PromoterMKK6::GUS and PromoterMPK4::GUS lines (Figure 7b). Previous research had identified a loss of function tes/stud/ atnack2 mutant that produces enlarged pollen grains with irregular germination apertures (Figure 7a), which phenocopies the mpk4 mutant pollen grain morphology (see Discussion) (Hulskamp et al., 1997; Spielman et al., 1997; Yang et al., 2003a; Tanaka et al., 2004; Oh et al., 2008). In the tes/ stud/atnack2 mutant, a kinesin-like protein that is required for male-specific meiotic cytokinesis in Arabidopsis is disrupted. Interestingly, the tobacco orthologue of TES/STUD/ ATNACK2 was reported earlier to be the upstream activator of an MAP kinase cascade (Nishihama and Machida, 2000; Takahashi et al., 2004). In light of these parallels, we postulated that TES/STUD/ATNACK2 could be the upstream kinesin that specifically regulates male meiotic cytokinesis upstream of the AtMKK6–AtMPK4 cascade. To test this hypothesis, and to identify the MAPKKK(s) that might possibly connect TES/STUD/ATNACK2 and the AtMKK6–AtMPK4


AtMPK4 functions in male meiotic cytokinesis 901

Figure 7. Phenotypic investigation of the upstream components of AtMPK4 in pollen formation. (a) Scanning electron microscopic views of wild-type Col and WS, and mkk6-2 (Col), mpk4 (Ler), tes-1 (Col), tes-4 (WS) anther and pollen grains. All the images were acquired at the same magnification. Scale bar: 25 lm. (b) Histochemical analysis of AtKK6 Promoter::GUS and AtMPK4 Promoter::GUS gene expression in flowers of transgenic Arabidopsis plants.

Figure 8. Protein–protein interaction between candidate upstream components of the AtMPK4 signalling cascade in pollen formation. (a) Yeast two-hybrid assay. Proteins of interest were fused with either Gal4 DNA binding domain (DB) or the Gal4 transactivation domain (AD), as indicated. Successful co-transformants were selected from synthetic complete drop-out medium without leucine and tryptophan (Sc – LT) plates. Positive interactions were selected based on growth on Sc – LT – His + 3AT plates and on the more stringent Sc – LT – uracil plates. (b) Protein–protein interactions detected in the Arabidopsis protoplast BiFC assay.

ably acting as the upstream MAPKKK(s) for AtMKK6 and AtMPK4 in the context of male gametogenesis. module, we tested the interaction of two MAPKKKs, AtMEKK1 and AtANP3, with the C-terminal region of TES/ STUD/ATNACK2 (TES-C) (Ishikawa et al., 2002), and with both AtMKK6 and AtMPK4. AtMEKK1 was earlier reported to interact directly with AtMPK4 (Ichimura et al., 1998), and to be essential for AtMPK4 activation in different biological contexts (Ichimura et al., 2006; Nakagami et al., 2006; Suarez-Rodriguez et al., 2007). AtANP3, on the other hand, is the closest structural orthologue of NtNPK1 (Jouannic et al., 1999). Protein–protein interaction assays showed that AtMPK4 can indeed interact with AtMEKK1 directly, as previously described, but not with AtANP3, whereas TES-C and AtMKK6 could both interact with AtANP3, but not with AtMEKK1 (Figure 8a,b). These results are consistent with a model in which ANP3 serves as the link between TES/STUD/ ATNACK2 and the AtMKK6–AtMPK4 module. AtANP family members have previously been shown to be required for mitotic cytokinesis (Krysan et al., 2002), and we conclude from our results that one, or more, AtANP isoform is prob-

DISCUSSION AtMPK4 functions in male meiotic cytokinesis The process of pollen formation requires the coordinate action of numerous genes responsible for the establishment of stamen identity, for cell division and differentiation during anther and pollen development, and for executing programmed cell death in specific tissues to allow the release of mature pollen grains (Ma, 2005). AtMPK4 is already known to play a role in disease resistance (Petersen et al., 2000; Andreasson et al., 2005; Brodersen et al., 2006; SuarezRodriguez et al., 2007; Gao et al., 2008; Qiu et al., 2008a), and in this context, one feature of the mpk4 mutant phenotype is a massive increase in constitutive salicylic acid (SA) accumulation, accompanied by the suppression of jasmonic acid (JA)-induced gene expression (Petersen et al., 2000). Arabidopsis mutants defective in jasmonate biosynthesis or signalling are known to display male sterility (Feys et al., 1994; Zhao and Ma, 2000). However, they exert this effect by


902 Qingning Zeng et al. delaying anther dehiscence and inhibiting anther filament elongation (Zhao and Ma, 2000), which is distinct from the mpk4 mutant phenotype. In addition, the signal transduction mutant, snc1, which also displays a dwarf phenotype and constitutively accumulates high levels of SA (Li et al., 2001), shows normal levels of fertility (X. Li, personal communication), and produces pollen grains that are neither enlarged nor aborted (Figure S5a). These data suggest that mpk4 male sterility is not caused by either JA insensitivity or high levels of SA accumulation. In the early stages of anther development, the anther primodium forms four archesporial cells, each of which then develops into a primary sporogenous cell and a primary parietal cell (Scott et al., 2004). Mutations that affect cell differentiation or cell fate determination at this stage lead to male sterility. For example, two paralogous Arabidopsis MAP kinases, AtMPK3 and AtMPK6, are required for normal male gamete formation, as the mpk6)/)mpk3+/) mutant often fails to develop all four locules in a given anther, and as a result, produces fewer viable pollen than the WT (Hord et al., 2008). At the same time, this mutant is also female sterile as a result of the arrest of late-stage cell division in the ovule integuments (Wang et al., 2008a). Other genes, such as SPOROCYTELESS/NOZZLE (SPL/NZZ), EXCESS MICROSPOROCYTES 1/EXTRA SPOROGENOUS CELLS (EMS1/EXS) and TAPETUM DETERMINANT 1 (TPD1), are essential for the formation of sporogenous cells or the somatic tapetum cells (Schiefthaler et al., 1999; Yang et al., 1999, 2003b; Canales et al., 2002; Zhao et al., 2002), and mutations at these loci result in empty anther locules. The mpk4 mutant anther, however, is able to produce a functional tapetum and apparently normal MMCs (Figure 2a), suggesting that AtMPK4 does not play a critical role in the early stages of microsporogenesis. After meiosis, individual haploid microspores go through two rounds of mitosis. Mutations in GEMINI POLLEN 1/ MICROTUBULE ORGANIZATION 1 (GEM1/MOR1) affect the unequal division of mitosis I (PMI) (Park et al., 1998; Twell et al., 2002), whereas duo mutations block mitosis II (PMII) by terminating the generative cell cycle, resulting in pollen grains with one vegetative cell and one generative cell (Durbarry et al., 2005; Brownfield et al., 2009). The plasma membrane-localized MAPKKKs, MAP3Ke1 and MAP3Ke2A, also appear to act after PMI, and mutations in the corresponding genes result in non-viable pollen (Chaiwongsar et al., 2006). Our data, on the other hand, demonstrate that AtMPK4 is not required for either mitotic cytokinesis or for cell lineage determination, but instead plays a specific role in meiotic cytokinesis during pollen development. AtMPK4 in somatic cytokinesis The severe dwarfism phenotype of the mpk4 mutant (Petersen et al., 2000) suggests that AtMPK4 might play a role in both somatic and meiotic cytokinesis, and a recent study

found AtMPK4 to be physically associated with cortical microtubules and to regulate their bundling (Beck et al., 2010). Very recently, Kosetsu et al. (2010) reported that the mpk4-2 mutant in the Columbia background exhibits somatic cytokinesis defects in roots and cotyledons, with about 2% of the cotyledon cells possessing incomplete cell plates. However, the Ler mpk4 mutant did not have such deficiencies (Kosetsu et al., 2010). In addition, AtMPK4 seems to be essential for the correct timing and speed of somatic cytokinesis (Beck et al., 2011). The appearance of internal cell walls in the mature mpk4 pollen grains (Figure 5b) could therefore reflect a mistiming of the meiotic cytokinesis process. Together, these data are consistent with the ubiquitous expression of AtMPK4 transcripts (Figure S1b), but they also emphasize the distinction between the degrees to which AtMPK4 activity is essential for different cellular processes. For example, in order to form the MMCs, the anther primordia have to successfully complete a series of defined cell divisions and differentiation, and the mpk4 mutant anther is able to do so. AtMPK4 thus appears to be essential for male meiotic cytokinesis, and also for somatic cytokinesis in some, but not all, cells and tissues. Such specificity could potentially result from complementation in certain tissues by other distinctively localized AtMPKs that share partial functional redundancy with AtMPK4. Alternatively, AtMPK4 might recruit upstream activators and/or downstream targets that are uniquely expressed in those tissues/cells. mpk4 phenocopies tes/stud/atnack2 Consistent with a ‘recruitment specificity’ model, mutation of an Arabidopsis kinesin, TES/STUD/AtNACK2, distinctively results in a pollen phenotype marked by a defect in male meiotic cytokinesis (Hulskamp et al., 1997; Spielman et al., 1997; Yang et al., 2003a; Tanaka et al., 2004; Oh et al., 2008). Like mpk4, tes/stud/atnack2 floral organs form normal sporogenous cells and peripheral supporting tissues, indicating that neither gene is required for early anther development or for the generation of MMCs. In both mutants, MMCs do not produce an intersporal callose wall, but the mature pollen grains contain irregular internal cell walls, which has been described as an attempt at delayed meiotic cytokinesis (Spielman et al., 1997). Although meiotic cytokinesis is not completed in either mutant genotype, mitotic cytokinesis appears to proceed normally within the multinucleate microgametophyte, resulting in over-size mature pollen grains that contain more than one set of tricellular structures (Figures 6a and 7a), and that sometimes germinate to produce a single pollen tube (Figure S2c), despite containing extra sets of nuclei. The multinucleate microspores thus still retain the ability to complete asymmetric mitotic cell division (Spielman et al., 1997). Homozygous mutants of either locus are able to produce only a few viable seeds (Figure S2a, compared with Yang et al., 2003a: figure 2f). It is clear, therefore, that mpk4 phenocopies tes/stud/


AtMPK4 functions in male meiotic cytokinesis 903 atnack2, and that TES/STUD/AtNACK2 and MPK4 are very likely to operate in the same pathway in regulating meiotic cytokinesis. TES/STUD/AtNACK2 is upstream of the MAP kinase cascade AtANPs–AtMKK6–AtMPK4 It has been proposed that the tobacco kinesin, NtNACK, operates upstream of the so-called MAP NP-Q-RK cascade to modulate phragmoplast expansion during cytokinesis (Takahashi et al., 2004; Sasabe et al., 2006). NtNACK has two orthologues in Arabidopsis: AtNACK1/HINKEL, which is involved in somatic cytokinesis (Strompen et al., 2002; Takahashi et al., 2010), and is essential for embryogenesis (Strompen et al., 2002), and AtNACK2/TES/STUD. As mpk4 essentially phenocopies tes/stud/atnack2 in pollen development, we propose that TES/STUD/AtNACK2 acts upstream of an MAP kinase cascade that involves ANP3, AtMKK6 and AtMPK4. Several lines of evidence support this conclusion: the C terminus of TES/STUD/AtNACK2 (Ishikawa et al., 2002) can interact with one of the cytokinesis-related AtMAPKKK/AtANPs, AtANP3 (Krysan et al., 2002); AtANP3 interacts with AtMKK6 in the protoplast bimolecular fluorescence complementation (BiFC) assay (Figure 8b); and AtMKK6 can interact with and phosphorylate AtMPK4. It is noteworthy that no interaction between AtANP3 and AtMKK6 was detected in yeast cells, but only in the nucleus of Arabidopsis protoplasts (Figure 8a,b), which is consistent with the negative results observed in a recent yeast complementation study (Takahashi et al., 2010), and with the fact that NtNPK1 has been shown to be a nuclear-localized protein (Ishikawa et al., 2002). The ANP3–AtMKK6 interaction might therefore require plant-specific conditions or proteins. On the other hand, although two loss-of-function alleles of AtMKK6, anq-1/mkk6-1 and mkk6-2, produce malformed and enlarged pollen grains (Figure 7a) (Soyano et al., 2003), the anp3 mutant does not (Figure S5b). As the three Arabidopsis ANP family members share considerable redundancy (Krysan et al., 2002), it seems likely that other ANP family members can compensate for loss of ANP3 function in male meiotic cytokinesis signalling. Based on the original tobacco NP-Q-RK model, the Arabidopsis orthologue of NtNRK is predicted to be AtMPK13, rather than AtMPK4 (Soyano et al., 2003). However, biochemical evidence for the phosphorylation of AtMPK13 by AtMKK6 has been conflicting. Whereas AtMPK13 was shown to be activated by AtMKK6 in yeast cells, and in in vitro kinase assays, respectively (Melikant et al., 2004; Lin et al., 2010), other reports observed no AtMPK13 phosphorylation by AtMKK6 (Lee et al., 2008; Takahashi et al., 2010). In our kinase assay, we observed that co-incubation with AtMKK6 resulted in strong phosphorylation and increased activation of AtMPK4, but only very weak phosphorylation of AtMPK13 (Figure 1c). A T-DNA insertion mutant of AtMPK13 (SALK_130193; Figure S6) does not exhibit any morpholog-

ical phenotype, which would appear to make AtMPK13 an unlikely candidate for the downstream target of AtMKK6. However, the SALK_130193 allele disrupts only the long splice variant of AtMPK13, and not the short variant (Figure S6), leaving the biological role of MPK13 uncertain. Nevertheless, as loss of function of AtMPK4 alone causes the meiotic cytokinesis defect, it appears that AtMPK13 does not share overlapping functions with AtMPK4 in the context of pollen meiotic cytokinesis. In addition to AtMKK6, two other Arabidopsis MAPKKs, AtMKK1 and AtMKK2, are also known to interact with AtMPK4 (Lee et al., 2008), and like mpk4, the mkk1 mkk2 mutant is severely dwarfed (Gao et al., 2008; Qiu et al., 2008b). However, the double mkk1 mkk2 mutant displays premature senescence and is ultimately lethal. In addition, we found that the MAPKKK acting upstream of AtMKK1/2, AtMEKK1, does not interact with TES/STUD/AtNACK2. The AtMEKK1–AtMKK1/2–AtMPK4 pathway may therefore help regulate disease and stress responses in Arabidopsis (Suarez-Rodriguez et al., 2007; Gao et al., 2008; Qiu et al., 2008b; Pitzschke et al., 2009), whereas the TES/STUD/ AtNACK2-AtANPs-AtMKK6-AtMPK4 module regulates cytokinesis. Further studies will be needed to elucidate exactly how the plant regulates the activation of two different signalling modules that each engage AtMPK4 as part of both disease resistance responses and developmental signalling. EXPERIMENTAL PROCEDURES Plant growth conditions and maintenance of the mpk4 mutant For all experiments, Arabidopsis seeds were surface sterilized, stratified at 4C for 2–4 days and germinated on agar-solidified halfstrength Murashige and Skoog (MS) medium plates [½MS salt, 1% sucrose, 0.5 g L)1 2-(N-morpholino)ethanesulfonic acid (MES) and 0.7% agar, pH 5.7]. The plates were kept in the growth room at 22– 24C under a 16-h light/8-h dark cycle. One-week-old seedlings were transferred to soil and kept in a growth chamber at 20C under a 16-h light/8-h dark cycle. The mpk4 mutant used in this study was originally described by Petersen et al. (2000), and seeds were obtained from the ABRC (http://abrc.osu.edu). Heterozygous MPK4+/) plants were selected based on kanamycin resistance and confirmed by genomic PCR, which can amplify both the MPK4 and GUS fragments. Homozygous plants were selected from the progeny of the heterozygous plants based on their dwarf phenotype, and RT-PCR was used to confirm that no MPK4 transcript was present. The PromoterMPK4::HA-MPK4/mpk4 seeds were generated and kindly provided by M. Petersen (Petersen et al., 2000).

RNA extraction and gene cloning Total RNA from different plant tissues was isolated with the RNeasy Plant Mini Kit (Qiagen, http://www.qiagen.com) according to the manufacturer’s instructions. Reverse transcription was performed using a first-strand cDNA synthesis kit (Invitrogen, http://www. invitrogen.com). Genes of interest were amplified from the cDNA and cloned into the pCR8/GW/TOPO vector (Invitrogen). The following primers were used for the cloning. TES-C forward, 5¢-AATGT


904 Qingning Zeng et al. TGTGTCTGCTAATTCAGCC-3¢; TES-C reverse, 5¢-CTAGAGATGCAA CAAGTTGGATATG-3¢; ANP3 forward, 5¢-ATGCAGGATATTCTCGGA TCGG-3¢; ANP3 reverse, 5¢-CTATCCTTTGTGGCCTGATAATGG-3¢. The MPKs and MKKs of interest were cloned into the pCR8 vector earlier (Lee et al., 2008).

Generation of Promoter::GUS lines and histochemical GUS assay A DNA fragment representing the 740-bp region immediately upstream of the AtMKK6 coding region was cloned into the pCAMBIA1381Z vector, and 908-bp upstream of the AtMPK4 coding region was cloned into a modified pUC19 vector and then subcloned into the pZP211 binary vector (Wang et al., 2008b) to generate the corresponding Promoter::GUS lines. The following primers were used to amplify the promoter region of AtMKK6: forward, 5¢-ccggaattcGCTCTCTCTCTCTCTCTCTACAGCGAG-3¢; reverse, 5¢cgcggatccTTTTTTTCTTTGGTTTCTTCCTTGG-3¢; and of AtMPK4: forward, 5¢-cgcgtcgacTCAATCGGTGCTAAGCTA-3¢; reverse, 5¢cgcccatggCGGAGCAAAATTCCTCAC-3¢. Histochemical staining for GUS activity was performed by submerging the inflorescence in 0.5 mg ml)1 X-Gluc (5-bromo-4chloro-3-indolyl-b-D-glucuronic acid, cyclohexylammonium salt) in sodium phosphate buffer (50 mM, pH 7.0 and 0.1% Triton X-100) for 4 h, followed by clearing in 75% ethanol. Images were acquired on a dissecting microscope with mounted camera.

Generation of ProMPK4::MPK4/mpk4 transgenic lines The promoter region of AtMPK4 was amplified and inserted into pUC19 using SalI and NcoI digestion sites as described above. The AtMPK4 CDS was amplified and inserted into the previous vector, using NcoI and ClaI digestion sites. The following primers were used for the cloning of the AtMPK4 CDS: forward, 5¢-cgcCCATG GcgATGTCGGCGGAGAGTTGTTTC-3¢; reverse, 5¢-cgcATCGATTC ACACTGAGTCTTGAGG-3¢. The whole ProMPK4::MPK4 fragment was then enzyme-digested and ligated into the digested pZP211 binary vector, which was transformed into mpk4 mutant plants by floral dip, using numerous plants. T1 generation plants were selected on plates containing both kanamycin (for the mpk4 mutation) and gentamycin (for the transgene). To select for ProMPK4::MPK4/mpk4 plants, genomic DNA was used to amplify the AtMPK4 gene by PCR. Plants with no endogenous genomic AtMPK4 fragment but with the exogenous AtMPK4 CDS were selected for phenotypic analysis.

Staining and microscopy Pollen viability staining was performed according to the method described by Alexander (1969), and the images were acquired on a Leica DM6000 microscope. For Toluidine Blue staining, samples were fixed and then embedded in Spurr’s resin (Spurr, 1969). Semi-thick transverse sections (0.5 lm) of the developing anthers were obtained on an Ultracut instrument (Leica EM UC6) and stained with 0.05% Toluidine Blue in 0.1 M phosphate buffer at pH 6.8 (O’Brien et al., 1964). Scanning electron microscopic images were acquired on a Hitachi S-2600N Variable Pressure Scanning Electron Microscope. For nuclear staining, fresh pollen grains were collected by shaking the opened flowers in 0.1 M sodium phosphate buffer, pH 7.0, collecting the grains by centrifugation, and treating these with DAPI staining buffer (0.1 M sodium phosphate, pH 7.0, 1 mM EDTA, 0.1% Triton X-100, 1.0 lg ml)1 DAPI) (Park et al., 1998). For Aniline Blue staining, flower tissues were fixed in 10% acetic acid in ethanol for 1.5 h and softened by submerging in 1 M NaOH overnight. After washing in 50 mM potassium phosphate buffer (pH 7.5), pollen grains were stained with 0.01% Aniline Blue in the

same phosphate buffer (Besser et al., 2006) and observed using a confocal microscope under UV light. For transmission electron microscopy, WT and mutant flower buds were dissected and anthers were fixed by high-pressure freezing (Kaneda et al., 2008). Samples were then freeze-substituted in 2% osmium tetroxide and 8% dimethoxypropane in acetone from )80 to )20C for 4 days, followed by an overnight substitution at 4C. The temperature was gradually raised to 20C over 2 h, and samples were infiltrated with a gradient series of Spurr’s resin to a final concentration of 100% resin. The infiltrated samples were then polymerized in fresh Spurr’s resin at 60C. Blocks containing anthers were cut to 50-nm thick sections on a Reichert Ultracut E microtome. Slide-mounted sections were stained with uranyl acetate for 6 min and then with lead citrate for 12 min. Images were acquired on a Hitachi H7600 transmission electron microscope.

Recombinant protein production and in vitro kinase assay The CA AtMKK6 was generated by QuickChange site-directed mutagenesis (Stratagene, now Agilent, http://www.genomics.agilent.com) and confirmed by sequencing. Specifically, the conserved S/TXXXXXS/T sites in these MKKs were mutated to DXXXXXE. Native form or mutated genes were cloned into the pGEX 4T-1 or 4T-2 vectors, which express the recombinant proteins with an N-terminal GST tag. Escherichia coli strain BL-21 was transformed with each expression construct and protein production was induced by adding 0.5 mM isopropyl-1-thio-b-D-galactopyranoside (IPTG) to the bacterial cultures at an OD600 of 0.4–0.6, followed by incubation at 30C for 4 h. The recombinant glutathione-S-transferase (GST) fusion proteins were purified from the cell extract using glutathioneconjugated Sepharose 4B. Protein concentrations were determined with the Bio-Rad (http://www.bio-rad.com) protein analysis system using BSA as a standard, and the purity of the isolated proteins was assessed by Coomassie Brilliant Blue (CBB) staining after separation on 10% SDS-PAGE gels. For the in vitro kinase assays, 1 lg purified GST-MPK was incubated with or without 0.5 lg CA MKK6 in 25 ll kinase reaction buffer (50 mM Tris–HCl, pH 7.5, 5 mM b-glycerolphosphate, 2 mM DTT, 10 mM MgCl2, 0.1 mM Na3VO4, 0.1 mM ATP and 5 lCi [c-32P]ATP) for 30 min at 30C. SDS-PAGE sample buffer was added to stop the reaction and boiled to denature the proteins for SDSPAGE analysis. Separated proteins were transferred to an ImmunBlotTM polyvinylidene fluoride (PVDF) membrane, which was used for autoradiography, and then CBB staining.

Yeast two-hybrid assay Yeast two-hybrid assays were conducted with the ProQuestTM yeast two-hybrid system (Invitrogen). cDNAs of interest were recombined from the pCR8 entry vector to either pDEST32 (pDB) or pDEST22 (pAD) vectors, as indicated. Selected pairs of vectors were introduced into the yeast strain MaV203, and co-transformants were selected on synthetic complete drop-out medium without leucine and tryptophan (Sc – LT). Positive interactions were selected on the less stringent Sc – LT medium lacking histidine, but including 25 mM 3-amino-1,2,4-triazole (3AT) plates (Sc – LT – His + 3AT), and on the more stringent Sc – LT medium lacking uracil (Sc – LT – uracil). Quantitative b-galactosidase assays were conducted according to the instruction manual using chlorophenol red b-Dgalactopyranoside (CPRG) as substrate.

BiFC assay A Gateway-compatible BiFC vector system was used in this study (http://www.bio.purdue.edu/people/faculty/gelvin/nsf/


AtMPK4 functions in male meiotic cytokinesis 905 protocols_vectors.htm). cDNAs of interest were recombined from the pCR8 vector to the destination vector containing either nYFP (pE3136) or cYFP (pE3130), as indicated. Each vector (10 lg) was transformed into freshly prepared Arabidopsis mesophyll protoplasts following the method described by Wang et al. (2005). Proteins were allowed to express and interact at room temperature in darkness for 20 h following transformation, after which both lightfield and fluorescence images were taken.

ACKNOWLEDGEMENTS This work was financially supported by grants to BEE and J-GC from the Natural Sciences and Engineering Research Council of Canada. QZ was the recipient of a University of British Columbia University Graduate Fellowship. We thank the Salk Institute Genomic Analysis Laboratory and the Arabidopsis Biological Resource Center for providing the mkk6, mpk4, mpk11 and mpk13 mutants. We are also grateful to Dr X. Li (Department of Botany, UBC) for sharing the snc1 mutant, Dr M. Petersen (Institute of Molecular Biology, Copenhagen University, Denmark) for the PromoterMPK4::HA-MPK4/mpk4 seeds and Dr P.J. Krysan (Genome Center of Wisconsin and Department of Horticulture, University of Wisconsin, USA) for the various anp mutants. We also thank Dr Somrudee Sritubtim for building the ProMKK6::GUS lines.

SUPPORTING INFORMATION Additional Supporting Information may be found in the online version of this article: Figure S1. AtMPK11 and its relationship to AtMPK4. Figure S2. Morphological phenotype of mpk4. Figure S3. Transmission electron micrographs. Figure S4. Identification of mkk6-2 T-DNA insertion mutant. Figure S5. Pollen grain phenotype of several mutants. Figure S6. AtMPK13 T-DNA insertion mutant. Appendix S1. Experimental procedures – additional information. Please note: As a service to our authors and readers, this journal provides supporting information supplied by the authors. Such materials are peer-reviewed and may be re-organized for online delivery, but are not copy-edited or typeset. Technical support issues arising from supporting information (other than missing files) should be addressed to the authors.

REFERENCES Alexander, M.P. (1969) Differential staining of aborted and nonaborted pollen. Stain Technol. 44, 117–122. Andreasson, E., Jenkins, T., Brodersen, P. et al. (2005) The MAP kinase substrate MKS1 is a regulator of plant defense responses. EMBO J. 24, 2579– 2589. Balasubramanian, M.K., Bi, E. and Glotzer, M. (2004) Comparative analysis of cytokinesis in budding yeast, fission yeast and animal cells. Curr. Biol. 14, R806–R818. Beck, M., Komis, G., Muller, J., Menzel, D. and Samaj, J. (2010) Arabidopsis homologs of nucleus- and phragmoplast-localized kinase 2 and 3 and mitogen-activated protein kinase 4 are essential for microtubule organization. Plant Cell, 22, 755–771. Beck, M., Komis, G., Ziemann, A., Menzel, D. and Samaj, J. (2011) Mitogenactivated protein kinase 4 is involved in the regulation of mitotic and cytokinetic microtubule transitions in Arabidopsis thaliana. New Phytol. 189, 1069–1083. Besser, K.v., Frank, A.C., Johnson, M.A. and Preuss, D. (2006) Arabidopsis HAP2 (GCS1) is a sperm-specific gene required for pollen tube guidance and fertilization. Development, 133, 4761–4769. Bo¨gre, L., Calderini, O., Binarova, P. et al. (1999) A MAP kinase is activated late in plant mitosis and becomes localized to the plane of cell division. Plant Cell, 11, 101–113.

Brodersen, P., Petersen, M., Bjorn Nielsen, H., Zhu, S., Newman, M.A., Shokat, K.M., Rietz, S., Parker, J. and Mundy, J. (2006) Arabidopsis MAP kinase 4 regulates salicylic acid- and jasmonic acid/ethylene-dependent responses via EDS1 and PAD4. Plant J. 47, 532–546. Brownfield, L., Hafidh, S., Durbarry, A., Khatab, H., Sidorova, A., Doerner, P. and Twell, D. (2009) Arabidopsis DUO POLLEN3 is a key regulator of male germline development and embryogenesis. Plant Cell, 21, 1940–1956. Calderini, O., Glab, N., Bergounioux, C., Heberle-Bors, E. and Wilson, C. (2001) A novel tobacco mitogen-activated protein (MAP) kinase kinase, NtMEK1, activates the cell cycle-regulated p43Ntf6 MAP kinase. J. Biol. Chem. 276, 18139–18145. 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. Chaiwongsar, S., Otegui, M.S., Jester, P.J., Monson, S.S. and Krysan, P.J. (2006) The protein kinase genes MAP3K epsilon 1 and MAP3K epsilon 2 are required for pollen viability in Arabidopsis thaliana. Plant J. 48, 193–205. Clark, K.A. and Krysan, P.J. (2010) Chromosomal translocations are a common phenomenon in Arabidopsis thaliana T-DNA insertion lines. Plant J. 64, 990–1001. Durbarry, A., Vizir, I. and Twell, D. (2005) Male germ line development in Arabidopsis. duo pollen mutants reveal gametophytic regulators of generative cell cycle progression. Plant Physiol. 137, 297–307. Enns, L.C., Kanaoka, M.M., Torii, K.U., Comai, L., Okada, K. and Cleland, R.E. (2005) Two callose synthases, GSL1 and GSL5, play an essential and redundant role in plant and pollen development and in fertility. Plant Mol. Biol. 58, 333–349. Feys, B., Benedetti, C.E., Penfold, C.N. and Turner, J.G. (1994) Arabidopsis mutants selected for resistance to the phytotoxin coronatine are male sterile, insensitive to methyl jasmonate, and resistant to a bacterial pathogen. Plant Cell, 6, 751–759. Gao, M., Liu, J., Bi, D., Zhang, Z., Cheng, F., Chen, S. and Zhang, Y. (2008) MEKK1, MKK1/MKK2 and MPK4 function together in a mitogen-activated protein kinase cascade to regulate innate immunity in plants. Cell Res. 18, 1190–1198. Hord, C.L., Sun, Y.J., Pillitteri, L.J., Torii, K.U., Wang, H., Zhang, S. and Ma, H. (2008) Regulation of Arabidopsis early anther development by the mitogen-activated protein kinases, MPK3 and MPK6, and the ERECTA and related receptor-like kinases. Mol. Plant, 1, 645–658. Hulskamp, M., Parekh, N.S., Grini, P., Schneitz, K., Zimmermann, I., Lolle, S.J. and Pruitt, R.E. (1997) The STUD gene is required for male-specific cytokinesis after telophase II of meiosis in Arabidopsis thaliana. Dev. Biol. 187, 114–124. Ichimura, K., Mizoguchi, T., Irie, K., Morris, P., Giraudat, J., Matsumoto, K. and Shinozaki, K. (1998) Isolation of ATMEKK1 (a MAP kinase kinase kinase)-interacting proteins and analysis of a MAP kinase cascade in Arabidopsis. Biochem. Biophys. Res. Commun. 253, 532–543. Ichimura, K., Casais, C., Peck, S.C., Shinozaki, K. and Shirasu, K. (2006) MEKK1 is required for MPK4 activation and regulates tissue-specific and temperature-dependent cell death in Arabidopsis. J. Biol. Chem. 281, 36969– 36976. Ishikawa, M., Soyano, T., Nishihama, R. and Machida, Y. (2002) The NPK1 mitogen-activated protein kinase kinase kinase contains a functional nuclear localization signal at the binding site for the NACK1 kinesin-like protein. Plant J. 32, 789–798. Jouannic, S., Hamal, A., Leprince, A.S., Tregear, J.W., Kreis, M. and Henry, Y. (1999) Plant MAP kinase kinase kinases structure, classification and evolution. Gene, 233, 1–11. Ju¨rgens, G. (2005) Cytokinesis in higher plants. Annu. Rev. Plant Biol. 56, 281– 299. Kaneda, M., Rensing, K.H., Wong, J.C., Banno, B., Mansfield, S.D. and Samuels, A.L. (2008) Tracking monolignols during wood development in lodgepole pine. Plant Physiol. 147, 1750–1760. Kosetsu, K., Matsunaga, S., Nakagami, H., Colcombet, J., Sasabe, M., Soyano, T., Takahashi, Y., Hirt, H. and Machida, Y. (2010) The MAP kinase MPK4 is required for cytokinesis in Arabidopsis thaliana. Plant Cell, 22, 3778–3790. Krysan, P.J., Jester, P.J., Gottwald, J.R. and Sussman, M.R. (2002) An Arabidopsis mitogen-activated protein kinase kinase kinase gene family encodes essential positive regulators of cytokinesis. Plant Cell, 14, 1109– 1120.


906 Qingning Zeng et al. Lee, J.S., Huh, K.W., Bhargava, A. and Ellis, B.E. (2008) Comprehensive analysis of protein–protein interactions between Arabidopsis MAPKs and MAPK kinases helps define potential MAPK signalling modules. Plant Signal. Behav. 3, 1037–1041. Li, X., Clarke, J.D., Zhang, Y. and Dong, X. (2001) Activation of an EDS1-mediated R-gene pathway in the snc1 mutant leads to constitutive, NPR1-independent pathogen resistance. Mol. Plant Microbe Interact. 14, 1131–1139. Lin, W.-Y., Matsuoka, D., Sasayama, D. and Nanmori, T. (2010) A splice variant of Arabidopsis mitogen-activated protein kinase and its regulatory function in the MKK6-MPK13 pathway. Plant Sci. 178, 245–250. Ma, H. (2005) Molecular genetic analyses of microsporogenesis and microgametogenesis in flowering plants. Annu. Rev. Plant Biol. 56, 393–434. Melikant, B., Giuliani, C., Halbmayer-Watzina, S., Limmongkon, A., HeberleBors, E. and Wilson, C. (2004) The Arabidopsis thaliana MEK AtMKK6 activates the MAP kinase AtMPK13. FEBS Lett. 576, 5–8. Nakagami, H., Soukupova, H., Schikora, A., Zarsky, V. and Hirt, H. (2006) A Mitogen-activated protein kinase kinase kinase mediates reactive oxygen species homeostasis in Arabidopsis. J. Biol. Chem. 281, 38697–38704. Nakashima, M., Hirano, K., Nakashima, S., Banno, H., Nishihama, R. and Machida, Y. (1998) The expression pattern of the gene for NPK1 protein kinase related to mitogen-activated protein kinase kinase kinase (MAPKKK) in a tobacco plant: correlation with cell proliferation. Plant Cell Physiol. 39, 690–700. Nishihama, R. and Machida, Y. (2000) The MAP kinase cascade that includes MAPKKK-related protein kinase NPK1 controls a mitotic proces in plant cells. Results Probl. Cell Differ. 27, 119–130. Nishihama, R. and Machida, Y. (2001) Expansion of the phragmoplast during plant cytokinesis: a MAPK pathway may MAP it out. Curr. Opin. Plant Biol. 4, 507–512. Nishihama, R., Ishikawa, M., Araki, S., Soyano, T., Asada, T. and Machida, Y. (2001) The NPK1 mitogen-activated protein kinase kinase kinase is a regulator of cell-plate formation in plant cytokinesis. Genes Dev. 15, 352– 363. O’Brien, T.P., Feder, N. and McCully, M.E. (1964) Polychromatic staining of plant cell walls by toluidine blue O. Protoplasma, 59, 368–373. Oh, S.A., Bourdon, V., Das ‘pal, M., Dickinson, H. and Twell, D. (2008) Arabidopsis kinesins HINKEL and TETRASPORE act redundantly to control cell plate expansion during cytokinesis in the male gametophyte. Mol. Plant, 1, 794–799. Otegui, M. and Staehelin, L.A. (2000) Cytokinesis in flowering plants: more than one way to divide a cell. Curr. Opin. Plant Biol. 3, 493–502. Otegui, M.S. and Staehelin, L.A. (2004) Electron tomographic analysis of postmeiotic cytokinesis during pollen development in Arabidopsis thaliana. Planta, 218, 501–515. Park, S.K., Howden, R. and Twell, D. (1998) The Arabidopsis thaliana gametophytic mutation gemini pollen1 disrupts microspore polarity, division asymmetry and pollen cell fate. Development, 125, 3789–3799. Petersen, M., Brodersen, P., Naested, H. et al. (2000) Arabidopsis map kinase 4 negatively regulates systemic acquired resistance. Cell, 103, 1111–1120. Pitzschke, A., Djamei, A., Bitton, F. and Hirt, H. (2009) A major role of the MEKK1-MKK1/2-MPK4 pathway in ROS signalling. Mol. Plant, 2, 120–137. Qiu, J.L., Fiil, B.K., Petersen, K. et al. (2008a) Arabidopsis MAP kinase 4 regulates gene expression through transcription factor release in the nucleus. EMBO J. 27, 2214–2221. Qiu, J.L., Zhou, L., Yun, B.W., Nielsen, H.B., Fiil, B.K., Petersen, K., Mackinlay, J., Loake, G.J., Mundy, J. and Morris, P.C. (2008b) Arabidopsis mitogenactivated protein kinase kinases MKK1 and MKK2 have overlapping functions in defense signaling mediated by MEKK1, MPK4, and MKS1. Plant Physiol. 148, 212–222. 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. Sasabe, M., Soyano, T., Takahashi, Y., Sonobe, S., Igarashi, H., Itoh, T.J., Hidaka, M. and Machida, Y. (2006) Phosphorylation of NtMAP65-1 by a MAP kinase down-regulates its activity of microtubule bundling and

stimulates progression of cytokinesis of tobacco cells. Genes Dev. 20, 1004–1014. 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. Scott, R.J., Spielman, M. and Dickinson, H.G. (2004) Stamen structure and function. Plant Cell, 16(Suppl), S46–S60. Soyano, T., Nishihama, R., Morikiyo, K., Ishikawa, M. and Machida, Y. (2003) NQK1/NtMEK1 is a MAPKK that acts in the NPK1 MAPKKK-mediated MAPK cascade and is required for plant cytokinesis. Genes Dev. 17, 1055–1067. Spielman, M., Preuss, D., Li, F.L., Browne, W.E., Scott, R.J. and Dickinson, H.G. (1997) TETRASPORE is required for male meiotic cytokinesis in Arabidopsis thaliana. Development, 124, 2645–2657. Spurr, A.R. (1969) A low-viscosity epoxy resin embedding medium for electron microscopy. J. Ultrastruct. Res. 26, 31–43. Strompen, G., El Kasmi, F., Richter, S., Lukowitz, W., Assaad, F.F., Jurgens, G. and Mayer, U. (2002) The Arabidopsis HINKEL gene encodes a kinesinrelated protein involved in cytokinesis and is expressed in a cell cycledependent manner. Curr. Biol. 12, 153–158. Suarez-Rodriguez, M.C., Adams-Phillips, L., Liu, Y., Wang, H., Su, S.H., Jester, P.J., Zhang, S., Bent, A.F. and Krysan, P.J. (2007) MEKK1 is required for flg22-induced MPK4 activation in Arabidopsis plants. Plant Physiol. 143, 661–669. Takahashi, Y., Soyano, T., Sasabe, M. and Machida, Y. (2004) A MAP kinase cascade that controls plant cytokinesis. J. Biochem. 136, 127–132. Takahashi, Y., Soyano, T., Kosetsu, K., Sasabe, M. and Machida, Y. (2010) HINKEL kinesin, ANP MAPKKKs and MKK6/ANQ MAPKK, which phosphorylates and activates MPK4 MAPK, constitute a pathway that is required for cytokinesis in Arabidopsis thaliana. Plant Cell Physiol. 51, 1766–1776. Tanaka, H., Ishikawa, M., Kitamura, S., Takahashi, Y., Soyano, T., Machida, C. and Machida, Y. (2004) The AtNACK1/HINKEL and STUD/TETRASPORE/ AtNACK2 genes, which encode functionally redundant kinesins, are essential for cytokinesis in Arabidopsis. Genes Cells, 9, 1199–1211. Twell, D., Park, S.K., Hawkins, T.J., Schubert, D., Schmidt, R., Smertenko, A. and Hussey, P.J. (2002) MOR1/GEM1 has an essential role in the plantspecific cytokinetic phragmoplast. Nat. Cell Biol. 4, 711–714. Wang, S., Tiwari, S.B., Hagen, G. and Guilfoyle, T.J. (2005) AUXIN RESPONSE FACTOR7 restores the expression of auxin-responsive genes in mutant Arabidopsis leaf mesophyll protoplasts. Plant Cell, 17, 1979–1993. Wang, H., Liu, Y., Bruffett, K., Lee, J., Hause, G., Walker, J.C. and Zhang, S. (2008a) Haplo-insufficiency of MPK3 in MPK6 mutant background uncovers a novel function of these two MAPKs in Arabidopsis ovule development. Plant Cell, 20, 602–613. Wang, S., Hubbard, L., Chang, Y., Guo, J., Schiefelbein, J. and Chen, J.G. (2008b) Comprehensive analysis of single-repeat R3 MYB proteins in epidermal cell patterning and their transcriptional regulation in Arabidopsis. BMC Plant Biol. 8, 81. 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. Yang, C.Y., Spielman, M., Coles, J.P., Li, Y., Ghelani, S., Bourdon, V., Brown, R.C., Lemmon, B.E., Scott, R.J. and Dickinson, H.G. (2003a) TETRASPORE encodes a kinesin required for male meiotic cytokinesis in Arabidopsis. Plant J. 34, 229–240. Yang, S.L., Xie, L.F., Mao, H.Z., Puah, C.S., Yang, W.C., Jiang, L., Sundaresan, V. and Ye, D. (2003b) Tapetum determinant1 is required for cell specialization in the Arabidopsis anther. Plant Cell, 15, 2792–2804. Zhao, D. and Ma, H. (2000) Male fertility: a case of enzyme identity. Curr. Biol. 10, R904–R907. 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. Zhou, C., Cai, Z., Guo, Y. and Gan, S. (2009) An arabidopsis mitogen-activated protein kinase cascade, MKK9-MPK6, plays a role in leaf senescence. Plant Physiol. 150, 167–177.
