Present address: Plant Sciences Division, School of Biosciences, University of Nottingham, Sutton Bonington Campus, Loughborough LE12 5RD, UK.
TETRASPORE encodes a kinesin required for male meiotic cytokinesis in Arabidopsis
Article first published online: 11 APR 2003
The Plant Journal
Volume 34, Issue 2, pages 229–240, April 2003
How to Cite
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. (2003), TETRASPORE encodes a kinesin required for male meiotic cytokinesis in Arabidopsis. The Plant Journal, 34: 229–240. doi: 10.1046/j.1365-313X.2003.01713.x
- Issue published online: 11 APR 2003
- Article first published online: 11 APR 2003
- Received 13 November 2002; revised 13 January 2003; accepted 24 January 2003.
A key step in pollen formation is the segregation of the products of male meiosis into a tetrad of microspores, each of which develops into a pollen grain. Separation of microspores does not occur in tetraspore (tes) mutants of Arabidopsis thaliana, owing to the failure of male meiotic cytokinesis. tes mutants thus generate large ‘tetraspores’ containing all the products of a single meiosis. Here, we report the positional cloning of the TES locus and details of the role played by the TES product in male cytokinesis. The predicted TES protein includes an N-terminal domain homologous to kinesin motors and a C-terminus with little similarity to other proteins except for a small number of plant kinesins. These include the Arabidopsis HINKEL protein and NACK1 and two from tobacco (Nishihama et al., 2002), which are involved in microtubule organization during mitotic cytokinesis. Immunocytochemistry shows that the characteristic radial arrays of microtubules associated with male meiotic cytokinesis fail to form in tes mutants. The TES protein therefore is likely to function as a microtubule-associated motor, playing a part either in the formation of the radial arrays that establish spore domains following meiosis, or in maintaining their stability.
Meiosis, a specialized cell division that results in halving of chromosome number and re-assortment of alleles, is essential for sexual reproduction in all plants. The meiotic cycle consists of two rounds of nuclear division (meiosis I and II) following a single round of DNA synthesis so that each diploid meiocyte generates four genetically distinct haploid cells (John, 1990). In seed plants, megasporocytes (in the ovule) or microsporocytes (in the anther) undergo meiosis, giving rise to mega- or microspores, which develop into embryo sacs or pollen grains, respectively. Meiosis also marks the transition from the sporophytic to the gametophytic generation in the life cycle of all plants (Dickinson, 1994).
The plant life cycle involves several different cytokinetic mechanisms, and meiosis in microsporocytes features a unique type of cytokinesis not found elsewhere in plant development. In mitotically dividing cells of the sporophyte, several cytoskeletal arrays are involved in marking the plane of division and mediating cytokinesis (reviewed in Brown and Lemmon, 2001; Heese et al., 1998; Staehelin and Hepler, 1996; Vantard et al., 2000). In G2 phase of the cell cycle, a transient pre-prophase band (PPB) appears as a ring of cortical microtubules (MTs) and actin microfilaments (MFs) that invariably predicts the future division site; during M phase, a bipolar spindle separates the two sets of chromosomes, and at mitotic anaphase, the phragmoplast, a ring of opposing sets of MTs with overlapping plus (fast-growing) ends and co-aligned (but not overlapping) MFs, forms in the spindle midzone. Golgi-derived vesicles containing wall materials are then transported along the phragmoplast MTs (and possibly also MFs) to the site of the phragmoplast, which is gradually replaced by the cell plate, expanding outward towards the parent cell walls to finally separate the daughter cells. The main component of the nascent cell plate is callose (a homogeneous β,1–3 glucan), which is later degraded and replaced by cellulose and its derivatives.
Male meiotic cytokinesis differs from mitotic cytokinesis in several ways, the most striking being that on entering meiosis a microsporocyte deposits callose between its plasma membrane and cellulose wall. Callose is also the major constituent of the intersporal walls formed after meiosis, which are later degraded to release the microspores (Echlin and Godwin, 1968; Stieglitz, 1977). Most monocotyledonous plants undergo successive microsporocyte cytokinesis in which walls are formed between the dyad cells after meiosis I and the microspores of the nascent tetrad after meiosis II. In contrast, most dicotyledons, including Arabidopsis thaliana, undergo simultaneous cytokinesis of the nascent tetrad, which means that no walls are formed until the end of meiosis II. In simultaneous cytokinesis, the intersporal walls first appear as ingrowths from the callose wall surrounding the microsporocyte and then expand centripetally until the microspores are separated (Brown and Lemmon, 1988, 2001; Owen and Makaroff, 1995).
The division plane is also regulated differently in mitosis and male meiosis. In meiotic divisions (as well as in some mitotic divisions in reproductive tissues), no PPB is observed (Wick, 1991). Instead, after each meiotic division in species with successive cytokinesis (Dickinson and Sheldon, 1984), or after meiosis II in those with simultaneous cytokinesis, each microspore nucleus becomes surrounded by a radial array of MTs that partition the surrounding cytoplasm into ‘spore domains’ (Brown and Lemmon, 1988, 2001). Cytokinesis proceeds along the planes defined by the intersection of the arrays, as vesicles contributing membranes and wall components coalesce at the spore domain interfaces.
Previously, we identified the tetraspore (tes) mutant of Arabidopsis in a screen for plants with defects in male meiosis (Spielman et al., 1997). In tes mutants, and also stud (std) mutants described by Hülskamp et al. (1997), the nuclear divisions of male meiosis occur normally, but male meiotic cytokinesis is disrupted while other cell divisions appear to be unaffected. In plants homozygous for strong mutant alleles (tes-1, tes-3, and tes-4, and stud-1–3), microsporocytes secrete a callose wall and undergo the nuclear divisions of meiosis, but no membranes or walls of any type form between the microspores after meiosis II. In the weak allele tes-2, partial walls form but these are insufficient to separate the microspores (Spielman et al., 1997). In tes and std mutants, failure of male meiotic cytokinesis causes all four meiotic products to begin male gametophytic development in a common cytoplasm. Some of these ‘tetraspores’ nevertheless form functional pollen grains, although with many abnormalities including polyploid sperm. TES and STD were both mapped to similar intervals on chromosome 3. Here, we report the cloning of the TES/STD locus, which encodes a protein with a putative kinesin motor domain.
Several steps of plant cell division involve proteins in the kinesin superfamily (kinesin structure and function are reviewed in Endow, 1999; Goldstein and Philp, 1999; Lawrence et al., 2002; Reddy, 2001; Sack et al., 1999; Vale and Fletterick, 1997). Kinesins conventionally consist of a tetramer of two heavy chains and two light chains; the heavy chain contains an N-terminal motor domain with a catalytic core that binds MTs and hydrolyses ATP to generate force (and usually movement along the MT) and a neck domain that affects the direction of movement. There is also a central stalk region of α-helical coiled coils involved in dimerization, and a C-terminal region, sometimes called a tail domain, that binds the light chains and interacts with cargo. Many proteins with kinesin motor domains have been found in eukaroytes; these are referred to as ‘kinesin-related proteins’ or simply ‘kinesins’, and are classified into subfamilies according to sequence and position of the motor (Kim and Endow, 2000). Most kinesins have an N-terminal motor and move towards the plus ends of MTs, but motors may be C-terminal (in minus-end directed kinesins) or internal. Kinesins perform functions including transport of vesicles, organelles, or chromosomes along MTs, cross-linking and antiparallel sliding of MT arrays, and, possibly, mediation of MT polymerization and depolymerization. Cell division-related functions attributed to kinesins include spindle stability and elongation, vesicle transport to the site of division, and chromosome movements.
An analysis of the complete Arabidopsis genome sequence identified 61 kinesins by homology to the motor domain (Reddy and Day, 2001). We found that one of these, AtF7K15.60, is the TES gene. We also confirmed that TES and STD are the same locus. The predicted TES protein includes an N-terminal domain with homology to kinesin motors and a C-terminal region with homology only to a small number of other plant kinesins, including the Arabidopsis HINKEL (HIK) protein (Strompen et al., 2002) and the NACK1 and NACK2 kinesins from tobacco (Nishihama et al. 2002), which are involved in MT organization during mitotic cytokinesis. Immunolocalization of MTs in wild-type and tes microsporocytes undergoing meiosis shows that tes mutants fail to form the radial MT arrays associated with cytokinesis of tetrads. TES thus encodes a kinesin required to establish the MT-based spore domains that are essential for male meiotic cytokinesis in Arabidopsis.
Positional cloning of TES
To refine the map position of TES, we generated a mapping population from the cross tes-1/tes-1 (Col-3) ×TES/TES (Ler) and identified 958 homozygous mutants in the F2 generation. Using a series of CAPS markers (Konieczny and Ausubel, 1993), we mapped TES to an interval spanned by two overlapping bacterial artificial chromosomes (BACs) covering 204 kb, flanked on the centromeric side by the T7 end of BAC F7K15 and the telomeric side by the T7 end of BAC T5C2 (Figure 1). We could not detect any further recombinants using markers within these two BACs (data not shown).
Annotation of the genomic sequence released by the Munich Information Center for Protein Sequences (MIPS) (http://mips.gsf.de/proj/thal/) identified a predicted kinesin on BAC F7K15 (F7K15.60 = At3g43210). Because the tes mutant phenotype suggests that TES may be involved in MT organization or vesicle trafficking (Spielman et al., 1997), and kinesins have roles in both of these processes (Goldstein and Philp, 1999), this indicated that F7K15.60 could be the TES gene. As a preliminary assay for a mutation at this locus, genomic DNA was amplified by PCR from tes-1 (Col-3), TES (Col-3), tes-4 (Ws2), and TES (Ws2) alleles and digested with a selection of restriction enzymes, and the banding pattern compared on agarose gels. Digestion with DdeI showed band shifts in both mutant tes alleles with respect to the corresponding wild-type fragments (data not shown). Sequencing the F7K15.60 locus in tes-1–4 and std-1, along with their respective ecotypes, showed base changes or deletions in all five mutant alleles (see Experimental procedures for allelism of tes and std mutants, and also the Section ‘Sequence of wild-type and mutant TES alleles’, below).
Complementation of tes-1 mutants with the candidate TES gene
To test whether F7K15.60 is the TES gene, a 7180 bp fragment of genomic DNA containing the predicted coding and regulatory sequences was transformed into tes-1 homozygotes. Kanamycin-resistant progeny of transformed mutants had a wild-type phenotype with respect to pollen size and seed set. Self-seeds collected from these plants were then grown without kanamycin selection and scored for (i) pollen size and (ii) presence of the mutant tes-1 and wild-type TES sequences, as determined by a CAPS marker sensitive to a 7 bp deletion in the tes-1 allele (see Experimental procedures). This assay showed co-segregation of the wild-type phenotype with presence of the transgene; of 24 plants scored in a single family, eight had the mutant phenotype and the mutant sequence only, while the others had a wild-type phenotype and both mutant and wild-type sequences (data not shown). Plants in this family with normal-sized pollen also had tetrads with cross-walls (showing rescue of the primary defect caused by the tes mutation), pollen with two sperms and one vegetative nucleus (the normal number), and normal seed set (Figure 2a,c,e), while plants with big pollen had tetrads with no cross-walls, extra pollen nuclei, and a high proportion of aborted seeds (Figure 2b,d,f). Therefore, the F7K15.60 construct complemented the tes-1 mutation and rescued its indirect as well as direct effects.
Sequence of wild-type and mutant TES alleles
Genomic DNA was sequenced from the five mutant and three wild-type alleles, and partial cDNAs (consisting of the predicted coding region) sequenced from tes-1 and tes-2 (both in Col-3), and wild-type Col-3. The genomic sequence of Col-3 is identical to the published sequence for Col-0; all references to the TES sequence hereafter indicate the Col-0/Col-3 nucleotide sequence unless otherwise stated. The transcriptional start and stop sites were investigated with 5′ and 3′ rapid amplification of cDNA ends (RACE). The 3′ untranslated region (UTR) consists of 147 nucleotides (not shown). The 5′ UTR of the TES mRNA transcript spans 498 nucleotides, including a 266 nucleotide intron extending from −272 to −7 deduced by comparison of the genomic and cDNA sequences (Figure 3a).
There are two adjacent start codons at the junction of the 5′ UTR and the coding sequence (Figure 3a). Normally, initiation of translation occurs at the first AUG in the mRNA transcript; however, in vertebrate mRNAs, it has been found that the first start codon may be skipped if it is not in the correct sequence context, with an A or U 3 bases upstream of the AUG and a G immediately following it being particularly important (reviewed in Kozak, 1999). Because the second but not first AUG in the TES mRNA occurs in the consensus context (Figure 3a), the Arabidopsis Genome Initiative (AGI) prediction that the F7K15.60 protein begins with a single methionine seems likely to be correct. However, sequencing of the TES protein would be required to confirm its start site.
Assuming that translation begins at the second start codon, the TES coding region consists of 4590 nucleotides spanning 14 exons (Figure 3b). The sequenced cDNA within the predicted coding region is 2814 nucleotides, slightly longer than the sequence predicted by AGI. This is because of an error in the predicted junction of intron 12/exon 13: we found that the cDNA includes 15 nucleotides 5′ to the predicted splice site. Based on the sequenced cDNA, the TES locus is predicted to encode a kinesin of 937 amino acids. The protein includes an N-terminal sequence with homology to the kinesin motor domain and a C-terminal region with two coiled coil domains; these results are presented in more detail below.
The wild-type Ler and Ws2 genomic sequences at the TES locus are unusually divergent from the Col sequence, but strikingly, they are completely identical to each other. In this 4570 bp sequence, we found 79 nucleotide substitutions and 14 insertions or deletions (indels) of one to eight nucleotides in the Ler/Ws2 sequence with respect to Col. Changes in the genomic sequence of the mutant alleles with respect to the Col or Ler/Ws2 sequence, as appropriate, and the resulting changes to the predicted proteins, are shown in Table 1. tes-2, tes-3, and std-1, all from ethyl methyl sulphonate (EMS) mutagenized populations (Hülskamp et al., 1997; Spielman et al., 1997), have single nucleotide substitutions, while tes-1 (fast neutron) and tes-4 (T-DNA) have deletions of 7 and 10 nucleotides, respectively (we previously found that the tes-4 mutant phenotype did not co-segregate with the T-DNA present in its family, suggesting there was no insertion in the TES gene; unpublished results). All mutations are in exons (Figure 3b), and only the tes-2 mutation, which causes a partial loss-of-function phenotype (Spielman et al., 1997), occurs in the kinesin motor domain.
|tes-1||7nt del 4508–14||Frameshift from 911 STOP 923|
|tes-2||g>a 1840||R>K 306|
|tes-3||c>t 3575||STOP 663|
|tes-4||10 nt del 3343–52||Frameshift from 585 STOP 600|
|std-1||c>t 3287||STOP 567|
TES encodes a kinesin with an N-terminal motor domain
The TES protein is predicted to contain a kinesin motor domain approximately from residue 26 to 355; this region includes consensus sequences for nucleotide and MT binding (Figure 4). At the 3′ end of the motor domain is a region with similarities to the neck of ‘orphan’ kinesins (Vale and Fletterick, 1997). Outside of the motor domain, there are two predicted coiled coil domains at 385–434 and 691–720.
blastp and tblastn searches (Altschul et al., 1997) (http://www.ncbi.nlm.nih.gov/BLAST/) were performed against the GenBank and EMBL databases. As the motor domain is well conserved among kinesins (Vale and Fletterick, 1997), the catalytic core and a possible neck domain of the motor were omitted from the TES sequence used in the search (see Figure 4). The most closely related proteins to TES are NACK1 and NACK2 from tobacco (Nishihama et al., 2002). NACK2 has an identity of 60% and similarity of 78% to TES outside the motor domain (64 and 80% overall), while NACK1 is 53% identical and 70% similar outside the motor domain (58 and 74% overall). Of the predicted 61 kinesins in the Arabidopsis genome (Reddy and Day, 2001), the protein encoded by F15H18.12 (At1g18370) on chromosome 1 is most similar to TES, with an identity of 52 and similarity of 69% outside the motor domain (57 and 73% overall). This kinesin has recently been identified as HIK (Strompen et al., 2002). An alignment of the predicted TES, NACK2, and HIK proteins is shown in Figure 4. Other conceptual proteins with homology to TES are a kinesin predicted from rice genomic DNA, BAB32972.1 (Sasaki et al., 2000), with identity and similarity outside the motor of 48 and 68%, respectively (56 and 74% overall), and two predicted from maize partial cDNAs, KIN1 and KIN2 (Lawrence et al., 2002), with identity/similarity scores outside the motor of 31/49% and 46/66%, respectively (part of the motor domain is missing in each of these proteins, making overall comparison impossible).
TES is widely expressed but enriched in flower buds
RNA blot analysis shows that TES transcripts are present in many parts of the plant but expression levels are variable (Figure 5). TES mRNA is most abundant in inflorescences containing buds undergoing male meiosis. There is less transcript in older buds, open flowers, and 2–3-week-old plants (both aerial portions and roots). Expression is barely detectable in fully expanded leaves and in siliques 3 to 5 days after flowering.
To further investigate the TES expression pattern, wild-type plants were transformed with the TES upstream and partial coding regions fused to a β-glucuronidase (GUS) reporter gene (see Experimental procedures). GUS expression was detected in many organs containing dividing cells, including inflorescence and floral meristems (Figure 6a), floral buds, vegetative meristems, young leaves, root tips, lateral root primordia, and embryos and endosperm in developing seeds (not shown). Expression was seen in developing pollen, including tetrads (Figure 6b), but after the uninucleate microspore stage it began to fade, becoming indetectable in mature anthers (Figure 6c).
Radial MT systems are not formed in tes-1 mutant tetrads
Male meiotic cytokinesis in flowering plants involves radial MT systems (RMS) emanating from each microspore nucleus in the nascent tetrad (Brown and Lemmon, 1988, 2001). The failure of this cytokinesis in tes mutants, along with the discovery that TES encodes a kinesin, suggests that RMS in tes mutants could be defective, as some kinesins are involved in MT organization (Goldstein and Philp, 1999). Indirect immunofluorescence was used to localize MTs during and after tetrad formation in wild-type (Figure 7a,c,e) and tes-1 mutant anthers (Figure 7b,d,f). Despite the fact that spindle formation and chromosome segregation during male meiosis I and II were indistinguishable in wild-type and tes-1 microsporocytes (data not shown), organized RMS were never seen in tes-1 mutant tetrads. The MTs of the RMS define discrete cytoplasmic domains in wild-type tetrads and remain in position as intersporal walls form (Figure 7c). In tes-1 tetrads, these MTs clearly fail to establish stable boundaries for the cytoplasmic domains, but instead accumulate around the nuclear surfaces (Figure 7d). Finally, these MT arrays collapse into a tangled mass in the undivided cytoplasm of multinucleate microspores. Strikingly, this collapse is accompanied by loss of nuclear spacing (Figure 7f).
TES encodes a plant-specific kinesin required for male meiotic cytokinesis
The predicted TES protein (Figure 4) contains the nucleotide and MT-binding motifs diagnostic of kinesin motors (Kull et al., 1996; Sablin et al., 1996; Woehlke et al., 1997; http://www.proweb.org/kinesin/Logos.html). There is also a region with similarity to the neck domains of ‘orphan’ kinesin motors, such as centromere-associated protein E (CENP-E) (Vale and Fletterick, 1997). The N-terminal position of the motor domain indicates that TES is likely to be directed towards the plus ends of MTs (Vale and Fletterick, 1997). The coiled coil motifs C-terminal to the motor suggest that the TES protein may form homo- or heterodimers, like other kinesins (Vale and Fletterick, 1997).
At least 40 genes and proteins are known to be involved (directly or indirectly) in cytokinesis of plant cells (reviewed in Nacry et al., 2000). Kinesins are involved in several steps in cell division in all eukaryotes, and roles in somatic cytokinesis have been found for several plant kinesins (Goldstein and Philp, 1999; Reddy, 2001). Among these are NACK1 and NACK2 from tobacco, the most closely related kinesins to TES. NACK1 protein accumulates at the phragmoplast equator, and mitotic cytokinesis is disrupted by overexpression of a dominant negative mutant NACK1 cDNA or by simultaneous repression of NACK1 and NACK2 gene expression (Nishihama et al., 2002). HIK (F15H18.12), the Arabidopsis kinesin with the highest similarity to TES, is also required for cytokinesis. The hik mutant embryos display incomplete and misoriented cell walls and die as seedlings. Immunolocalization of tubulin in mutant cells shows unusual persistence of phragmoplast MTs, suggesting that HIK function is required for reorganization of these MTs during lateral expansion of the cell plate (Strompen et al., 2002). Several more distantly related kinesins, including AtPAKRP1, AtPAKRP2, and KCBP (ZWICHEL (ZWI)) in Arabidopsis, TKRP125 in tobacco, and DcKRP120-2 in carrot have roles in cytokinesis; immunolocalization and interference by antibodies suggest various functions for these, including crosslinking and sliding of phragmoplast MTs and vesicle transport to the cell plate (Asada et al., 1997; Barroso et al., 2000; Lee and Liu, 2000; Lee et al., 2001; Vos et al., 2000).
TES is required for establishment or maintenance of the RMS involved in male meiotic cytokinesis, and maybe plays a role in the sporophyte
In Arabidopsis, as in most dicots, the spores produced by male meiosis are divided in the nascent tetrad by simultaneous cytokinesis. It has been shown for Arabidopsis as well as other dicot species that simultaneous cytokinesis involves RMS emanating from the microspore nuclei, which mark the boundaries of the microspore cytoplasts and determine the planes of division (Brown and Lemmon, 1988, 2001). In tes-1 mutant tetrads, RMS are not formed; instead the MTs become increasingly disorganized as the undivided tetrad matures into a multinucleate microspore (Figure 7). This indicates that the TES product is required, directly or indirectly, for establishment or maintenance of the RMS involved in male meiotic cytokinesis.
Although the tes mutant phenotype indicates a role specific to cytokinesis in male tetrads, the observation that TES is expressed in many tissues (Figures 5 and 6) raises the possibility that the TES kinesin has additional functions not revealed by the mutants studied so far. RNA blot analysis (Figure 5), TES::GUS reporter activity (Figure 6), and RT-PCR (not shown) all indicate that TES is expressed throughout the plant life cycle, especially in dividing cells. Similar discrepancies between mutant phenotypes and expression domains are seen for other Arabidopsis kinesins. ATK1 is expressed in vegetative tissues but the presumed null atk1-1 mutation only affects spindle function in male meiosis (Chen et al., 2002), and the ZWI (KCBP) kinesin is also expressed in many parts of the plant, but zwi mutants only show defects in trichome morphology (Oppenheimer et al., 1997). Given the large number of kinesins encoded in the Arabidopsis genome (Reddy and Day, 2001), it is possible that there is extensive redundancy among these proteins, so that mutation only uncovers part of each kinesin's function. There is also precedent in other eukaryotes for redundancy among kinesins involved in cell division (reviewed in Goldstein and Philp, 1999). In this connection, it is significant that we may not have yet investigated a null allele of TES. Of the five alleles described here, only tes-2, which appears to retain partial TES function (Spielman et al., 1997), has a mutation in the motor domain (although it should be noted that the tes-3, tes-4, and std-1 mutations all introduce a stop codon in or near the middle third of the protein; Figure 3b and Table 1).
The tes-3, tes-4, and std-1 mutations produce a severely truncated protein, but the tes-1 mutation introduces a frameshift only 26 residues from the end. This could interfere with cargo-binding function, which is thought to reside in the kinesin tail (Vale and Fletterick, 1997). The tes-2 mutation causes a change at residue 306 from arginine (R) to lysine. R306 in TES falls within a conserved region of the motor domain involved in MT binding, and directed mutation of the homologous residue in human kinesin heavy chain (R284) decreases affinity for MTs (Woehlke et al., 1997). Therefore, the tes-2 mutation is likely to interfere with the MT-binding ability of the protein.
Phylogenetic relationships of the TES kinesin
Database searches using the TES protein without its motor domain reveal strong similarities to very few proteins: only two kinesins from tobacco (NACK1 and NACK2, Nishihama et al., 2002), one kinesin from Arabidopsis (HIK; Strompen et al., 2002), one from rice (BAB32972.1; Sasaki et al., 2000), and two from maize (KIN1 and KIN2; Lawrence et al., 2002) show more than 30% similarity to TES over the C-terminal region. With the exception of NACK2, the other kinesins are more similar to HIK than they are to TES. Our findings are in good agreement with a phylogeny of all 61 kinesins in Arabidopsis based on the motor domains (Reddy and Day, 2001), which places TES (F7K15.60) closest to HIK (F15H18.12), within a group of eight kinesins that do not fall into any recognized subfamily. Based on alignments of complete protein sequences, TES has also been placed with HIK as the only Arabidopsis kinesin in a small ‘HIK subgroup’, which contains just plant kinesins (Strompen et al., 2002). Therefore, TES and a small group of closely related proteins represent a novel group of plant-specific kinesins, which may have a role in MT organization during cytokinesis.
The TES locus is highly polymorphic between Col and Ler/Ws2 ecotypes
The number of polymorphisms between the Col and Ler/Ws2 alleles of TES (79 nucleotide substitutions and 14 indels in the 4590 bp coding region, giving an average of one change every 49 bases) is striking in light of the finding of Konieczny and Ausubel (1993) that Col and Ler differ on an average by one change every 261 bases, which may even be an overestimate of polymorphism levels (Bergelson et al., 1998). However, it is not unprecedented: PAT1, a gene involved in tryptophan biosynthesis, shows even more divergence between Col and Ler alleles, with 90 nucleotide differences and 12 indels from 1 to 81 bp in a 2627 bp interval, an average of one change every 26 bases (Rose et al., 1997).
TES function in other species
Mutants with defective male meiotic cytokinesis have been reported in other angiosperm species. Both the ms1 mutant of soybean (Albertsen and Palmer, 1979) and the jumbo pollen (jp) mutant of alfalfa (McCoy and Smith, 1983) fail to undergo cytokinesis in tetrads, resulting in the production of coenocytic microspores with extra nuclei. The similar mutant phenotypes suggest that MS1 and JP could be homologous to TES; to our knowledge neither have been cloned.Tavoletti et al. (2000) reported that the jumbo pollen trait caused by the jp mutation could be separated from the multinucleate microspore trait in segregating populations. However, our finding that complementation of tes mutants with the TES locus alone restores the normal number of pollen nuclei as well as normal pollen size (Figure 2) confirms that the tes mutation is the cause of multinucleation as well as big pollen size.
Zea mays (maize) and Oryza sativa (rice), like most other monocots, have successive male meiotic cytokinesis: walls are formed between meiotic products at the dyad as well as the tetrad stage. In both these species, radial microtubule arrays are observed after the nuclear divisions of meiosis I as well as meiosis II (Steiger and Cande, 1990; Xu and Ye, 1998). The predicted kinesins with a strong similarity to TES in the maize and rice genomes could, therefore, also be involved in male meiotic cytokinesis. However, as the maize and rice kinesins are more similar to HIK than they are to TES (data not shown), they may be involved in mitotic cytokinesis. It is not yet known whether hik mutations would affect meiotic cytokinesis because mutant plants die as seedlings (Strompen et al., 2002).
Plant materials and growth conditions
tes-1 to tes-4 mutants and growth conditions for plants are described in Spielman et al. (1997). std-1 is described in Hülskamp et al. (1997). Wild-type Col-0, Ler and Ws2 seeds were obtained from the Nottingham Arabidopsis Stock Centre (NASC, Nottingham, UK), and Col-3 seeds from Lehle Seeds (TX, USA).
Allelism of TES and STD
Allelism tests between tes and std mutants were conducted as the mutations were reported to cause similar phenotypes and both mapped to the same interval on chromosome 3 (Hülskamp et al., 1997; Spielman et al., 1997). std-1 homozygotes were crossed as seed parents with tes-4 heterozygotes as pollen parents (using a heterozygous pollen donor greatly increases the proportion of viable seed); of 24 F1 plants resulting from the cross, 13 had the tes/std big pollen phenotype. Crosses between std-1 and tes-1 homozygotes in both directions yielded only seven plants, and all of these had the big pollen phenotype. Both sets of crosses indicated that the TES and STD genes were the same.
Positional cloning of TES
F2 plants from the mapping cross tes-1/tes-1 (Col-3) ×TES/TES (Ler) were scored for the tes mutant phenotype, and DNA was extracted using the protocol described in Dean et al. (1992). CAPS markers (Konieczny and Ausubel, 1993) were used to distinguish Col-3 from Ler DNA. Information on the AtPox and T21P20 markers is available at http://www.arabidopsis.org. New CAPS markers generated during the mapping of TES (see Figure 1) are available on request.
To test for polymorphisms between mutant and wild-type alleles at the F7K15.60 locus (see Results), a 1.7 kb region (F7K15-60F2) was amplified using forward primer 5′-GGATCAAGTGAGTGTGAAC-3′ and reverse primer 5′-CTAGAGATGCAACAAGTTGG-3′, digested with a selection of restriction enzymes, and separated on a 2.8% agarose gel; polymorphisms were detected after digestion with DdeI.
The F7K15-60F2 PCR product (see above) was used to screen a binary cosmid (BC) library of the Arabidopsis genome by the Gene Transfer Clone Identification and Distribution (GeTCID) service provided by the UK Genomic Arabidopsis Resource Network (GARNET, York, UK). The probe hybridized to four clones. The putative TES coding region was amplified from one of these, BC clone 66E8, and also from IGF (Institut fuer Genbiologische Forschung) BAC F7K15 (obtained from the Arabidopsis Biological Resource Center (ABRC, Ohio, USA)), using forward primer 5′-ATGGGACCTCCGAGAACTCCG-3′ and reverse primer, as was done for F7K15-60F2 mentioned above. The PCR products were digested with restriction enzymes and the banding patterns compared to confirm that BC clone 66E8 contained the entire putative TES open reading frame (ORF). DNA was extracted from clone 66E8 using the method provided by the GeTCID service, and digested with PacI and SacI to release an 8355 bp fragment including the putative TES coding sequence. The digested cosmid DNA was separated on a 0.6% agarose gel and the band of interest was purified using a QIAquick Gel Extraction Kit (Qiagen, Crawley, W. Sussex, UK), cloned between the PacI and SacI sites of puCAP (van Engelen et al., 1995), and transformed into Escherichia coli strain DH5α. The insert contained a small region of the predicted ORF downstream of the putative TES locus, which was excised from the vector by digestion with XbaI followed by ligation with T4 ligase (Promega, Southampton, UK). This produced a puCAP vector containing a 7180 bp insert consisting of the 4590 bp putative TES coding region, 2180 bp of upstream sequence, and 410 bp of downstream sequence. The 7180 bp fragment was released from the vector by digestion with PacI and AscI, and cloned between the PacI and AscI sites of pBINPLUS (van Engelen et al., 1995) which was then transformed into Agrobacterium tumefaciens strain GV3101 by electroporation (Shen and Forde, 1989). Agrobacterium tumefaciens was transformed into tes-1/tes-1 homozygous mutants using the floral dip method (Clough and Bent, 1998), and seeds were collected from self-pollinated plants and sown on germination medium containing 1% sucrose and 100 µg ml−1 kanamycin for selection of transformants. Kanamycin-resistant plants were scored for rescue of the mutant phenotype as described in Results.
For detection of the tes-1 mutant and wild-type TES alleles, as described, DNA was extracted from plants as for mapping, and a 490 bp product was amplified by PCR using forward primer 5′-CGAACAGCACTTGGCTGAGC-3′ and reverse primer 5′-GAGATGCAACAAGTTGGATATG-3′ with an annealing temperature of 57°C, followed by digestion with SacI and separation on a 2% agarose gel. The wild-type product was cut into two fragments of approximately 410 and 80 bp, but the tes-1 product was uncut because of deletion of the restriction site by the tes-1 mutation.
For comparison of tetrads in rescued and non-rescued plants, inflorescences were fixed in 3 : 1 ethanol:acetic acid, transferred to 70, 85, and 95% ethanol, and embedded in JB-4 glycol methacrylate resin (Polysciences, Warrington, PA, USA). Five micrometer-sections were cut in ribbons using glass knives made from microscope slides on a Microm HM 335E rotary microtome, according to Ruzin (1999), and stained in a mix containing three parts of 2.5 µg ml−1 DAPI in Vectashield antifade solution (Vector Laboratories, Peterborough, UK) and one part of decolorized 0.1% aniline blue in 0.1 m K3PO4·H2O. Specimens were mounted under coverslips and viewed with a Zeiss Axiophot microscope (Oberkochen, Germany) using a 50 W mercury lamp and 365 nm excitation, 395 nm dichroic, 420 mm long-pass emission filters. Mature pollen grains from rescued and non-rescued plants were fixed in 3 : 1 ethanol:acetic acid, transferred to 70% ethanol, stained in the DAPI/Vectashield solution alone, and viewed as above.
For genomic and cDNA sequences, a series of overlapping 2 kb fragments was amplified from the appropriate template, purified using the QIAquick Gel Extraction Kit (Qiagen), and sequenced using primers approximately 600 bp apart. The primers were designed using the published Col-0 F7K15 sequence (http://www.arabidopsis.org).
3′ and 5′ rapid amplification of cDNA ends (RACE)
Total RNA was prepared from flower buds of Col-0 plants with an RNeasy Plant Mini Kit according to the manufacturer's protocol (Qiagen).
For 5′ RACE, cDNA was synthesized with a gene-specific primer TES 51 (5′-TGCAACGTCCCTAGAACCAC-3′), purified and dA-tailed as described in the 5′/3′ RACE kit instruction manual (Roche, Welwyn Garden City, UK). 5′ RACE cDNA was amplified using oligo d(T)-anchor primer (5′-GACCACGCGTATCGATGTCGACTTTTTTTTTTTTTTTTV-3′) and the second gene-specific primer TES 52 (5′-ACGATAGTCTCATCATCGGGA-3′); 1 µl of the first PCR reaction was used in a second round of PCR using the anchor primer (5′-GACCACGCGTATCGATGTCGAC-3′) and a third gene-specific primer TES 53 (5′-TCTCATCCGAACAGTGACAAG-3′). The amplified products were cloned into pGEM-T Easy (Promega) and sequenced.
3′ RACE cDNA was synthesized from inflorescence RNA using the oligo d(T)-anchor primer (5′-GACCACGCGTATCGATGTCGACTTTTTTTTTTTTTTTTV-3′) according to the manufacturer's instructions. 3′ RACE products were amplified using the gene-specific primer TES 31 (5′-CCGCCGGGTGGGAAAAAAACAGAG-3′) and the PCR anchor primer (5′-GACCACGCGTATCGATGTCGAC-3′); 1 µl of this PCR reaction was used in a second round of PCR with gene-specific primer TES 32 (5′-CACAACATGCAACACGTAAGA-3′) and the PCR anchor primer; the PCR products were cloned into pGEM-T Easy and sequenced.
RNA blot analysis
RNA blot analysis was carried out according to Sambrook et al. (1989) except where otherwise stated. Total RNA was extracted from various tissues of wild-type plants using the RNeasy Plant Mini Kit (Qiagen), loaded at 15 µg lane−1 on a 1% agarose gel, separated by electrophoresis, and transferred onto GeneScreen Plus Hybridization Transfer Membranes (PerkinElmer, Cambridge, UK) using the manufacturer's protocol. A TES probe was prepared by amplifying an 884 bp region from cDNA using the F7K15-60F2 forward primer as before and reverse primer 5′-GTACTCCGAAGCAGCGCCTGTCTCG-3′. A GAPDH probe (Mizukami and Fischer, 2000) was prepared as a loading control. Probes were labeled with 32P-dCTP (ICN, Basingstoke, UK) using the Rediprime II random prime labeling system (Amersham Pharmacia, Little Chalfont, Buckinghamshire, UK) according to the manufacturer's instructions. Membranes were pre-hybridized for 3 h and hybridized for 2 h at 65°C in Rapid-hyb buffer (Amersham Pharmacia) and washed according to the manufacturer's instructions, and exposed on Kodak X-OMAT-AR film (Kodak, Hemel Hempstead, UK) at −80°C for 4 days (TES probe) or 4 h (GAPDH probe). Membrane stripping was carried out according to the Rapid-hyb instructions.
TES::GUS reporter construct and analysis
A region consisting of 2.7 kb of sequence upstream of the TES translation initiation site followed by the first 7 exons and 6 introns of the coding region was amplified from Col-0 genomic DNA by PCR using forward primer 5′-AAACTCGAGAAATCACCGAAGATGCAGA-3′ and reverse primer 5′-AAAGGTACCTGCTATTTCCCTGAGACTGCT-3′. The amplified fragment was cloned into the pGEM-T vector (Promega), excised by digestion with XhoI and KpnI, and subcloned into the XhoI and KpnI sites of pBJ60 (gift of Bart Janssen, Horticulture and Food Research Institute of New Zealand) to produce plasmid pTESI6::GUS. The TESI6::GUS expression cassette was excised from pBJ60 as a NotI fragment and subcloned into the NotI site of the binary vector pBJ40. This vector was then introduced into A. tumefaciens strain GV3101 carrying the pAD1289 plasmid conferring overexpression of VirG (Hansen et al., 1994) by electroporation (Shen and Forde, 1989), and the bacteria transformed into Arabidopsis by the floral dipping method (Clough and Bent, 1998). Seeds were collected from self-pollinated plants and sown on germination media containing 1% sucrose and 50 µg ml−1 kanamycin for selection of transformants. Kanamycin-resistant plants were analysed for GUS activity: samples were incubated in 90% acetone on ice for 30 min, washed in Tris–NaCl buffer (100 mm Tris–HCl pH 7.2, 50 mm NaCl), vacuum infiltrated with Tris–NaCl buffer containing 0.1% Triton X-100, 2 mm each of K3Fe(CN6) and K4Fe(CN)6, 2 mm 5-bromo-4-chloro-3-indolyl-beta-d-glucuronic acid, cyclohexylammonium salt (Melford Laboratories, Ipswich, Suffolk, UK), and 100 µg ml−1 chloramphenicol, and incubated at 37°C overnight. Stained samples were dehydrated to 95% ethanol through a graded ethanol series, with a fixation step in FAA (3.7% formaldehyde, 50% ethanol, 5% acetic acid) between 50 and 70% ethanol, and embedded and sectioned as above. Sections were mounted under coverslips in DPX (Sigma) and viewed on a Zeiss Axiophot microscope using dark field illumination.
Immunolocalization of tubulin
Plant materials were prepared from tes-1/tes-1 and wild-type Col-3 flower buds according to methods published by Brown and Lemmon (1995). Coverslips with stained materials were mounted on a microscope slide in Prolong brand antifade reagent (Molecular Probes) and fluorescence was examined using a Bio-Rad 1024 confocal Kr/Ar laser scanning microscope. Illustrations were prepared by transferring Bio-Rad files to a Macintosh PowerPC using nih Image software.
We gratefully acknowledge the following: Martin Hülskamp (University of Cologne, Germany) for std1–1 seeds; Daphne Preuss and Greg Copenhaver (University of Chicago, USA) for sharing unpublished CAPS markers; Rachel Hodge (University of Leicester, UK) for help with mapping; Anuj Bhatt (University of Oxford, UK) for patient advice on many intellectual and technical issues; W.E. Friedman (University of Colorado, Boulder, USA) for teaching the resin-sectioning techniques; the Arabidopsis Biological Resource Center (Ohio, USA) and the Institut fuer Genbiologische Forschung for BAC clones, Ian Bancroft (John Innes Centre, Norwich, UK) and Plant Bioscience Ltd (Norwich, UK) for BC clones, and the Nottingham Arabidopsis Stock Centre (Nottingham, UK) for Col-0, Ler, and Ws2 seeds. This work was funded by the British Biotechnology and Biological Sciences Research Council (C-YY, MS, SG, JPC) and the National Science Foundation (USA) award MCB-9726968 (RCB, BEL).
- 1979) A comparative light- and electron-microscopic study of microsporogenesis in male sterile (MS1) and male fertile soybeans (Glycine max (L.) Merr.). Am. J. Bot. 66, 253–265. and (
- 1997) Gapped blast and psi-blast: a new generation of protein database search programs. Nucl. Acids Res. 25, 3389–3402. , , , , , and (
- 1997) TKRP125, a kinesin-related protein involved in the centrosome-independent organization of the cytokinetic apparatus in tobacco BY-2 cells. J. Cell Sci. 110, 179–189. , and (
- 2000) Two kinesin-related proteins associated with the cold-stable cytoskeleton of carrot cells: characterization of a novel kinesin, DcKRP120-2. Plant J. 24, 859–868. , , , , and (
- 1998) Genetic variation within and among populations of Arabidopsis thaliana. Genetics, 148, 1311–1323. , , and (
- 1995) Predicting coiled coils by use of pairwise residue correlations. Proc. Natl. Acad. Sci. USA, 92, 8259–8263 (available at http://nightingale.lcs.mit.edu/cgi-bin/score). , , , , and (
- 1988) Microtubules associated with simultaneous cytokinesis of coenocytic microsporocytes. Am. J. Bot. 75, 1848–1856. and (
- 1995) Methods in plant immunolight microscopy. Meth. Cell Biol. 49, 85–107. and (
- 2001) The cytoskeleton and spatial control of cytokinesis in the plant life cycle. Protoplasma, 215, 35–49. and (
- 2002) The Arabidopsis ATK1 gene is required for spindle morphogenesis in male meiosis. Development, 129, 2401–2409. , , , , , , and (
- 1998) Floral dip: a simplified method for Agrobacterium-mediated transformation of Arabidopsis thaliana. Plant J. 16, 735–743. and (
- 1992) Behaviour of the maize transposable element Ac in Arabidopsis thaliana. Plant J. 2, 69–81. , , , and (
- 1994) The regulation of alternation of generation in flowering plants. Biol. Rev. 69, 419–422. (
- 1984) A radial pattern of microtubuler extending between the nuclear envelope and the plasma membrane during early male haplophase in flowering plants. Planta, 161, 86–90. and (
- 1968) The ultrastructure and ontogeny of pollen in Helleborus foetidus L. II. Pollen grain development through the callose special wall stage. J. Cell Sci. 3, 175–186. and (
- 1999) Microtubule motors in spindle and chromosome motility. Eur. J. Biochem. 262, 12–18. (
- 1995) pBinPlus: an improved plant transformation vector based on pBin19. Transgenic Res. 4, 288–290. , , , , and (
- 1999) The road less traveled: emerging principles of kinesin motor utilization. Annu. Rev. Cell Dev. Biol. 15, 141–183. and (
- 1994) Constitutive expression of the virulence genes improves the efficiency of plant transformation by Agrobacterium. Proc. Natl. Acad. Sci. USA, 91, 7603–7607. , and (
- 1998) Cytokinesis in flowering plants: cellular process and developmental integration. Curr. Opin. Plant Biol. 1, 486–491. , and (
- 1997) The STUD gene is required for male-specific cytokinesis after telophase II of meiosis in Arabidopsis thaliana. Dev. Biol. 187, 114–124. , , , , , and (
- 1990) Meiosis. Cambridge: Cambridge University Press. (
- 2000) A kinesin family tree. J. Cell Sci. 113, 3681–3682. and (
- 1993) A procedure for mapping Arabidopsis mutations using co-dominant ecotype-specific PCR-based markers. Plant J. 4, 403–410. and (
- 1999) Initiation of translation in prokaryotes and eukaryotes. Gene, 234, 187–208. (
- 1996) Crystal structure of the kinesin motor domain reveals a structural similarity to myosin. Nature, 380, 550–555. , , , and (
- 2002) Maximum likelihood methods reveal conservation of function among closely related kinesin families. J. Mol. Evol. 54, 42–53. , , and (
- 2001) A novel plant kinesin-related protein specifically associates with the phragmoplast organelles. Plant Cell, 13, 2427–2439. , and (
- 2000) Identification of a phragmoplast-associated kinesin-related protein in higher plants. Curr. Biol. 10, 797–800. and (
- 1983) Genetics, cytology, and crossing behavior of an alfalfa (Medicago sativa) mutant resulting in failure of the post-meiotic cytokinesis. Can. J. Genet. Cytol. 25, 390–397. and (
- 2000) Plant organ size control: AINTEGUMENTA regulates growth and cell numbers during organogenesis. Proc. Natl. Acad. Sci. USA, 97, 942–947. and (
- 2000) Genetic dissection of cytokinesis. Plant Mol. Biol. 43, 719–733. , and (
- 2002) Expansion of the cell plate in plant cytokinesis requires a kinsin-like protein/MAPKKK complex. Cell, 109, 87–99. , , et al. (
- 1997) Essential role of a kinesin-like protein in Arabidopsis trichome morphogenesis. Proc. Natl. Acad. Sci. USA, 94, 6261–6266. , , , , , and (
- 1995) Ultrastructure of microsporogenesis and microgametogenesis in Arabidopsis thaliana (L.) Heynh. Ecotype Wassilewskija (Brassicaceae). Protoplasma, 185, 7–21. and (
- 2001) Molecular motors and their functions in plants. Int. Rev. Cytol. 204, 97–178. (
- 2001) Kinesins in the Arabidopsis genome: a comparative analysis among eukaryotes. BMC Genomics, 2, 2 (available at http://www.biomedcentral.com/1471-2164/2/2). and (
- 1997) An allelic series of blue fluorescent trp1 mutants of Arabidopsis thaliana. Genetics, 145, 197–205. , and (
- 1999) Plant Microtechnique and Microscopy. Oxford: Oxford University Press. (
- 1996) Crystal structure of the motor domain of the kinesin-related motor ncd. Nature, 380, 555–559. , , , and (
- 1999) Motor proteins of the kinesin family: structures, variations, and nucleotide binding sites. Eur. J. Biochem. 262, 1–11. , and (
- 1989) Molecular Cloning: a Laboratory Manual, 2nd edn. New York: Cold Spring Harbor Press. , and (
- 2000) http://rgp.dna.affrc.go.jp. , and (
- 1989) Efficient transformation of Agrobacterium species by high voltage electroporation. Nucl. Acids Res. 17, 8385. and (
- 1997) TETRASPORE is required for male meiotic cytokinesis in Arabidopsis thaliana. Development, 124, 2645–2657. , , , , and (
- 1996) Cytokinesis in higher plants. Cell, 84, 821–824. and (
- 1990) Microtubule distribution in dv, a maize meiotic mutant defective in the prophase to metaphase transition. Dev. Biol. 138, 231–242. and (
- 1977) Role of β-1,3-glucanase in post-meiotic microspore release. Dev. Biol. 57, 87–97. (
- 2002) The Arabidopsis HINKEL gene encodes a kinesin-related protein involved in cytokinesis and is expressed in a cell cycle-dependent manner. Curr. Biol. 12, 153–158. , , , , , and (
- 2000) Mapping the jp (jumbo pollen) gene and QTLs involved in multinucleate microspore formation in diploid alfalfa. Theor. Appl. Genet. 101, 372–378. , , , and (
- 1997) The design plan of kinesin motors. Annu. Rev. Cell Dev. Biol. 13, 745–777. and (
- 2000) Cell cycle regulation of the microtubular cytoskeleton. Plant Mol. Biol. 43, 691–703. , and (
- 2000) The kinesin-like calmodulin-binding protein is differentially involved in cell division. Plant Cell, 12, 979–990. , , and (
- 1991) Spatial aspects of cytokinesis in plant cells. Curr. Opin. Cell Biol. 3, 253–256. (
- 1997) Microtubule interaction site of the kinesin motor. Cell, 90, 207–216. , , , , and (
- 1998) Changes in the pattern of organization of microtubules during microspore formation in rice (Oryza sativa L.). Acta Bot. Sinica, 40, 585–590. and (
TETRASPORE (TES) genomic DNA sequence (Colombia-3) EMBL accession no. AJ508243.
TETRASPORE (TES) cDNA sequence (Colombia-3) EMBL accession no. AJ495781.
TETRASPORE (TES) genomic DNA sequence (Landsberg erecta) EMBL accession no. AJ496182.
TETRASPORE (TES) genomic DNA sequence (Ws2) EMBL accession no. AJ507734.
TETRASPORE (TES) predicted protein sequence (Col-3) EMBL accession no. CAD42234.
TETRASPORE (TES) predicted protein sequence (Landsberg erecta) EMBL accession no. CAD42658.
TETRASPORE (TES) predicted protein sequence (Ws2) EMBL accession no. CAD45645.