The Ubiquitin-Specific Protease Subfamily UBP3 ... - Plant Physiology

5 downloads 7288 Views 850KB Size Report
... Osterlund et al., 2000; Seo et al., 2003; Al-Sady et al., 2006), the cell-cycle check- ... [OA] Open Access articles can be viewed online without a sub- scription.
The Ubiquitin-Specific Protease Subfamily UBP3/UBP4 Is Essential for Pollen Development and Transmission in Arabidopsis1[W][OA] Jed H. Doelling*, Allison R. Phillips, Gulsum Soyler-Ogretim, Jasen Wise, Jennifer Chandler, Judy Callis, Marisa S. Otegui, and Richard D. Vierstra Division of Plant and Soil Sciences (J.H.D., G.S.-O.) and Department of Pathology (J.W.), West Virginia University, Morgantown, West Virginia 26506; Department of Genetics (A.R.P., R.D.V.) and Department of Botany (M.S.O.), University of Wisconsin, Madison, Wisconsin 53706; and Section of Molecular and Cellular Biology, University of California, Davis, California 95616 (J. Chandler, J. Callis)

Deubiquitinating enzymes are essential to the ubiquitin (Ub)/26S proteasome system where they release Ub monomers from the primary translation products of poly-Ub and Ub extension genes, recycle Ubs from polyubiquitinated proteins, and reverse the effects of ubiquitination by releasing bound Ubs from individual targets. The Ub-specific proteases (UBPs) are one large family of deubiquitinating enzymes that bear signature cysteine and histidine motifs. Here, we genetically characterize a UBP subfamily in Arabidopsis (Arabidopsis thaliana) encoded by paralogous UBP3 and UBP4 genes. Whereas homozygous ubp3 and ubp4 single mutants do not display obvious phenotypic abnormalities, double-homozygous mutant individuals could not be created due to a defect in pollen development and/or transmission. This pollen defect was rescued with a transgene encoding wild-type UBP3 or UBP4, but not with a transgene encoding an active-site mutant of UBP3, indicating that deubiquitination activity of UBP3/UBP4 is required. Nuclear DNA staining revealed that ubp3 ubp4 pollen often fail to undergo mitosis II, which generates the two sperm cells needed for double fertilization. Substantial changes in vacuolar morphology were also evident in mutant grains at the time of pollen dehiscence, suggesting defects in vacuole and endomembrane organization. Even though some ubp3 ubp4 pollen could germinate in vitro, they failed to fertilize wild-type ovules even in the absence of competing wildtype pollen. These studies provide additional evidence that the Ub/26S proteasome system is important for male gametogenesis in plants and suggest that deubiquitination of one or more targets by UBP3/UBP4 is critical for the development of functional pollen.

The growth and development of plants are exquisitely regulated by the abundance of key proteins, the levels of which are precisely controlled by the combined action of various synthetic and catabolic pathways. One essential pathway for selective protein degradation in plants and animals involves the use of the small protein ubiquitin (Ub) as a tag to target specific proteins for breakdown by the large multicatalytic protease, the 26S proteasome. The Ub/26S proteasome 1 This work was supported by the West Virginia University and the West Virginia University College of Agriculture (Hatch; grant to J.H.D.); the National Science Foundation (grant nos. IBN0212659 to J.C. and MCB–0619736 to M.S.O.); the National Research Initiative of the U.S. Department of Agriculture Cooperative State Research, Education, and Extension Service (grant no. 2005–00930 to R.D.V.); and by a Louis and Elsa Thomsen Wisconsin Distinguished Predoctoral Fellowship (to A.R.P.). * Corresponding author; e-mail [email protected]. The author responsible for distribution of materials integral to the findings presented in this article in accordance with the policy described in the Instructions for Authors (www.plantphysiol.org) is: Jed H. Doelling ([email protected]). [W] The online version of this article contains Web-only data. [OA] Open Access articles can be viewed online without a subscription. www.plantphysiol.org/cgi/doi/10.1104/pp.106.095323

system (UPS) has been shown to be critical for much of plant physiology and development, where it plays key roles in such diverse processes as the cell cycle, photomorphogenesis, embryogenesis, hormone signaling, senescence, plant-microbe interactions, and disease resistance, to name a few (for recent reviews, see Smalle and Vierstra, 2004; Hoecker, 2005; Zeng et al., 2006). The list of plant proteins known to be degraded by the UPS is growing rapidly. Important targets include the DELLA family of GA-signaling repressors (Sun and Gubler, 2004), the EIN3/EIL1 transcriptional effectors of ethylene perception (Binder et al., 2007), the HY5, phyA, LAF1, and PIF3 regulators of photomorphogenesis (Clough and Vierstra, 1997; Osterlund et al., 2000; Seo et al., 2003; Al-Sady et al., 2006), the cell-cycle checkpoint proteins cyclin A3 and cyclin B1 (Genschik et al., 1998), the Aux/IAA family of auxin perception repressors (Parry and Estelle, 2006), the cold-regulated transcription factor ICE1 (Dong et al., 2006), and the ABI3 and ABI5 regulators of abscisic acid signaling (Zhang et al., 2005; Stone et al., 2006). The complete list of protein targets in plants may eventually number in the thousands (Smalle and Vierstra, 2004). In the UPS, Ub becomes covalently attached to various target proteins via the ATP-dependent reaction cascade involving the sequential action of three enzyme classes,

Plant Physiology, November 2007, Vol. 145, pp. 801–813, www.plantphysiol.org Ó 2007 American Society of Plant Biologists

801

Doelling et al.

Ub-activating enzymes (E1s), Ub-conjugating enzymes (E2s), and Ub-protein ligases (E3s; Smalle and Vierstra, 2004). Ub attachment occurs through an isopeptide bond linking the carboxyl group of the C-terminal Gly residue of Ub to the e-amino group of one or more internal Lys within the target. In some situations, only a single Ub is attached. This monoubiquitination appears to serve a variety of functions, including endocytosis, internalization of plasma membrane proteins, and transcriptional regulation (Sigismund et al., 2004). For proteins targeted to the 26S proteasome, a poly-Ub chain is typically assembled using one of several Lys (e.g. Lys-48) within previously attached Ubs as the concatenation site. E3s or E2/E3 complexes are responsible for recognizing specific target proteins at appropriate times and places and thus control much of the specificity of the UPS. Large numbers of E3s exist in various eukaryotes, suggesting that each may be responsible for recognizing only one or a few specific targets (Smalle and Vierstra, 2004). In Arabidopsis (Arabidopsis thaliana), for example, more than 1,300 distinct E3s are predicted, implying that the UPS is particularly important during the life cycle of plants (Gagne et al., 2002; Stone et al., 2005; Gingerich et al., 2005). In addition to the pathway that conjugates Ub, reactions that remove attached Ub are also essential to the UPS (Wilkinson, 2000; Smalle and Vierstra, 2004). This cleavage is catalyzed by a large and diverse collection of deubiquitinating enzymes (DUBs) that specifically releases Ub linked via peptide or isopeptide bonds. In this capacity, DUBs have a number of functions. One is to help generate Ub monomers by processing the initial translation products of the UBQ genes, which typically encode Ub as fusions to itself or other proteins (Callis et al., 1995). A second is to recycle Ub in free functional forms by releasing Ubs from polyubiquitinated proteins during their degradation and then disassembling the freed poly-Ub chains (Amerik and Hochstrasser, 2004). A third role is to reverse ubiquitination by substrate-specific cleavage of Ub moieties. Individual DUBs that act in these various capacities have been described (e.g. Doelling et al., 2001; Amerik and Hochstrasser, 2004; Nijman et al., 2005; Wee et al., 2005; van der Horst et al., 2006; Zhang et al., 2006). A large collection of distinct DUB types exists in eukaryotes, including the Ub C-terminal hydrolases, the otubain- and ataxin-related proteins, the 26S proteasome subunit RPN11, and the ubiquitin-specific proteases (UBPs; Amerik and Hochstrasser, 2004; Nijman et al., 2005). UBPs are a diverse class of DUBs defined through sequence analysis by the presence of several conserved motifs, including signature Cys and His boxes that contain catalytically essential Cys and His residues, respectively, and a sequential stretch of Gln, Gly, Leu, and Phe motifs of unknown function that are defined by the presence of these respective amino acids (Wilkinson, 1997; Yan et al., 2000). Yeast (Saccharomyces cerevisiae) expresses 16 different UBPs (Wilkinson, 1997), whereas Arabidopsis is predicted to contain at least 27 (Yan et al., 802

2000). The Arabidopsis UBP collection can be further classified based on amino acid sequence alignments into 14 subfamilies, with each subfamily containing one to five members (Yan et al., 2000). Whereas the DUB activities of several Arabidopsis UBPs have been confirmed in vitro (Chandler et al., 1997; Rao-Naik et al., 2000; Yan et al., 2000; Doelling et al., 2001; Sridhar et al., 2007), little is currently known about the specific functions and/or substrates of each member. Recently, histone H2b has been shown to be a specific target of the nuclear-localized UBP26 (Sridhar et al., 2007). A systematic genetic analysis of the 16 yeast UBPs revealed that none are essential, suggesting that many have overlapping functions (Amerik et al., 2000). Exceptions are ScUBP4 (Doa4) and ScUBP14, in which loss-of-function mutants have pleiotropic consequences (Amerik et al., 1997). To help elucidate the specific functions of plant UBPs, we and others have begun a similar reverse-genetic analysis in Arabidopsis. Initial studies of mutants comprising five of the 14 subfamilies revealed that UBPs have much more specific roles in plants as compared to yeast. Double mutants missing the UBP1 and UBP2 subfamily are indistinguishable from wild-type plants when grown under normal conditions, but are hypersensitive to the amino acid analog canavanine, suggesting a role for this pair in abnormal protein turnover (Yan et al., 2000). UBP14 is essential during embryo development; plants lacking UBP14 activity are unable to dismantle free multi-Ub chains, leading to disrupted endosperm cellularization and embryo lethality at the globular stage (Doelling et al., 2001; Tzafrir et al., 2002). UBP12 of the UBP12/UBP13 subfamily and UBP7 of the UBP6/UBP7 subfamily restored Arabidopsis ATL2induced toxicity to yeast ubp15D and ubp6D mutants, respectively (Aguilar-Henonin et al., 2006). The absence of UBP26 was shown to increase levels of monoubiquitinated histone H2b; this change was accompanied by an enrichment of histone H3 Lys-4 dimethylation and trimethylation common in active chromatin relative to histone H3 Lys-9 dimethylation common in silent chromatin (Sridhar et al., 2007). Here, we report genetic characterization of the Arabidopsis UBP3 and UBP4 genes that encode a pair of highly related UBPs. Previously, studies revealed that both proteins have DUB activity that will cleave artificial substrates bearing one or more Ubs linked via a-peptide bonds (Chandler et al., 1997; RaoNaik et al., 2000). Using a pair of T-DNA insertion mutations that block accumulation of full-length transcripts, we found that UBP3 and UBP4 together are essential for proper formation and/or function of the male gametophyte. Double-mutant pollen exhibited a variety of phenotypes, such as impaired mitosis II, reduced pollen tube germination efficiency, abnormal vacuole and endomembrane structures at pollen dehiscence, and failure to fertilize ovules. The phenotype of the ubp3 ubp4 mutants reinforces the importance of the UPS system in pollen development and suggests Plant Physiol. Vol. 145, 2007

Deubiquitination Regulates Pollen Development

that deubiquitination of one or more Ub conjugates is essential to the cell cycle of male gametes. RESULTS Isolation of the ubp3-1 and ubp4-1 Mutants

Sequence alignments revealed that UBP3 (At4g39910) and UBP4 (At2g22310) represent a distinct two-member subfamily of Arabidopsis UBPs, sharing 93% amino acid sequence identity with each other (Chandler et al., 1997). Like other UBPs, they contain obvious Cys and His boxes and the intervening Gln, Gly, Leu, and Phe motifs that help define these DUBs (Fig. 1A). However, UBP3/UBP4 are easily distinguished from other Arabidopsis UBPs by their small size (371 and 365 amino acids, respectively) and absence of obvious motifs/sequences beyond the core catalytic region (Yan et al., 2000). BLAST searches identified potential orthologs in a variety of other plant species, including full-length cDNAs from tomato (Solanum lycopersicum; 90% similarity), corn (Zea mays; 88% similarity), and rice (Oryza sativa; 88% similarity), indicating that these two UBPs are universally present within the plant kingdom. Their closest relatives outside of plants include Caenorhabditis elegans R10E11.3 and human and mouse UBPs, USP46 and UBH1, which all are equally small UBPs that share 70% to 72% similarity to UBP3/UBP4 (Hansen-Hagge et al., 1998). UBP3 and UBP4 potentially have both a bipartite nuclear localization sequence (NLS) near the N terminus (residues 83–100) and a SV40-type NLS near the C terminus (residues 244–250) and appear to be enriched in the nucleus (Chandler et al., 1997). UBP3/UBP4 genes are expressed throughout mature Arabidopsis plants, suggesting that they play a general role in plant growth and development (Chandler et al., 1997). The Genevestigator DNA microarray database (http://www.genevestigator.ethz.ch) reported UBP3/UBP4 transcripts in various Arabidopsis tissues, with the levels of UBP3 mRNA about 3 times more abundant than those for UBP4 (Zimmermann et al., 2004). However, the expression patterns of UBP3 and UBP4 were found to differ during pollen development (Honys and Twell, 2004). Whereas UBP3 transcripts were detected during all four stages of pollen development, UBP4 transcripts were not detected in tricellular and mature pollen. A graphic representation of array expression data for UBP3 and UBP4 at four stages of pollen development and in various Arabidopsis tissues is shown in Figure 1B (Honys and Twell, 2004). Widespread distribution of the UBP3/UBP4 proteins was evident by immunoblot analysis of various Arabidopsis tissues with anti-UBP3 antibodies (Chandler et al., 1997). To help define the functions of UBP3 and UBP4, we searched the available Arabidopsis T-DNA mutant populations for insertions within the corresponding genes. An insertion allele for each gene was identified in the Wassilewskija (Ws) ecotype and designated Plant Physiol. Vol. 145, 2007

ubp3-1 and ubp4-1. Sequence analyses revealed that the ubp3-1 allele contains a T-DNA insertion within the fourth intron, 1,435 bp downstream of the translation start codon, whereas the ubp4-1 allele contains a T-DNA insertion within the second intron, 565 bp downstream of the start codon (Fig. 1A). Genomic PCR showed that an intact left border is present at both ends of the T-DNA in the ubp3-1 allele, but only at the 3# end of the T-DNA in the ubp4-1 allele. Both ubp3-1 and ubp4-1 cosegregated with kanamycin resistance conferred by the neomycin phosphotransferase (NPTII) gene linked to the T-DNA. After three backcrosses with wild type, the progeny of individual hemizygous plants displayed a 3:1 ratio of kanamycin resistance, strongly suggesting that a single T-DNA insertion site was present in each line. Reverse transcription (RT)-PCR analysis of RNA isolated from homozygous ubp3-1 and ubp4-1 plants failed to amplify the corresponding full-length transcripts (Fig. 1C). Taken together with the fact that both T-DNAs interrupt the UBP3 and UBP4 genes in between the nucleotide sequences encoding the essential Cys and His boxes, we concluded that these mutations represent null alleles even if partial transcripts are expressed and translated (Fig. 1A). Individuals homozygous for ubp3-1 or ubp4-1 could be generated, indicating that neither gene by itself is essential in Arabidopsis. The morphology and entire life cycle of homozygous ubp3-1 and ubp4-1 plants were indistinguishable from wild-type Ws (Fig. 1D; data not shown). Given the likelihood that UBP3 and UBP4 serve redundant functions, we attempted to generate a doublehomozygous ubp3/ubp4 plant. Self-fertilization of double-hemizygous plants (UBP3/ubp3-1; UBP4/ubp4-1) failed to generate double-homozygous mutant individuals. However, individuals that appeared to be phenotypically normal were identified that contained a single wild-type copy of either UBP3 or UBP4 (genotypes UBP3/ubp3-1; ubp4-1/ubp4-1 and ubp3-1/ubp3-1; UBP4/ubp4-1, respectively). Of the 40 segregating progeny screened from each of the UBP3/ubp3-1; ubp4-1/ ubp4-1 or ubp3-1/ubp3-1; UBP4/ubp4-1 parents, none had the ubp3-1/ubp3-1; ubp4-1/ubp4-1 genotype, suggesting that double-homozygous mutant individuals are not viable. The x 2 test revealed the probability of this failure happening by chance to be ,0.001 if all genotypes have equal fitness. This lethality suggested that UBP3 and UBP4 are collectively important for Arabidopsis embryo development and/or male/female gametophyte function. Either UBP3 or UBP4 Is Required for Pollen Transmission

Because approximately one-half of the progeny from a self-fertilized UBP3/ubp3-1; ubp4-1/ubp4-1 parent had the UBP3/ubp3-1; ubp4-1/ubp4-1 genotype, whereas the other half had the UBP3/UBP3; ubp4-1/ubp4-1 genotype, we reasoned that UBP3/UBP4 is essential for either male or female gametes, but not both. 803

Doelling et al.

Careful dissection of siliques from self-fertilized UBP3/ubp3-1; ubp4-1/ubp4-1 or ubp3-1/ubp3-1; UBP4/ ubp4-1 flowers did not reveal aborted seeds, suggesting that male gametes with the ubp3-1 ubp4-1 haploid genotype were defective. To show that mutant male gametes were defective, we performed reciprocal crosses between wild-type Ws and either UBP3/ubp3-1; ubp4-1/ubp4-1 or ubp3-1/ ubp3-1; UBP4/ubp4-1 plants. Given that the UBP3 and UBP4 genes are located on different chromosomes (IV and II, respectively) and thus should segregate independently, one-half of the progeny of a UBP3/ubp3-1; ubp4-1/ubp4-1 3 UBP3/UBP3; UBP4/UBP4 cross should be the UBP3/ubp3-1; UBP4/ubp4-1 genotype, whereas the other half should be the UBP3/UBP3; UBP4/ubp4-1 genotype if the gametes are immune to missing functional copies of both UBP3 and UBP4 (Fig. 2A). This distribution was obtained if wild-type plants served as pollen donors, indicating that female gametogenesis was normal (Fig. 2B; data not shown). Similarly, roughly one-half of the progeny (18 of 40) of crosses between ubp3-1/ubp3-1; UBP4/ubp4-1 females and UBP3/UBP3; UBP4/UBP4 males contained both the ubp3-1 and ubp4-1 mutant alleles. However, when UBP3/ubp3-1; ubp4-1/ubp4-1 or ubp3-1/ubp3-1; UBP4/ ubp4-1 plants supplied the pollen, we failed to find any individuals among the 200 or 99 progeny, respectively, that resulted from the simultaneous transmission of both ubp3-1 and ubp4-1 via the same pollen grain (x2 probability that this failure was due to chance alone is ,0.001; Fig. 2C; data not shown). Such a segregation imbalance indicated that UBP3/UBP4 is required for male gametogenesis/pollen function. Complementation with UBP3 or UBP4 Can Rescue ubp3-1 ubp4-1 Pollen Function

Figure 1. Gene structure and genetic analysis of Arabidopsis UBP3 and UBP4. A, Gene structures showing the positions of the coding regions for the Cys, Gln, Gly, Leu, Phe, and His boxes, the positions of the ubp3-1 and ubp4-1 T-DNA insertions, and primer binding sites used for genotyping and RT-PCR analyses. Lines represent introns and nontranscribed DNA upstream and downstream of each gene. White and gray rectangles represent protein coding regions and 5#- and 3#-untranslated regions, respectively. The nucleotide sequence of each primer is presented in ‘‘Materials and Methods.’’ B, Bar graph depicting microarray expression data of UBP3 (black bars) and UBP4 (gray bars) as reported by Honys and Twell (2004). UNM, Uninucleate microspores; BCP, bicellular pollen; TCP, tricellular pollen; and MPG, mature 804

To verify that the lack of active UBP3/UBP4 protein was responsible for the pollen defect, we attempted to recover double-mutant pollen by introducing either a wild-type UBP4 gene under the control of its own promoter (UBP4-T) or wild-type UBP3 cDNA under the control of the AtUBQ10 promoter (UBP3-T). Among the hygromycin-resistant progeny from self-fertilized individuals of genotypes UBP3/ubp3-1; ubp4-1/ubp4-1; UBP4-T and ubp3-1/ubp3-1; UBP4/ubp4-1; UBP3-T, we identified multiple double-homozygous mutant individuals (Supplemental Figure S1; data not shown), thus demonstrating successful complementation. To confirm that an enzymatically active copy of either UBP3 or UBP4 is required in Arabidopsis pollen,

pollen grains. C, RT-PCR analysis of total RNA from wild-type Ws plants and plants homozygous for either the ubp3-1 or ubp4-1 T-DNA insertions. Genomic DNA from wild-type Ws plants confirmed the activity of the primers, whereas RT-PCR analysis of the UBP14 transcripts verified the integrity of the initial RT reactions. D, Phenotype of 3-week-old ubp3-1 and ubp4-1 plant seedlings as compared to wildtype Ws Arabidopsis. Plant Physiol. Vol. 145, 2007

Deubiquitination Regulates Pollen Development

Figure 2. ubp3-1 ubp4-1 double-mutant pollen is defective in pollen transmission. A, Punnett square diagrams of the reciprocal crosses. Upper and lower cases represent the wild-type and mutant alleles, respectively. Possible pollen genotypes are indicated to the left of the squares, possible ovule genotypes above the squares, and possible offspring genotypes within the squares. The gray X in the progeny square marks the UBP3/ubp3-1; UBP4/ubp4-1 genotype that was not found when pollen from a UBP3/ubp3-1; ubp4-1/ubp4-1 donor was used. B and C, PCR genotyping of 20 individual offspring resulting from each reciprocal cross. DNA isolated from individual offspring was genotyped at the UBP3 locus using two PCR reactions: UBP3-2 1 UBP3-4 to detect the presence of a wild-type allele (UBP3) and UBP3-2 1 JL-202 to detect the presence of a mutant ubp3-1 allele (ubp3-1). DNA from homozygous mutant ubp3-1 (m) or wild-type (w) individuals was used as control. See Figure 1 for the positions of the primers.

we attempted to rescue the ubp3-1 ubp4-1 genotype with a UBP3 transgene encoding an active-site mutant altered in the Cys box (Cys-32-Ser; C32S). It has been well established for a number of UBPs that such a mutation blocks UBP activity toward Ubs linked by either peptide or isopeptide linkages (Chandler et al., 1997; Rao-Naik et al., 2000; Yan et al., 2000; Doelling et al., 2001, Sridhar et al., 2007). Despite numerous attempts, a double-homozygous ubp3-1/ubp3-1; ubp4-1/ ubp4-1 mutant that also contained the UBP3-C32S-T transgene was not identified among the self-fertilized progeny of UBP3/ubp3-1; ubp4-1/ubp4-1 plants that were homozygous for the UBP3-C32S-T locus. FurPlant Physiol. Vol. 145, 2007

thermore, when UBP3/ubp3-1; ubp4-1/ubp4-1 individuals homozygous for UBP3-C32S-T were used as pollen donors in crosses with wild-type plants, none of the 47 progeny tested contained the ubp3-1 allele (x 2 test ,0.001; data not shown). As opposed to the need for active UBP3 or UBP4 in pollen, it was also remotely possible that our failure to rescue the ubp3-1/ubp3-1; ubp4-1/ubp4-1 genotype with the UBP3-C32S-T transgene was caused either by a tight linkage of the UBP3-C32S-T transgene to the UBP3 locus that would have prevented independent segregation or by a failure to express the transgene at the right time and place. We ruled out a tight linkage between UBP3-C32S-T and UBP3 based on observations that selfed UBP3/ubp3-1; ubp4-1/ubp4-1; UBP3-C32S-T plants produced pollen of the UBP3 ubp4-1 UBP3C32S-T and UBP3 ubp4-1 genotypes in approximately equal proportions. Twelve of 20 progeny from a UBP3/ UBP3; UBP4/UBP4 3 UBP3/ubp3-1; ubp4-1/ubp4-1; UBP3-C32S-T cross contained the UBP3-C32S-T transgene and eight did not, whereas all 20 progeny were homozygous wild type at the UBP3 locus (data not shown). We considered improper UBP3-C32S-T transgene expression unlikely because UBP3-T driven by the same UBQ10 promoter could rescue the ubp3-1 ubp4-1 defect. Nevertheless, it remained possible that the UBP3-C32S-T transgene integrated into a transcriptionally silent region. To demonstrate UBP3-C32S-T transgene expression, we performed RT-PCR analysis on the transgene using total RNA isolated from adult ubp3-1/ubp3-1; UBP4/ubp4-1; UBP3-C32S-T plants as the template. As can be seen in Figure 3, mRNA derived from the UBP3-C32S-T locus could be readily amplified. Collectively, our genetic analyses indicated that the Ub-specific protease activity derived from UBP3/UBP4 is essential for proper pollen function. ubp3 ubp4 Mutant Pollen Germinate Less Efficiently and Are Defective in Sperm Production

To help define the defect in ubp3 ubp4 mutant pollen more precisely, dehiscent pollen collected from UBP3/ ubp3-1; ubp4-1/ubp4-1 plants were compared to those obtained from their UBP3/UBP3; ubp4-1/ubp4-1 siblings with regard to germination efficiency, morphology, and the number of sperm nuclei. To aid in the analyses, we introgressed the quartet1-2 (qrt1-2) mutation into the ubp3/ubp4 mutant background (Preuss et al., 1994). This mutation prevents physical separation of the pollen grains generated from each microspore mother cell, thus allowing analysis of the four meiotic products as a stable pollen tetrad. Whereas all four pollen grains in a tetrad from a UBP3/UBP3; ubp4-1/ubp4-1; qrt1-2/qrt1-2 anther should be normal, two of the four from a UBP3/ubp3-1; ubp4-1/ubp4-1; qrt1-2/qrt1-2 anther should be defective if UBP3/UBP4 activity is essential for pollen development and/or function. Germination assays revealed that pollen from UBP3/ ubp3-1; ubp4-1/ubp4-1 plants germinated less efficiently than pollen obtained from UBP3/UBP3; ubp4-1/ubp4-1 805

Doelling et al.

Figure 3. The mutant UBP3-C32S-T transgene is transcribed. Total RNA isolated from wild-type Ws, homozygous ubp3-1/ubp3-1; UBP4/ UBP4 plants, and ubp3-1/ubp3-1; UBP4/ubp4-1 plants containing the UBP3-C32S-T transgene were subjected to RT-PCR analysis using primers 3-5 and 3-6. 1 and 2, RT reaction was carried out with or without reverse transcriptase, respectively. Lane gDNA indicates the PCR product using genomic DNA as a template and lane M contains a DNA size ladder with approximate lengths indicated on the right. The arrow designates the RT-PCR-amplified UBP3 cDNA product.

plants (58% for UBP3/ubp3-1; ubp4-1/ubp4-1 versus 83% for UBP3/UBP3; ubp4-1/ubp4-1 [Table I]). Interestingly, the efficiency for the UBP3/ubp3-1; ubp4-1/ubp4-1 background in multiple experiments was .50%, suggesting that some pollen with the ubp3-1 ubp4-1 genotype could germinate. The viability of ubp3-1 ubp4-1 pollen was also supported by the inspection of individual tetrads. Here, an occasional tetrad was found from the UBP3/ubp3-1; ubp4-1/ubp4-1; qrt1-2/qrt1-2 parent that had three and sometimes four pollen tubes emerging upon hydration (Fig. 4A). Light microscopy of individual pollen grains in the tetrads from UBP3/ ubp3-1; ubp4-1/ubp4-1; qrt1-2/qrt1-2 parents failed to identify any gross morphological detects at the time of pollen dehiscence, except that we did observe a slightly higher proportion of collapsed pollen grains in tetrads from UBP3/ubp3-1; ubp4-1/ubp4-1; qrt1-2/qrt1-2 plants (109 of 1,004 pollen grains [11%]) as compared to those from UBP3/UBP3; ubp4-1/ubp4-1; qrt1-2/qrt1-2 plants (17 of 748 pollen grains [2.3%]). Staining of the

shrunken grains with 4#,6-diamidino-2-phenylindole (DAPI) failed to detect pollen nuclei, suggesting that these collapsed grains lacked nuclei or that staining was obstructed (Fig. 4G). The male gametophyte in Arabidopsis reaches maturity at the time of anther dehiscence where it contains three haploid cells: a vegetative cell that directs pollen tube growth and two internal sperm cells that are needed for double fertilization of the ovule (McCormick, 2004). The sperm cells are the mitotic products of the haploid generative cell; they and the generative cell can easily be distinguished from the vegetative cell by their relatively smaller nuclei that stain intensely with DAPI (McCormick, 2004; Fig. 4B). When tetrads derived from self-fertilized UBP3/ ubp3-1; ubp4-1/ubp4-1; qrt1-2/qrt1-2 plants were visualized by DAPI staining, we discovered a substantial defect in sperm cell development. Whereas all four pollen grains in 70% of the tetrads obtained from UBP3/UBP3; ubp4-1/ubp4-1; qrt1-2/qrt1-2 plants contained the two obvious sperm nuclei, only 3.6% of the tetrads from UBP3/ubp3-1; ubp4-1/ubp4-1; qrt1-2/qrt1-2 plants had this normal composition (Fig. 4, B and C; Supplemental Table S1). Most of the tetrads from the UBP3/ubp3-1; ubp4-1/ubp4-1; qrt1-2/qrt1-2 parents contained one or more pollen grains with only two obvious nuclei, a diffusively staining larger vegetative cell nucleus and a strongly staining smaller nucleus that either represented the precursor generative cell or a single sperm cell (Fig. 4, D–F). For example, we could detect only a single generative/sperm nucleus in two of the four pollen grains in 28% of the tetrads from selfed UBP3/ubp3-1; ubp4-1/ubp4-1; qrt1-2/qrt1-2 plants versus 3% of the tetrads from selfed UBP3/UBP3; ubp4-1/ubp4-1; qrt1-2/qrt1-2 plants. In 67% of the pollen tetrads from UBP3/ubp3-1; ubp4-1/ubp4-1; qrt1-2/qrt1-2 plants that contained either two or three pollen grains with two sperm nuclei each, the remaining one or two pollen grains contained a single generative/sperm nucleus, suggesting that UBP3/UBP4 is important for pollen mitosis II. When the same data were analyzed without regard to tetrad association, 90% of the 748 pollen grains from UBP3/UBP3; ubp4-1/ubp4-1; qrt1-2/qrt1-2 plants contained two sperm nuclei. In contrast, only 48% of the 1,004 pollen grains from UBP3/ubp3-1; ubp4-1/ubp4-1; qrt1-2/qrt1-2 plants had both sperm nuclei, whereas an additional 38% contained a single generative/sperm

Table I. In vitro pollen germination In vitro pollen germination comparing the germination efficiency of pollen harvested from UBP3/UBP3; ubp4-1/ubp4-1 plants with their UBP3/ubp3-1; ubp4-1/ubp4-1 siblings. Exp

1 2 3 Total 806

UBP3/UBP3; ubp4-1/ubp4-1

UBP3/ubp3-1; ubp4-1/ubp4-1

Germ

Total

% Germ

Germ

Total

% Germ

126 331 116 573

164 384 140 688

77 86 83 83

99 256 131 486

240 379 213 832

41 68 62 58 Plant Physiol. Vol. 145, 2007

Deubiquitination Regulates Pollen Development

Figure 4. Germination and nuclei content of ubp3-1 ubp4-1 pollen. A, In vitro germination of pollen tetrads from UBP3/ubp3-1; ubp4-1/ubp4-1; qrt1-2/qrt1-2 plants. Each image shows an individual tetrad where three or four pollen grains germinated. The bar represents 50 mm. B to G, DAPI staining of pollen from UBP3/UBP3; ubp4-1/ubp4-1; qrt1-2/qrt1-2 (33) and UBP3/ubp3-1; ubp4-1/ubp4-1; qrt1-2/qrt1-2 (33) plants showing that some pollen grains contain two sperm nuclei, others only one generative/ sperm nucleus, and some without any visible sperm nuclei (left). Photographs of each pollen tetrad under visible light (right) are shown. Fluorescence of the DAPI under UV light. B, Tetrad from a UBP3/UBP3; ubp4-1/ubp4-1; qrt1-2/qrt1-2 plant where each pollen grain contains two sperm nuclei. Shown are tetrads from a UBP3/ubp3-1; ubp4-1/ubp4-1; qrt1-2/qrt1-2 plant where four (C), three (D), or two (E and F) pollen grains contain two sperm nuclei, but the remaining pollen grains most often contain only one generative/sperm nucleus. G, Tetrad from a UBP3/ubp3-1; ubp4-1/ubp4-1; qrt1-2/qrt1-2 plant where one pollen grain contains two sperm nuclei, one pollen grain contains only a vegetative nucleus, and the other two pollen grains are visibly shrunken and do not show any nuclei.

nucleus. Another 11% of the pollen grains were smaller in which no DAPI staining was detected (Fig. 4G). We could not detect sperm nuclei in the remaining 3.7% of the pollen grains, although they had normal morphology. The absence of two sperm cells in a high percentage of pollen from self-fertilized UBP3/ubp3-1; ubp4-1/ ubp4-1; qrt1-2/qrt1-2 plants likely reflected a failure of the generative cell to complete mitosis II to produce the two sperm cells (McCormick, 2004). ubp3 ubp4 Mutant Pollen Have Altered Cellular Structure

To further survey the defects in ubp3-1 ubp4-1 pollen, we examined high-pressure frozen/freeze-substituted pollen grains from UBP3/ubp3-1; ubp4-1/ubp4-1; qrt1-2/ qrt1-2, ubp3-1/ubp3-1; UBP4/ubp4-1; qrt1-2/qrt1-2, and qrt1-2/qrt1-2 plants for their structural organization by transmission electron microscopy (Fig. 5). The pollen grains were collected at two different stages of development: after mitosis I and at anther dehiscence when wild-type pollen have completed mitosis II to form two sperm cells. After mitosis I, pollen from plants of all three genotypes looked similar with dense cytoplasm and abundant organelles (Fig. 5, A, D, and G). Because all four pollen grains in tetrads from ubp3-1/ ubp3-1; UBP4/ubp4-1; qrt1-2/qrt1-2 and UBP3/ubp3-1; ubp4-1/ubp4-1; qrt1-2/qrt1-2 plants exhibited dense cytoplasm following mitosis I (Fig. 5, D and G) and most mutant pollen contain at least one sperm/generative nucleus at dehiscence (Fig. 4, C–F), we concluded that ubp3-1 ubp4-1 mutant pollen likely develop normally to mitosis I. At anther dehiscence, approximately 50% Plant Physiol. Vol. 145, 2007

of the pollen grains from both mutant plants exhibited very dense cytoplasm packed with vesicles, endoplasmic reticulum membranes, mitochondria, and lipid bodies that was indistinguishable from qrt1-2/qrt1-2 pollen (Fig. 5, B and C). However, the remaining pollen grains showed some structural alterations specifically in the organization of the endomembrane system. Among these pollen grains, the most striking difference was the highly enlarged and dilated vacuolar system, where some vacuoles even appeared to have encapsulated portions of cytoplasm and organelles (Fig. 5, E, F, H, and I). A smaller fraction of pollen grains (approximately 15%) from the UBP3/ubp3-1; ubp4-1/ubp4-1; qrt1-2/qrt1-2 plants was completely collapsed at this stage (Fig. 5, J and K). Such collapsed pollen was not evident in the samples from ubp3-1/ ubp3-1; UBP4/ubp4-1; qrt1-2/qrt1-2 plants. UBP3/UBP4 Also Have Roles in Pollen Transmission

Although some of the ubp3-1 ubp4-1 double-mutant pollen appeared to contain two sperm cells and some of these mutant pollen grains likely germinate (at least in vitro; Fig. 4A), we failed to find progeny produced by the fertilization of a wild-type ovule with this pollen (see above). This failure suggested that UBP3/ UBP4 also have important roles in pollen function downstream of pollen mitosis II and subsequent germination. Possible additional defects include: (1) poor competition with wild-type and single-mutant pollen (UBP3 UBP4, UBP3 ubp4-1, or ubp3-1 UBP4); (2) failure to germinate on the stigma and/or penetrate into the 807

Doelling et al.

Figure 5. Transmission electron microscope analysis of high-pressure frozen/freeze-substituted pollen. The endomembrane organization of pollen from qrt1-2/qrt1-2 (A), ubp3-1/ubp3-1; UBP4/ubp4-1; qrt1-2/qrt1-2 (D), and UBP3/ubp3-1; ubp4-1/ ubp4-1; qrt1-2/qrt1-2 (G) plants appears similar following pollen mitosis I. Upper case and lower case numbers indicate wildtype and mutant alleles, respectively. D, Pollen tetrad containing two pollen grains in which the generative cell (GC) is still attached to the intine wall. The pollen grain depicted in G is slightly more mature in that the generative cell has already been internalized. VN, Vegetative nucleus. Pollen from qrt1-2/qrt1-2 (B and C), ubp3-1/ubp3-1; UBP4/ubp4-1; qrt1-2/qrt1-2 (E and F), and UBP3/ubp3-1; ubp4-1/ubp4-1; qrt1-2/qrt1-2 plants (H–K) following anther dehiscence show differences in vacuole and endomembrane organization. B and C, Pollen from qrt1-2/qrt1-2 plants contain dense cytoplasm packed with lipid bodies (LB), vesicles (V), mitochondria (M), and endoplasmic reticulum (ER) membranes. E, F, H, and I, Pollen grains from ubp3-1/ubp3-1; UBP4/ubp4-1; qrt1-2/qrt1-2 (E and F) and UBP3/ubp3-1; ubp4-1/ubp4-1; qrt1-2/qrt1-2 (H and I) plants showing an enlarged vacuolar system (asterisks). J and K, Collapsed pollen grains from UBP3/ubp3-1; ubp4-1/ubp4-1; qrt1-2/qrt1-2 plants. Scale bars in A, B, D, E, G, H, J, and K 5 5 mm; in C, F, and I 5 2 mm.

style; (3) aberrant pollen tube guidance to the micropyle in the ovules; and/or (4) defects in fertilization of the female gametophyte (Johnson et al., 2004; Lalanne et al., 2004). We addressed the potential problem of competition by placing only a few pollen tetrads from UBP3/ubp3-1; ubp4-1/ubp4-1; qrt1-2/qrt1-2 individuals onto the stigmas of emasculated wild-type flowers. Of the 100 progeny seedlings generated by such limited pollination where an average of three to 10 seeds was obtained per silique (as opposed to .20 seeds per silique for wild-type Ws), none was found by PCR genotyping to carry the ubp3-1 allele (data not shown). As a consequence, we concluded that UBP3/UBP4 also have roles in pollen germination, pollen tube growth and guidance, and/or ovule fertilization. 808

DISCUSSION

DUBs perform a number of essential functions in the UPS, including the initial synthesis of Ub monomers, recycling Ubs bound to proteins or concatenated into poly-Ub chains, and release of Ub from ubiquitinated proteins as a way to reverse the action of ubiquitination. Whereas the two former activities affect the global functions of Ub by maintaining the pool of free functional Ub, the latter can have important regulatory consequences by affecting specific targets. Here, we show that a subfamily of Arabidopsis UBPs encoded by the UBP3 and UBP4 genes plays an essential role in pollen development. The two DUBs are enriched in the nucleus using one or both potential NLSs and are Plant Physiol. Vol. 145, 2007

Deubiquitination Regulates Pollen Development

widely distributed among various Arabidopsis tissues. Whereas the single-homozygous ubp3-1 and ubp4-1 mutants are normal, the formation of double-homozygous plants was prohibited by a block in pollen development/function. This defect can be rescued by wildtype UBP3 or UBP4, but not by a UBP3 active-site mutant, demonstrating that the DUB activity of UBP3/ UBP4 is required. Homozygous ubp3/ubp4 mutants were affected at various steps in pollen maturation, with a striking defect evident in the production of the two sperm nuclei from the generative cell. Histochemical staining of pollen nuclei revealed that a majority of double-mutant pollen grains contain a single generative/sperm nucleus in addition to the vegetative nucleus, suggesting that pollen mitosis II, which creates the two sperm cells, was substantially abrogated. These aberrant nuclei stained intensely with DAPI, suggesting that their chromosomes remained condensed like those in sperm and generative nuclei (McCormick, 2004). Whether these single sperm/generative cells completed S-phase, but were blocked in cytokinesis and were thus diploid, or whether they were blocked prior to S-phase and thus remained haploid, is not yet known. Transmission electron microscopy revealed substantial changes in the endomembrane system of double-mutant pollen. Additional defects were also found with respect to germination and fertilization. Although ubp3-1 ubp4-1 pollen can germinate (albeit poorly), we have not yet found evidence that they can successfully fertilize wild-type ovules. In contrast to the strong effects on pollen, development and fertilization of the ovule were unaffected in the ubp3-1 ubp4-1 background, indicating that female gametogenesis does not obligatorily require UBP3/ UBP4. It also should be stressed that UBP3 and UBP4 are widely distributed in other tissues besides anthers (Chandler et al., 1997; this report) and thus are likely to have important roles in Arabidopsis outside of male gametogenesis. Such possibilities were not investigated here because the pollen-defective phenotype prevented generation of a ubp3/ubp4 double-homozygous mutant plant. With respect to UBP3/UBP4, it remains unclear how their absence blocks pollen maturation. UBP3/UBP4 are best related to C. elegans R10E11.3, human and mouse USP46, and human and mouse UBH1 (HansenHagge et al., 1998). At the present time, the specific functions of these UBPs are not known. The observation that ubp3-1 ubp4-1 pollen arrest at various stages with a major block at mitosis II suggests that they are not inhibited at a particular stage, but became inhibited once levels of UBP3/UBP4 drop below a critical threshold as the male gametophyte divides and grows. Their enrichment in nuclei would discount a role in the initial synthesis of Ub monomers by processing the translation products of UBQ genes and favors a role that involves the release of Ub attached to itself or other targets via isopeptide linkages. One scenario that better agrees with the wide distribution of UBP3/UBP4 throughout Arabidopsis Plant Physiol. Vol. 145, 2007

plants and the lack of additional motifs beyond the catalytic core that could provide target specificity is that UBP3/UBP4 provides general activity to the UPS. In particular, UBP3/UBP4, like UBP14 (Doelling et al., 2001), could assist in disassembling poly-Ub chains and thus play a central role in maintaining an ample supply of Ub monomers. Whereas loss of UBP14 activity induces dramatic stabilization of Ub conjugates and free poly-Ub chains, which subsequently blocks early embryogenesis in Arabidopsis (Doelling et al., 2001), loss of UBP3/UBP4 activity could become acute much earlier or be more specific for the male gametophyte. UBP3/UBP4, like UBP14, could digest Lys-48linked chains that are important for 26S proteasome recognition (Doelling et al., 2001), or it could prefer poly-Ub linked by any one of the other six Lys that are used in vivo for chain assembly (Peng et al., 2003; Maor et al., 2007; S. Saracco and R.D. Vierstra, unpublished data). Lys-63 chains, in particular, have been linked to chromatin remodeling and DNA repair (Hofmann and Pickart, 1999). Unfortunately, without a method to collect sufficient quantities of ubp3-1 ubp4-1 pollen, this scenario cannot yet be tested. Another more intriguing scenario is that UBP3/ UBP4 have a specific role in deubiquitinating a key regulator of mitosis and/or pollen development. Lack of this DUB activity in the ubp3-1 ubp4-1 double mutant could destabilize a ubiquitinated target by discouraging its recognition by the 26S proteasome, or block or enhance target protein function by preventing removal of the Ub moieties. The UPS is known to perform a number of essential functions throughout mitosis that could be related to the defect in pollen mitosis II. Here, the UPS removes a variety of proteins involved in DNA replication, mitotic entry, anaphase entry and progression, and exit from mitosis (Genschik et al., 1998; Castro et al., 2005; Fulop et al., 2005). The anaphasepromoting E3 complex, in particular, plays an important role in the ubiquitination and degradation of various cell-cycle checkpoint proteins (Capron et al., 2003). In addition, a number of DUBs have intriguing connections to the cell cycle, DNA regulation, and/or gamete development. Examples include the deubiquitination of (1) Drosophila Lqf by the UBP Faf that controls eye cell fate by programmed cell death; (2) histone H2B by yeast UBP8 that affects nucleosome packing; and (3) p53 by HAUSP that attenuates cell division and oncogenesis (for review, see Amerik and Hochstrasser, 2004). In a similar fashion, UBP3/UBP4 could transiently stabilize a particular checkpoint protein via the removal of Ub chains. DUBs have also been implicated in intracellular trafficking and endomembrane assembly through the modification of factors that direct these processes (for review, see Millard and Wood, 2006). Two examples are Drosophila FAM and mammalian VCIP135. Drosophila epsin is a trafficking accessory protein involved in endocytosis that can be deubiquitinated and stabilized by FAM (Cadavid et al., 2000). VCIP135 reverses a ubiquitination event that occurs during disassembly 809

Doelling et al.

of Golgi cisternae prior to mitosis, thereby allowing for membrane reassembly following mitosis (Wang et al., 2004). Given the substantial changes in intracellular morphology of the pollen, UBP3/UBP4 could somehow be involved in directing endomembrane dynamics by releasing Ub signals on one or more critical factors. Also of potential interest is the amino acid sequence similarity of the catalytic core domain of UBP3/UBP4 (Cys, Gln, Gly, Leu, Phe, and His boxes) with Aspergillus nidulans CreB (67% similarity), which plays a role in the carbon metabolism of this fungus (Lockington and Kelly, 2001). Unlike UBP3/UBP4, CreB contains a 287-amino acid extension beyond the His box that presumably directs substrate specificity or cellular localization. CreB is believed to deubiquitinate CreA and other substrates during carbon catabolite repression (Boase and Kelly, 2004). Based on this similarity, UBP3/UBP4 could have an analogous role in the carbon metabolism of pollen, which is crucial to this specialized heterotrophic tissue because it completes its development inside the anther and after dehiscence on stored sugars (McCormick, 2004). The UPS has also been specifically implicated in pollen development/function. For example, during the development of maize pollen, the levels of Ub and Ub conjugates were found to decrease 10- to 50-fold when comparing young microspores without vacuoles to mature pollen grains (Callis and Bedinger, 1994). Speranza et al. (2001) reported that pollen tube growth in kiwifruit (Actinidia deliciosa) was significantly reduced by treatment with 26S proteasome-specific inhibitors such as MG-132 and epoxomicin, whereas Kim et al. (2006) recently showed that exogenous Ub can block pollen adhesion to lily styles. In Arabidopsis specifically, the core ASK1 subunit of SCF E3s that directs ubiquitination is required for male fertility (Yang et al., 1999; Devoto et al., 2002). And, finally, several E3s play critical roles in discriminating self from non-self pollen during the incompatibility reaction between pollen and the stigma/stylar tissue of the female reproductive tissue (Stone et al., 2003; Ushijima et al., 2003; Qiao et al., 2004). Defects in any one of these steps could generate one or more of the defects seen here for ubp3-1 ubp4-1 plants. With respect to pollen mitosis II, a number of Arabidopsis mutants (mad1, mad2, mad3, duo1, duo2, and cdka1) have been described that could be related to UBP3/UBP4 (Grini et al., 1999; Durbarry et al., 2005, Iwakawa et al., 2006; Nowack et al., 2006). mad1 mutants are disrupted in cell or nuclear division during mitosis II, the mad2 mutation affects differentiation of the generative cell, and mad3 mutants fail to form the normal intine layers of pollen grains (Grini et al., 1999). The duo1 mutation prevents entrance into mitosis following DNA replication, whereas duo2 mutants are defective in chromatid separation, suggesting that mitosis is arrested during prometaphase (Durbarry et al., 2005). Although several of the mad and duo loci have been mapped to specific chromosomes, the corresponding gene has been identified for only duo1 810

(Rotman et al., 2005). The DUO1 locus encodes a pollen-specific R2R3 MYB transcription factor that may help direct the transcription of genes that regulate the cell cycle. In mice, c-Myb can be ubiquitinated and degraded by the UPS (Bies and Wolff, 1997). Like ubp3/ ubp4 mutants, mad1 and mad2 mutants display a range of phenotypic abnormalities. cdka1 mutant pollen contains a single sperm cell due to failure of pollen mitosis II; however, this pollen can successfully fertilize ovules leading to arrested embryogenesis at the globular stage (Iwakawa et al., 2006; Nowack et al., 2006). The possibility that any of these loci represent targets of UBP3/UBP4 following their ubiquitination awaits further genetic analyses. Sequence analysis of UBP3 and UBP4 revealed that they, like USP46, UBH1, and CreB, contain N-terminal consensus sequences for myristoylation (GAAGSKLEKA, residues 2–11 in UBP3 (Boisson et al., 2003). This posttranslational modification involves the attachment of the saturated 14-carbon fatty acid myristic acid to an exposed N-terminal Gly following removal of the initiator Met. UBP3, in particular, was verified to be myristoylated (at least in vitro) using an Arabidopsis source of the myristoyl transferase activity (Boisson et al., 2003). Protein myristoylation serves a number of functions, including participating in signal transduction networks where a large proportion of targets are protein kinases. With respect to the potential location of UBP3/UBP4 in the nucleus, this addition could promote their association with specific ubiquitinated transcription factors or the 26S proteasome, which could also be myristoylated through the RPT2 subunit (Boisson et al., 2003). Clearly, the next steps in defining UBP3/UBP4 functions in the UPS will be to identify their interactors in Arabidopsis and define the role of myristoylation in their intracellular distribution. MATERIALS AND METHODS Isolation of the ubp3-1 and ubp4-1 T-DNA Insertion Mutants The ubp3-1 insertion mutant was identified in a PCR screen of the University of Wisconsin Arabidopsis (Arabidopsis thaliana) T-DNA population prepared with the Ws ecotype (Sussman et al., 2000), using either gene-specific primer UBP3-2 (TCCTTCGTTGCTTCGTCGCTTTGGACAGA) or UBP3-4 (AACTGTACATGGCCTTTTTCAAAGTATAC) in combination with the left-border T-DNA-specific primer JL-202 (http://www.biotech.wisc.edu/ NewServicesAndResearch/Arabidopsis). The ubp4-1 insertion mutant was identified in the T-DNA-transformed population of Arabidopsis ecotype Ws generated by Dr. Kenneth Feldman (obtained from the Arabidopsis Biological Resource Center [ABRC]) using primer UBP4-3 (GTAGCCATGATCAGTATTACTTGAGTACTC) in combination with the T-DNA left-border-specific primer L (Krysan et al., 1996). Both T-DNA insertion lines contained the associated NPTII gene, which allowed us to track each insertion by kanamycin resistance. Each mutant was backcrossed three times to wild-type Arabidopsis ecotype Ws to eliminate possible second-site mutations before creating the double mutant.

Plant Growth and Genetic Crosses Arabidopsis seeds were surface sterilized in 33% bleach, washed three times with sterile water, and incubated at 4°C for 2 d prior to germination on Gamborg B5 agar medium containing 2% Suc. If antibiotic selection was

Plant Physiol. Vol. 145, 2007

Deubiquitination Regulates Pollen Development

Transmission Electron Microscopy

desired, either kanamycin or hygromycin B was added to the growth medium at 25 mg/L. Plants were grown in a growth chamber at 21°C with a 16-h-light/ 8-h-dark photoperiod. When pollination and seed set were desired, seedlings were transferred to soil approximately 2 weeks after germination. Genetic crosses were performed by pollinating the stigmas of immature, emasculated flowers of the desired female parent with mature pollen from the desired male parent. The qrt1-2 mutation in the Col-3 background (seed stock no. 8846; ABRC; Preuss et al., 1994) was introgressed into the ubp3-1 ubp4-1 mutant to facilitate pollen tetrad analysis.

Whole anthers were removed from unopened and open flowers and immediately loaded into sample holders filled with 0.1 M Suc. Samples were frozen in a Baltec HPM 010 high-pressure freezer (Technotrade) and transferred into liquid nitrogen for storage. Freeze substitution and sample embedding were performed as described in Otegui and Staehelin (2004). Images were collected using a Philips CM120 scanning transmission electron microscope.

DNA and RNA Isolation, PCR Genotyping, and RT-PCR

Supplemental Data

Genomic DNA was isolated from leaves using the individual plant method (Krysan et al., 1996). Total RNA was isolated from soil-grown plants and purified by LiCl precipitation (Rapp et al., 1992) and DNAse RQ1 digestion (Promega) or by the RNeasy plant mini protocol (Qiagen). For PCR genotyping, the following primer combinations were used: (1) wild-type copy of UBP3 (UBP3-2 1 UBP3-4 pairs); (2) wild-type copy of UBP4 or the UBP4-T transgene (UBP3/4-1 [TCCAAACTCGAGAAAGCTCTCGGCGACC] 1 UBP4-3); (3) ubp3-1 allele (UBP3-4 1 JL-202); (4) wild-type copy of UBP4 (UBP3/4-1 1 UBP4-5 [ATCGCTGGTGTAAGCGTTGATTTTCGCGA]); (5) the ubp4-1 allele (UBP4-3 1 L); and (6) the UBP3, UBP3-T, and UBP3-C32S-T transgenes (UBP3/4-1 1 UBP3-3 [GTAACCATGGTCAGTATTGCTCGAATACTC]). RT-PCR was performed according to Yan et al. (2000), using Moloney murine leukemia virus reverse transcriptase (Promega) and primers UBP3-5 (AAGGTGAACGCTACTTTGGATCGAGAA) and UBP3-6 (TTCTGCTTTCTCAGCTTTCACTTGGTC) to detect UBP3 transcripts, or primers UBP4-2 (TACCATGGGCGCGGCGGGGTCCAAACTC) and UBP4-4 (GTGCTAGCTGTACATGGTTATGTTCTTCGAAGAGAG) to detect UBP4 transcripts.

The following materials are available in the online version of this article. Supplemental Figure S1. Complementation with UBP4 transgene. Supplemental Table S1. UBP3/4 is important for pollen mitosis II.

ACKNOWLEDGMENTS We thank Dr. Mark Johnson for helpful discussions and the ABRC and the University of Wisconsin Biotech Center for clones and mutant seeds. We also thank the anonymous reviewers for their helpful comments. Received December 28, 2006; accepted September 25, 2007; published September 28, 2007.

LITERATURE CITED Complementation of ubp3-1 ubp4-1 For complementation analysis of the ubp3-1 ubp4-1 mutant, transgenes encoding the full-length UBP3 or UBP4 protein or an active-site mutant of UBP3 (UBP3-C32S) were generated. For the UBP4-T transgene expressed under its own promoter, a 6-kb Acc65I-XbaI genomic DNA fragment from bacterial artificial chromosome T26C19 (ABRC) containing the UBP4 gene was ligated into the corresponding sites of the pCAMBIA 1301 plasmid (Medical Research Council Laboratory of Molecular Biology). For the UBP3-T transgene, a 1,350-bp BclI-KpnI fragment containing a full-length UBP3 cDNA (plasmid p8136; Chandler et al., 1997) was cloned downstream of a 996-bp 5# fragment of AtUBQ10 that ended just proximal to the coding sequence (p6056; Norris et al., 1993). The promoter-cDNA fragment was ligated into a BIN19derived plant transformation vector containing a 250-bp Sau3A I fragment from the nopaline synthase 3#-untranslated region. The UBP3-C24S-T transgene was identical to UBP3-T except that the Cys-32 codon (UGC) was replaced by one encoding Ser (AGC) using the mutagenic oligonucleotide CGGCAACACTAGCTACTGTAACAG and the Kunkel method (Kunkel, 1985). UBP4-T was introduced into Agrobacterium tumefaciens strain GV3101 and transformed into Ws plants containing one wild-type copy of UBP3 (UBP3/ ubp3-1; ubp4-1/ubp4-1) by the floral-dip method (Clough and Bent, 1998). Transformed seedlings were identified on Gamborg B5 agar medium containing 2% Suc and 25 mg/L hygromycin B. Hygromycin B-resistant plants containing the ubp3-1 insertion were transferred to soil and allowed to set seed. UBP3-T and UBP3-C32S-T transgenes were introduced into A. tumefaciens strain GV3101 and then into wild-type Arabidopsis ecotype Nossen-0 by the floral-dip method as above. Transformed plants were identified by kanamycin resistance. Only plants segregating 3:1 for kanamycin resistance were used in genetic crosses to generate plants harboring both the ubp3-1 and ubp4-1 mutation and the UBP3-T or UBP3-C32S-T transgenes.

Pollen Morphology and Germination Assays Pollen grains from open flowers were stained with 0.25 mg/mL DAPI for analysis of pollen morphology and nuclei number. Individual pollen grains and pollen tetrads were viewed using a fluorescent microscope equipped with the DAPI filter set. Pollen germination was assayed by suspension of pollen tetrads in medium containing 5 mM MES (pH 5.8), 1 mM KCl, 10 mM CaCl2, 0.8 mM MgSO4, 1.5 mM boric acid, 2% Suc, and 24% polyethylene glycol, which was modified from the basic germination medium reported by Fan et al. (2001). Pollen was germinated for 4 h at room temperature prior to the assessment of germination efficiency.

Plant Physiol. Vol. 145, 2007

Aguilar-Henonin L, Bravo J, Guzman P (2006) Genetic interactions of a putative Arabidopsis thaliana ubiquitin ligase with components of the Saccharomyces cerevisiae ubiquitin machinery. Curr Genet 50: 257–268 Al-Sady B, Ni W, Kircher S, Schafer E, Quail PH (2006) Photoactivated phytochrome induces rapid PIF3 phosphorylation prior to proteasomemediated degradation. Mol Cell 23: 439–446 Amerik AY, Hochstrasser M (2004) Mechanism and function of deubiquitinating enzymes. Biochim Biophys Acta 1695: 189–207 Amerik AY, Li SJ, Hochstrasser M (2000) Analysis of the deubiquitinating enzymes of the yeast Saccharomyces cerevisiae. Biol Chem 381: 981–992 Amerik A, Swaminathan S, Krantz BA, Wilkinson KD, Hochstrasser M (1997) In vivo disassembly of free polyubiquitin chains by yeast Ubp14 modulates rates of protein degradation by the proteasome. EMBO J 16: 4826–4838 Bies J, Wolff L (1997) Oncogenic activation of c-Myb by carboxyl-terminal truncation leads to decreased proteolysis by the ubiquitin-26S proteasome pathway. Oncogene 14: 203–212 Binder BM, Gagne JM, Walker JM, Emborg TJ, Hemmann G, Bleecker AB, Vierstra RD (2007) The Arabidopsis EIN3-binding F-box proteins, EBF1 and EBF2, have distinct but overlapping roles in ethylene signaling. Plant Cell 19: 509–523 Boase NA, Kelly JM (2004) A role for creD, a carbon catabolite repression gene from Aspergillus nidulans, in ubiquitination. Mol Microbiol 53: 929–940 Boisson B, Giglione C, Meinnel T (2003) Unexpected protein families including cell defense components feature in the N-myristoylome of a higher eukaryote. J Biol Chem 278: 43418–43429 Cadavid AL, Ginzel A, Fischer JA (2000) The function of the Drosophila fat facets deubiquitinating enzyme in limiting photoreceptor cell number is intimately associated with endocytosis. Development 127: 1727–1736 Callis J, Bedinger P (1994) Developmentally regulated loss of ubiquitin and ubiquitinated proteins during pollen maturation in maize. Proc Natl Acad Sci USA 91: 6074–6077 Callis JA, Carpenter TB, Sun CW, Vierstra RD (1995) Structure and evolution of genes encoding polyubiquitin and ubiquitin-like proteins in Arabidopsis thaliana ecotype Columbia. Genetics 139: 921–939 Capron A, Okresz L, Genschik P (2003) First glance at the plant APC/C, a highly conserved ubiquitin-protein ligase. Trends Plant Sci 8: 83–89 Castro A, Bernis C, Vigneron S, Labbe JC, Lorca T (2005) The anaphasepromoting complex: a key factor in the regulation of cell cycle. Oncogene 24: 314–325 Chandler JS, McArdle B, Callis J (1997) AtUBP3 and AtUBP4 are two closely related Arabidopsis thaliana ubiquitin-specific proteases present in the nucleus. Mol Gen Genet 255: 302–310

811

Doelling et al.

Clough RC, Vierstra RD (1997) Phytochrome degradation. Plant Cell Environ 20: 713–721 Clough SJ, Bent AF (1998) Floral dip: a simplified method for Agrobacteriummediated transformation of Arabidopsis thaliana. Plant J 16: 735–743 Devoto A, Nieto-Rostro M, Xie D, Ellis C, Harmston R, Patrick E, Davis J, Sherratt L, Coleman M, Turner JG (2002) COI1 links jasmonate signalling and fertility to the SCF ubiquitin-ligase complex in Arabidopsis. Plant J 32: 457–466 Doelling JH, Yan N, Kurepa J, Walker J, Vierstra RD (2001) The ubiquitinspecific protease UBP14 is essential for early embryo development in Arabidopsis thaliana. Plant J 27: 393–405 Dong CH, Agarwal M, Zhang Y, Xie Q, Zhu JK (2006) The negative regulator of plant cold responses, HOS1, is a RING E3 ligase that mediates the ubiquitination and degradation of ICE1. Proc Natl Acad Sci USA 103: 8281–8286 Durbarry A, Vizir I, 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 Fan LM, Wang YF, Wang H, Wu WH (2001) In vitro Arabidopsis pollen germination and characterization of the inward potassium currents in Arabidopsis pollen grain protoplasts. J Exp Bot 52: 1603–1614 Fulop K, Tarayre S, Kelemen Z, Harvath G, Kevei Z, Nikovics K, Bako L, Brown S, Kondorosi A, Kondorosi E (2005) Arabidopsis anaphasepromoting complexes. Cell Cycle 4: 1084–1092 Gagne JM, Downes BP, Shiu SH, Durski AM, Vierstra RD (2002) The F-box subunit of the SCF E3 complex is encoded by a diverse superfamily of genes in Arabidopsis. Proc Natl Acad Sci USA 99: 11519–11524 Genschik P, Criqui MC, Parmentier Y, Derevier A, Fleck J (1998) Cell cycle-dependent proteolysis in plants: identification of the destruction box pathway and metaphase arrest produced by the proteasome inhibitor MG132. Plant Cell 10: 2063–2076 Gingerich DJ, Gagne JM, Salter DW, Hellmann H, Estelle M, Ma L, Vierstra RD (2005) Cullins 3a and 3b assemble with members of the broad complex/tramtrack/bric-a-brac (BTB) protein family to form essential ubiquitin-protein ligases (E3s) in Arabidopsis. J Biol Chem 280: 18810–18821 Grini PE, Schnittger A, Schwarz H, Zimmermann I, Schwab B, Jurgens G, Hulskamp M (1999) Isolation of ethyl methanesulfonate-induced gametophytic mutants in Arabidopsis thaliana by a segregation distortion assay using the multimarker chromosome 1. Genetics 151: 849–863 Hansen-Hagge TE, Janssen JW, Hameister H, Papa FR, Zechner U, Seriu T, Jauch A, Becke D, Hochstrasser M, Bartram CR (1998) An evolutionarily conserved gene on human chromosome 5q33-q34, UBH1, encodes a novel deubiquitinating enzyme. Genomics 49: 411–418 Hoecker U (2005) Regulated proteolysis in light signaling. Curr Opin Plant Biol 8: 469–476 Hofmann RM, Pickart CM (1999) Noncanonical MMS2-encoded ubiquitinconjugating enzyme functions in assembly of novel polyubiquitin chains for DNA repair. Cell 96: 645–653 Honys D, Twell D (2004) Transcriptome analysis of haploid male gametophyte development in Arabidopsis. Genome Biol 5: R85 Iwakawa H, Shinmyo A, Sekine M (2006) Arabidopsis CDKA;1, a cdc2 homologue, controls proliferation of generative cells in male gametogenesis. Plant J 45: 819–831 Johnson MA, von Besser K, Zhou Q, Smith E, Aux G, Patton D, Levin JZ, Preuss D (2004) Arabidopsis hapless mutations define essential gametophytic functions. Genetics 168: 971–982 Kim ST, Zhang K, Dong J, Lord EM (2006) Exogenous free ubiquitin enhances lily pollen tube adhesion to an in vitro stylar matrix and may facilitate endocytosis of SCA. Plant Physiol 142: 1397–1411 Krysan PJ, Young JC, Tax F, Sussman MR (1996) Identification of transferred DNA insertions within Arabidopsis genes involved in signal transduction and ion transport. Proc Natl Acad Sci USA 93: 8145–8150 Kunkel TA (1985) Rapid and efficient site-specific mutagenesis without phenotypic selection. Proc Natl Acad Sci USA 82: 488–492 Lalanne E, Michaelidis C, Moore JM, Gagliano W, Johnson A, Patel R, Howden R, Vielle-Calzada J-P, Grossniklaus U, Twell D (2004) Analysis of transposon insertion mutants highlights the diversity of mechanisms underlying male progamic development in Arabidopsis. Genetics 167: 1975–1986 Lockington RA, Kelly JM (2001) Carbon catabolite repression in Aspergillus nidulans involves deubiquitination. Mol Microbiol 40: 1311–1321 Maor R, Jones A, Nuhse TS, Studholme DJ, Peck SC, Shirasu K (2007)

812

Multidimensional Protein Identification Technology (MudPIT) analysis of ubiquitinated proteins in plants. Mol Cell Proteomics 6: 601–610 McCormick S (2004) Control of male gametophyte development. Plant Cell 16: S142–S153 Millard SM, Wood SA (2006) Riding the DUBway: regulation of protein trafficking by deubiquitylating enzymes. J Cell Biol 173: 463–468 Nijman SMB, Luna-Vargas MPA, Velds A, Brummelkamp TR, Dirac AMG, Sixma TK, Bernards R (2005) A genomic and functional inventory of deubiquitinating enzymes. Cell 123: 773–786 Norris SR, Meyer SE, Callis J (1993) The intron of Arabidopsis thaliana polyubiquitin genes is conserved in location and is a quantitative determinant of chimeric gene expression. Plant Mol Biol 21: 895–906 Nowack MK, Grini PE, Jakoby MJ, Lafos M, Koncz C, Schnittger A (2006) A positive signal from the fertilization of the egg cell sets off endosperm proliferation in angiosperm embryogenesis. Nat Genet 38: 63–67 Osterlund MT, Hardtke CS, Wei N, Deng XW (2000) Targeted destabilization of HY5 during light-regulated development of Arabidopsis. Nature 405: 462–466 Otegui MS, Staehelin LA (2004) Electron tomographic analysis of postmeiotic cytokinesis during pollen development in Arabidopsis thaliana. Planta 218: 501–515 Parry G, Estelle M (2006) Auxin receptors: a new role for F-box proteins. Curr Opin Cell Biol 18: 152–156 Peng J, Schwartz D, Elias JE, Thoreen CC, Cheng D, Marsischky G, Roelofs J, Finley D, Gygi SP (2003) A proteomics approach to understanding protein ubiquitination. Nat Biotechnol 21: 921–926 Preuss D, Rhee SY, Davis RW (1994) Tetrad analysis possible in Arabidopsis with mutation of the QUARTET (QRT) genes. Science 264: 1458–1460 Qiao H, Wang H, Zhao L, Zhou J, Huang J, Zhang Y, Xue Y (2004) The F-box protein AhSLF-S2 physically interacts with S-RNases that may be inhibited by the ubiquitin/26S proteasome pathway of protein degradation during compatible pollination in Antirrhinum. Plant Cell 16: 582–595 Rao-Naik C, Chandler JS, McArdle B, Callis J (2000) Ubiquitin-specific proteases from Arabidopsis thaliana: cloning of AtUBP5 and analysis of substrate specificity of AtUBP3, AtUBP4, and AtUBP5 using Escherichia coli in vivo and in vitro assays. Arch Biochem Biophys 379: 198–208 Rapp JC, Baumgartner BJ, Mullet J (1992) Quantitative analysis of transcription and RNA levels of 15 barley chloroplast genes: transcription rates and mRNA levels vary over 300-fold, predicted mRNA stabilities vary 30-fold. J Biol Chem 267: 21404–21411 Rotman N, Durbarry A, Wardle A, Yang WC, Chaboud A, Faure JE, Berger F, Twell D (2005) A novel class of MYB factors controls sperm-cell formation in plants. Curr Biol 15: 244–248 Seo HS, Yang JY, Ishikawa M, Bolle C, Ballesteros ML, Chua NH (2003) LAF1 ubiquitination by COP1 controls photomorphogenesis and is stimulated by SPA1. Nature 423: 995–999 Sigismund S, Polo S, Di Fiore PP (2004) Signaling through monoubiquitination. Curr Top Microbiol Immunol 286: 149–185 Smalle J, Vierstra RD (2004) The ubiquitin 26S proteasome proteolytic pathway. Annu Rev Plant Biol 55: 555–590 Speranza A, Scoccianti V, Crinelli R, Calzoni GL, Magnani M (2001) Inhibition of proteasome activity strongly affects kiwifruit pollen germination. Involvement of the ubiquitin/proteasome pathway as a major regulator. Plant Physiol 126: 1150–1161 Sridhar VV, Kapoor A, Zhang K, Zhu J, Zhou T, Hasegawa PM, Bressan RA, Zhu JK (2007) Control of DNA methylation and heterochromatic silencing by histone H2B deubiquitination. Nature 447: 735–738 Stone SL, Anderson EM, Mullen RT, Goring DR (2003) ARC1 is an E3 ubiquitin ligase and promotes the ubiquitination of proteins during the rejection of self-incompatible Brassica pollen. Plant Cell 15: 885–898 Stone SL, Hauksdottir H, Troy A, Herschleb J, Kraft E, Callis J (2005) Functional analysis of the RING-type ubiquitin ligase family of Arabidopsis. Plant Physiol 137: 13–30 Stone SL, Williams L, Farmer LM, Vierstra RD, Callis J (2006) KEEP ON GOING, a RING E3 ligase essential for Arabidopsis growth and development, is involved in abscisic acid signaling. Plant Cell 18: 3415–3428 Sun TP, Gubler F (2004) Molecular mechanism of gibberellin signaling in plants. Annu Rev Plant Biol 55: 197–223 Sussman MR, Amasino RM, Young JC, Krysan PJ, Austin-Phillips S (2000) The Arabidopsis knockout facility at the University of WisconsinMadison. Plant Physiol 124: 1465–1467

Plant Physiol. Vol. 145, 2007

Deubiquitination Regulates Pollen Development

Tzafrir I, McElver JA, Liu C, Yang LJ, Wu JQ, Martinez A, Patton DA, Meinke DW (2002) Diversity of TITAN functions in Arabidopsis seed development. Plant Physiol 128: 38–51 Ushijima K, Sassa H, Dandekar AM, Gradziel TM, Tao R, Hirano H (2003) Structural and transcriptional analysis of the self-incompatibility locus of almond: identification of a pollen-expressed F-box gene with haplotypespecific polymorphism. Plant Cell 15: 771–781 van der Horst A, de Vries-Smits AMM, Brenkman AB, van Triest MH, van den Broek N, Colland F, Maurice MM, Burgering BMT (2006) FOXO4 transcriptional activity is regulated by monoubiquitination and USP7/ HAUSP. Nat Cell Biol 8: 1064–1073 Wang Y, Satoh A, Warren G, Meyer HH (2004) VCIP135 acts as a deubiquitinating enzyme during p97-p47-mediated reassembly of mitotic Golgi fragments. J Cell Biol 164: 973–978 Wee S, Geyer RK, Toda T, Wolf DA (2005) CSN facilitates Cullin-RING ubiquitin ligase function by counteracting autocatalytic adapter instability. Nat Cell Biol 7: 387–391 Wilkinson KD (1997) Regulation of ubiquitin-dependent processes by deubiquitinating enzymes. FASEB J 11: 1245–1256 Wilkinson KD (2000) Ubiquitination and deubiquitination: targeting of

Plant Physiol. Vol. 145, 2007

proteins for degradation by the proteasome. Semin Cell Dev Biol 11: 141–148 Yan N, Doelling JH, Falbel TG, Durski AM, Vierstra RD (2000) The ubiquitin-specific protease family from Arabidopsis. AtUBP1 and 2 are required for the resistance to the amino acid analog canavanine. Plant Physiol 124: 1828–1843 Yang M, Hu Y, Lodhi M, McCombie WR, Ma H (1999) The Arabidopsis SKP1-LIKE1 gene is essential for male meiosis and may control homologue separation. Proc Natl Acad Sci USA 96: 11416–11421 Zeng LR, Vega-Sanchez ME, Zhu T, Wang GL (2006) Ubiquitinationmediated protein degradation and modification: an emerging theme in plant-microbe interactions. Cell Res 16: 413–426 Zhang D, Zaugg K, Mak T, Elledge S (2006) A role for the deubiquitinating enzyme USP28 in control of the DNA-damage response. Cell 126: 529–542 Zhang X, Garreton V, Chua NH (2005) The AIP2 E3 ligase acts as a novel negative regulator of ABA signaling by promoting ABI3 degradation. Genes Dev 19: 1532–1543 Zimmermann P, Hirsch-Hoffmann M, Hennig L, Gruissem W (2004) GENEVESTIGATOR. Arabidopsis microarray database and analysis tool box. Plant Physiol 136: 2621–2632

813