Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 7-Day Trial for You or Your Team.

Learn More →

AtRabD2b and AtRabD2c have overlapping functions in pollen development and pollen tube growth

AtRabD2b and AtRabD2c have overlapping functions in pollen development and pollen tube growth Background: Rab GTPases are important regulators of endomembrane trafficking, regulating exocytosis, endocytosis and membrane recycling. Many Rab-like proteins exist in plants, but only a subset have been functionally characterized. Results: Here we report that AtRabD2b and AtRabD2c play important roles in pollen development, germination and tube elongation. AtrabD2b and AtrabD2c single mutants have no obvious morphological changes compared with wild-type plants across a variety of growth conditions. An AtrabD2b/2c double mutant is also indistinguishable from wild-type plants during vegetative growth; however its siliques are shorter than those in wild-type plants. Compared with wild-type plants, AtrabD2b/2c mutants produce deformed pollen with swollen and branched pollen tube tips. The shorter siliques in the AtrabD2b/2c double mutant were found to be primarily due to the pollen defects. AtRabD2b and AtRabD2c have different but overlapping expression patterns, and they are both highly expressed in pollen. Both AtRabD2b and AtRabD2c protein localize to Golgi bodies. Conclusions: These findings support a partially redundant role for AtRabD2b and AtRabD2c in vesicle trafficking during pollen tube growth that cannot be fulfilled by the remaining AtRabD family members. Background factor attachment protein receptor (SNARE) proteins to Ras-like small GTP-binding proteins (GTPases) regulate promote specificity of vesicle transport to target com- diverse processes in eukaryotic cells including signal partments and facilitate vesicle and target membrane transduction, cell proliferation, cytoskeletal organization fusion [7-13]. They are therefore essential for the trans- and intracellular membrane trafficking. GTPases are port of proteins and membrane through the endomem- activated by GTP binding and inactivated by subsequent brane system to their destination. hydrolysis of bound GTP to GDP, thus acting as mole- The Arabidopsis thaliana genome encodes 93 putative cular switches in these processes [1,2]. The Rab GTPase Ras superfamily proteins. Fifty-seven of these are Rab family is the largest and most complex within the Ras GTPases, more than in yeast but similar to the number protein superfamily. Rab GTPases are important regula- in humans [13,14]. According to their sequence similar- tors of endomembrane trafficking, regulating exocytosis, ity and phylogenetic clustering with yeast and mamma- endocytosis and membrane recycling processes in eukar- lian orthologs, these Rab proteins were assigned to eight yotic cells [3-6]. Rab GTPase functions have been exten- subfamilies, AtRabA to AtRabH, which can be further sively studied in yeast and mammalian systems. Both in divided into 18 subclasses [13]. Relatively few of the vivo and in vitro experiments have demonstrated that plant Rab orthologs have been investigated functionally. different Rab proteins function in distinct intracellular Most of these studies have used constitutively active membrane trafficking steps and they are hypothesized to (CA) and/or dominant negative (DN) mutations, gener- work together with soluble N-ethylmaleimide-sensitive ated by direct mutation of the conserved domain to restrict mutant GTPase proteins to the active GTP- * Correspondence: bassham@iastate.edu bound form (constitutively active) or inactive GDP- Department of Genetics, Development and Cell Biology, Iowa State bound form (dominant negative). Expression of CA or University, Ames, IA 50010, USA © 2011 Peng et al; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Peng et al. BMC Plant Biology 2011, 11:25 Page 2 of 16 http://www.biomedcentral.com/1471-2229/11/25 DN Rab GTPases can perturb the activity of the endo- RNA accumulation level across a variety of experimen- genous Rab, revealing their functional significance. For a tal conditions. Phenotypic analysis of knockout number of plant Rab GTPases, expression of their CA mutants indicates that they are at least partially func- and DN mutants in transformed plants, together with tionally redundant, and are important in pollen devel- protein localization information, has shown that these opment and pollen tube growth. The proteins both Rabs perform functions similar to those of their yeast localize to the trans-Golgi, consistent with their pro- posed role in trafficking from the ER to the Golgi and mammalian orthologs [15-19]. apparatus. Several reports indicate that Rab proteins are impor- tant for elongation of tip-growing cells in plants. For example, AtRabA4b is reported to localize to the tips of Results root hair cells and was proposed to regulate membrane The expression patterns of AtRabD2b and AtRabD2c are trafficking through a compartment involved in the closely correlated polarized secretion of cell wall components [18]. NtRab2 The four RabD family members in Arabidopsis share GTPase is important for trafficking between the endo- about 88% identity at the amino acid level. The accu- plasmic reticulum and Golgi bodies in tobacco pollen mulation pattern of the associated transcripts is quite tubes and may be specialized to optimally support the distinct across a wide variety of experimental condi- high secretory demands in these tip growing cells [16]. tions and developmental stages (MetaOmGraph, http:// NtRabA (Rab11) in tobacco is predominantly localized www.metnetdb.org/MetNet_MetaOmGraph.htm; [26]) to an inverted cone-shaped region at the pollen tube tip, (Table 1; Additional file 1, Table S1). AtRabD2b and and both constitutively active and dominant negative AtRabD2c expression patterns are correlated (at a mutants resulted in reduced tube growth rate, meander- Pearson correlation value of 0.72), whereas AtRabD1 ing pollen tubes, and reduced male fertility [20]. and AtRabD2a show very low correlation with the There are four genes in the Arabidopsis RabD subfam- others (Pearson correlation value of < 0.20). Based on ily, AtRabD1 (At3g11730), AtRabD2a (At1g02130, their high sequence similarity (99% amino acid iden- AtRab1b), AtRabD2b (At5g47200, AtRab1a) and tity) and the correlation between their mRNA accumu- AtRabD2c (At4g17530, AtRab1c) [13]. In mammals, the lation patterns, we hypothesized that AtRabD2b and orthologs of AtRabD, Rab1 isoforms, physically associate AtRabD2c might have some functional overlap that is with the ER, ER-Golgi intermediate compartment and not shared by AtRabD1 and AtRabD2a. Golgi and regulate membrane trafficking between the ER and Golgi complex [21]. Fluorescent protein fusions Identification of Null Mutations in the Genes AtRabD2b with AtRabD1, AtRabD2a and AtRabD2b localize to the and AtRabD2c Golgi and trans-Golgi network [22,23], and transient It was reported previously that an AtrabD2b AtrabD2c expression in plant cells of dominant negative mutants double mutant has no phenotype [22]. Based on our of rabD2a or rabD1 resulted in the inhibition of ER-to- correlation analysis above, we hypothesized that this Golgi trafficking [15,22,24], suggesting a related function mutant may have some more subtle defects that cannot for the plant Rab1 homologs. Pinheiro et al. [22] isolated be compensated for by the remaining family members. T-DNA insertion mutants in each of the AtRabD family To investigate this further, we identified T-DNA inser- genes and reported that each of the single and double tion mutants (Figure 1A) in AtrabD2b (3 alleles) and mutants lacked a detectable phenotype. By contrast, a AtrabD2c (1 allele). Homozygous lines for the T-DNA rabD2a rabD2b rabD2c triple mutant was lethal and a insertions were identified by PCR, using primers rabD1 rabD2b rabD2c triple mutant had stunted growth selected by iSct primers (http://signal.salk.edu/tdnapri- and low fertility, indicating that these gene family mem- mers.2.html), and the insertion sites were determined by bers perform important and overlapping functions. sequencing the PCR products (Figure 1B). Analysis of We previously hypothesized that closely related mRNA levels by RT-PCR indicated that AtrabD2b-1, genes with a high Pearson correlation in their RNA AtrabD2b-2 and AtrabD2c-1 are null mutants. However, accumulation level are functionally redundant, and showed that expression patterns of both the AtRabD2b Table 1 Pearson correlation between expression patterns and AtRabD2c genes are negatively correlated with the of AtRabD family members process of starch synthesis [25], whereas the expression AtRabD1 AtRabD2a AtRabD2b AtRabD2c patterns of the remaining RabD genes are not. We AtRabD1 100% therefore predicted that these two Rab proteins may AtRabD2a 17.08% 100% have redundant functions that are not shared by the AtRabD2b -3.4% 22.19% 100% other two AtRabD family members. Here we show that AtRabD2c 15.92% 25.23% 77.81% 100% AtRabD2b and AtRabD2c are highly correlated in their Peng et al. BMC Plant Biology 2011, 11:25 Page 3 of 16 http://www.biomedcentral.com/1471-2229/11/25 single mutants were crossed, F1 plants were allowed to AtrabD2b-1 self fertilize and the AtrabD2b-2/AtrabD2c-1 double ATG TGA mutant was identified from the F2 population by PCR using the primers for both AtrabD2b-2 and AtrabD2c-1. 100bp Hereafter, the AtrabD2b-2/AtrabD2c-1 double mutant A tRabD2b AtrabD2b-3 AtrabD2b-2 2055bp will be referred to as AtrabD2b/2c, and AtrabD2b-2 and ATG TAA AtrabD2c-1 single mutants will be referred to as AtrabD2b and AtrabD2c respectively. A tRabD2c 100bp 2060bp Siliques Are Shorter in the AtrabD2b/2c Double Mutant AtrabD2c-1 than in Either Single Mutant or in Wild-Type Lines Col-0 AtrabD2c AtrabD2b AtrabD2b/2c To evaluate phenotypes associated with the AtrabD2b M M 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 and AtrabD2c mutants, homozygous AtrabD2b (three alleles, AtrabD2b-1, AtrabD2b-2 and AtrabD2b-3), AtrabD2c and AtrabD2b/2c mutants, along with wild- type siblings, were grown on agar plates with or without various hormone, nutrient and light treatments. We tested over 50 of the conditions described in the Gantlet C website (http://www.gantlet.org); however, no significant phenotypic differences were observed in the seedlings for any of the mutant alleles (data not shown). In addition, we tested the seedling phenotype on media with or with- AtRabD2c out sucrose or vitamin B5 and, consistent with previous reports [22], no obvious phenotypes were observed. By contrast, AtrabD2b/2c double mutant lines AtRabD2b showed a phenotype associated with reproduction. In these lines, siliques were shorter when grown either under continuous light or long day (16h light/8h dark) Tubulin conditions. Neither the AtrabD2b nor the AtrabD2c single mutant alleles displayed a short silique pheno- Figure 1 Characterization of AtrabD2b and AtrabD2c mutations. type. The length of AtrabD2b/2c siliques was 70% of A, Gene map. The scaled linear map depicts the 8 exons as boxes that of wild-type, AtrabD2b or AtrabD2c single mutant and the 7 introns as lines between the boxes for both the lines (Figure 2; P < 0.01 by Student’s t-test). To evalu- AtRabD2b and AtRabD2c genes. The positions of the translational ate whether this reduced silique size is associated with start and stop codons in exon 1 and exon 8, respectively, are noted. The locations of the T-DNA insertions (not drawn to scale) in the a seed defect, siliques from AtrabD2b/2c, wild-type, genes are indicated. B, Genotypes of T-DNA insertion mutants. AtrabD2b and AtrabD2c mutant lines were opened at Genomic DNA was isolated from the indicated single and double 10 DAF (days after flowering). Consistently, no defects mutants and amplified by PCR. Primer pairs used were as following: in the seeds of either AtrabD2b or AtrabD2c single lane 1, AtrabD2c-LP1 and AtrabD2c-RP1; lane 2, AtrabD2c-RP1 and LBb1; lane 3, AtrabD2b-LP1 and AtrabD2b-RP1; lane 4, AtrabD2b-RP1 mutants were observed. However, approximately half and LBb1. C, Analysis of transcripts from AtrabD2b-1, AtrabD2c-1 and of the ovules in the AtrabD2b/2c double mutant were AtrabD2b/2c mutants. Total RNA from leaves of wild-type plants, not fertilized (Figure 3). Consistent with this observa- AtrabD2c-1, AtrabD2b-1 and AtrabD2b/2c was amplified by RT-PCR. tion, the AtrabD2b/2c mutant plants produced a smal- Primer pairs for AtRabD2c were AtRabD2c-F and AtRabD2c-R, primer ler quantity of seeds than wild-type plants or single pairs for AtRabD2b were AtRabD2b-F and AtRabD2b-R. Tubulin was used as control. mutants (Figure 3; Additional file 2, Figure S1). These results are consistent with a functional overlap between AtRabD2b and AtRabD2c that cannot be ful- the AtrabD2b-3 mutation had no effect on AtRabD2b filled by AtRabD1 or AtRabD2a. RNA accumulation (Figure 1C and data not shown). Progeny from AtrabD2b-1 and AtrabD2c-1 heterozy- Complementation of AtrabD2b/2c Mutant Phenotype gotes showed a T-DNA segregation ratio of approxi- To demonstrate that the AtrabD2b/2c mutant phenotype mately 3:1 based on kanamycin resistance, consistent is due to the mutations in the AtRabD2b and AtRabD2c with a single insertion. AtrabD2b-2 wassuppliedasa genes, constructs containing either AtRabD2b or homozygous line. To generate AtrabD2b AtrabD2c dou- AtRabD2c, each expressed from their native promoter, ble mutants, AtrabD2b-2 and AtrabD2c-1 homozygous were introduced into the AtrabD2b/2c double mutant. AtrabD2c Col-0 AtrabD2b/2c AtrabD2b Peng et al. BMC Plant Biology 2011, 11:25 Page 4 of 16 http://www.biomedcentral.com/1471-2229/11/25 Figure 2 The AtrabD2b/2c double mutant shows a striking shorter silique phenotype. A, Vegetative growth of AtrabD2b, AtrabD2c and AtrabD2b/2c plants. B, Inflorescence of AtrabD2b/2c and wild-type plants. Scale bars = 850 μm. C, Siliques from the AtrabD2b/2c mutant and wild-type plants; arrows indicate the sequence of siliques from the oldest to the youngest. Scale bars = 850 μm. D, Siliques (from 6 to 14 ) of the first inflorescence for wild type, single and double mutants were measured for each plant, with 10 plants measured for each genotype. Error bars indicate standard deviation. Peng et al. BMC Plant Biology 2011, 11:25 Page 5 of 16 http://www.biomedcentral.com/1471-2229/11/25 Figure 3 There are many non-fertilized ovaries in the AtrabD2b/2c double mutant. Individual siliques of wild type, single and double mutant plants were dissected and examined under the microscope. Arrows heads indicate unfertilized ovaries. Inset, seeds produced by a single plant. Scale bars = 200 μm. Both constructs were able to rescue the silique length pollen visualized under the SEM. Surprisingly, even the phenotype of the mutant (Figure 4A, C) and restored the AtrabD2b and AtrabD2c single mutant lines produce seed fertilization defect (Figure 4B) and seed number aberrant pollen at a level of about 10%. This is unex- (Additional file 2, Figure S1), confirming that the loss of pected, as the AtrabD2b and AtrabD2c single mutants AtRabD2b and AtRabD2c is responsible for these have normal-appearing siliques and seed quantities simi- phenotypes. lar to the wild-type plants. A likely explanation is that there are sufficient normal pollen grains in the single AtrabD2b/2c, AtrabD2b and AtrabD2c Pollen Have Defects mutants to efficiently fertilize the ovaries in the in Morphology and Pollen Tube Elongation AtrabD2b and AtrabD2c single mutants. Two possibilities could explain the unfertilized embryos We originally identified AtRabD2b and AtRabD2c seen in the AtrabD2b/2c double mutants. One possibi- because the transcript accumulation patterns of these lity is that the pollen bears a defect that leads to pollen two genes correlate with those of many genes associated sterility and inability to fertilize the embryos. Alterna- with starch metabolism. Indeed, the AtrabD2b/2c double tively, ovules may bear an abnormality such that their mutant pollen stained less intensely with IKI than wild- fertilization is reduced. To distinguish between these type pollen (Figure 5C), suggesting a decreased starch two possibilities, we observed the pollen by scanning content in the AtrabD2b/2c mutant pollen. This is con- sistent with the expression correlation, although the rea- electron microscopy (SEM). All of the pollen from wild- son for this phenotype is unclear. type plants looked normal, whereas more than 20% of the AtrabD2b/2c pollen exhibited an irregular, collapsed A single flower of Arabidopsis produces thousands of morphology (Figure 5A). We also observed that some pollen grains, but usually there are less than 100 abnormal pollen grains from the AtrabD2b/2c double embryos in one silique. If only 20% of the pollen grains mutant were devoid of nuclei, as indicated by DAPI are abnormal, we would not expect the strikingly staining, whereas all pollen from wild-type (Figure 5B) reduced fertility seen in the AtrabD2b/2c double andsinglemutantplants(data notshown)havenuclei. mutant. We therefore looked for additional explanations This defective pollen may be the severely collapsed for the reduced fertility. To evaluate germination and Peng et al. BMC Plant Biology 2011, 11:25 Page 6 of 16 http://www.biomedcentral.com/1471-2229/11/25 and 6B). Moreover, the pollen tubes of the AtrabD2b/2c double mutant were much shorter than those of wild- type plants or either single mutant (P < 0.01), and the single mutants had shorter pollen tubes than wild-type plants (Figure 6E; P < 0.01 for both mutants). Even though the AtrabD2b and AtrabD2c single mutants had collapsed pollen, shorter pollen tubes and swollen tips, their siliques were normal compared with wild-type plants. We hypothesize that the single mutants may still have sufficient normal pollen to enable all embryos to be fertilized. The in vitro pollen germination phenotypes were confirmed by analyzing pollen tube growth after in vivo pollination (Figure 6C). Open flowers from wild- type or AtrabD2b/2c mutant plants were incubated overnight on agar medium. The AtrabD2b/2c mutant flowers had reduced pollen germination and decreased pollen tube length compared with wild-type plants, sug- gesting that pollen germination and pollen tube growth may also be defective in vivo. Pollen and Pollen Tube Defects Cause the Shorter Siliques in the AtrabD2b/2c Mutant To investigate whether the unfertilized seeds are due to the observed pollen abnormality, or whether the ovary also has defects that might contribute to the reduced rate of fertilization, we crossed wild-type and AtrabD2b/ 2c mutant plants. If the shorter silique phenotype is borne only by the abnormal pollen, wild-type plant pol- len should rescue the AtrabD2b/2c mutant silique phe- Figure 4 Complementation of the double mutant phenotype. notype to a normal length (AtrabD2b/2c mutant female A, Siliques are shown from wild-type plants, AtrabD2b and AtrabD2c flower crossed with wild-type plant pollen). In contrast, single mutants, the AtrabD2b/2c double mutant and the AtrabD2b/ the AtrabD2b/2c mutant plant pollen crossed with a 2c double mutant complemented with either AtRabD2b or wild-type female would mimic the mutant phenotype of AtRabD2c. Scale bars = 0.5 cm. B, Individual siliques of rescued lines were dissected and examined under the microscope. Scale bars = decreased fertilization (wild-type female flower crossed 600 μm. C, Siliques (from 6 to 14 ) of the first inflorescence for the with AtrabD2b/2c mutant pollen). Alternatively, if the indicated genotypes were measured for each plant, with 10 plants ovary also has some abnormality, wild-type pollen measured for each genotype. Error bars indicate standard deviation. would not completely rescue the mutant phenotype, and AtrabD2b/2c mutant pollen would not mimic the tube growth of the pollen grains, pollen was germinated mutant phenotype. The results of these crosses indicated in vitro. After overnight incubation, almost all of the that pollen from wild-type plants can rescue the pollen from wild-type plants germinated and showed a AtrabD2b/2c short silique phenotype, and the pollen typical tip growth. However, about 10% of the pollen from AtrabD2b/2c can bestow the shorter silique phe- from the AtrabD2b/2c mutant did not germinate at all notype on wild-type plants (Figures 7A and 7C). Specifi- and 50% of the pollen germinated but did not grow api- cally, about half of the seeds were not fertilized in the cally as did pollen of wild-type plants (Figure 6A and siliques that developed from wild-type pistils fertilized 6B). Instead, these pollen tubes were shorter and had by AtrabD2b/2c pollen (Figure 7B). In contrast, the sili- swollen tips, some burst (≈5%), and others branched ques from AtrabD2b/2c usually had about 50% unferti- (≈2%; Figure 6A,B and 6D). The germination rate of the lized ovules, but when these pistils were fertilized by pollen from the single mutants was similar to the wild- wild-type pollen, all seeds looked normal, and the sili- type pollen. However, approximately 20% of the germi- ques were longer than those siliques in the same inflor- nating pollen also had swollentips(Figure 6Aand6B), escence which were self-fertilized (Figures 7). These although the phenotype was not as severe as the results confirm that the unfertilized ovaries are mostly, AtrabD2b/2c double mutant; burst or branched tubes if not exclusively, caused by pollen defects in the were never observed in either single mutant (Figure 6A AtrabD2b/2c mutant. Peng et al. BMC Plant Biology 2011, 11:25 Page 7 of 16 http://www.biomedcentral.com/1471-2229/11/25 Figure 5 Pollen defects in AtrabD2b, AtrabD2c and AtrabD2b/D2c mutants. A, Fresh pollen was examined by SEM. B, DAPI staining of pollen. Fresh pollen grains were stained with DAPI and photographed under the fluorescence microscope. Arrow indicates a pollen grain from the AtrabD2b/2c mutant that lacks a nucleus. C, IKI staining of pollen, demonstrating reduced staining of the AtrabD2b/2c double mutant pollen compared with wild-type pollen. Scale bars = 10 μm (A); 50 μm (B,C). In silico and GUS Analysis of AtRabD2b and AtRabD2c including high expression in floral organs and particu- Expression larly in the stamen (Figure 8; [25,26]). If AtRabD2b and AtRabD2c are involved in pollen To directly examine the spatial expression pattern of development and pollen tube growth, they are expected the AtRabD2b and AtRabD2c genes, transgenic lines to be co-expressed in pollen and pollen tubes. Public containing promoter:GFP/GUS constructs for each gene microarray data indicates that both AtRabD2b and were analyzed for GUS activity at various stages of AtRabD2c are expressed throughout development, development from germination to senescence. As Peng et al. BMC Plant Biology 2011, 11:25 Page 8 of 16 http://www.biomedcentral.com/1471-2229/11/25 Figure 6 Pollen tube elongation defects in AtrabD2b, AtrabD2c and AtrabD2b/2c mutants. A, Pollen was germinated in vitro for 6 hours and examined by SEM. B, Germinated pollen was stained with aniline blue then observed under an epifluorescence microscope. C. Open flowers from an AtrabD2b/2c mutant plant, along with a wild-type plant, were incubated overnight on medium then examined by fluorescence microscopy. D. Close up view of pollen tubes in the AtrabD2b/2c mutant. E. Pollen was germinated in vitro and pollen tube length measured after an overnight incubation using SIS Pro software (OSIS, Lakewood, CO) (n > 200). Error bars indicate standard deviation. Scale bars = 10 μm (A); 50 μm (B, C); 20 μm (D). indicated by the in silico analyses, both AtRabD2b and only expressed in the trichomes, while AtRabD2b was AtRabD2c were expressed widely during development. not expressed in these cells (Figure 9A). In flowers, GUS staining further indicated that in cotyledons, AtRabD2b was expressed in sepals, stamen and stigma, rosette leaves and cauline leaves, AtRabD2b expression while AtRabD2c was expressed in sepal, stamen, stigma was localized predominantly in vascular tissues (Figure and style (Figure 9E, F). This dichotomy of expression 9B), whereas AtRabD2c was expressed ubiquitously in suggests that AtRabD2b and AtRabD2c may function cotyledons and in mature leaves throughout the entire independently of each other in certain cells. Both genes leaf. Interestingly, in emerging leaves, AtRabD2c was were expressed in pollen grains and germinating pollen Peng et al. BMC Plant Biology 2011, 11:25 Page 9 of 16 http://www.biomedcentral.com/1471-2229/11/25 Figure 7 Wild-type pollen can restore the shorter siliques of the AtrabD2b/2c mutant to normal length.Wild-typeand AtrabD2b/2c double mutant plants were crossed and silique length measured after 10 days. A. Inflorescences from a cross between a wild-type plant and AtrabD2b/2c mutant. The blue arrow indicates a silique in which a wild-type pistil was fertilized with AtrabD2b;AtrabD2c pollen. The red arrow indicates a silique in which the AtrabD2b/2c mutant pistil was fertilized with wild-type pollen. B. Siliques from the crosses at 10 DAP (days after pollination) were dissected and examined under a stereo microscope. White arrowheads indicate unfertilized embryos found upon pollination of wild-type plants with AtrabD2b/2c pollen. C, More than 20 siliques were measured for each plant. Error bars indicate standard deviation. Scale bars= 850 μm (A); 500 μm (B). Peng et al. BMC Plant Biology 2011, 11:25 Page 10 of 16 http://www.biomedcentral.com/1471-2229/11/25 Signal Intensity Figure 8 In silico expression analysis of AtRabD2b and AtRabD2c. The spatial and temporal expression profiles of AtRabD2b and AtRabD2c were analyzed using Genevestigator anatomy (A) and development (B) tools, respectively. Numbers along the X axis represent the developmental stage: 1, germinated seed; 2, seedlings; 3, young rosette; 4, developed rosette; 5, bolting; 6, young flower; 7, developed flower; 8, flowers and siliques; 9, mature siliques. Signal Intensity Peng et al. BMC Plant Biology 2011, 11:25 Page 11 of 16 http://www.biomedcentral.com/1471-2229/11/25 Figure 9 Temporal and spatial expression pattern of AtRabD2b and AtRabD2c. Transgenic plants were generated that express the GUS gene driven by a 954bp or 558bp fragment upstream of the AtRabD2b or AtRabD2c start codon, respectively. GUS activity (blue color) was analyzed in cotyledons, young leaves, (A, B), old leaves, roots (C, D), flowers, pistils and germinated pollen (E, F). (Figure 9E, F), which is consistent with their role in pol- Subcellular localization of AtRabD2b and AtRabD2c len development and pollen tube growth. They both In mammals, different AtRabD orthologs (Rab1 isoforms) also showed expression in roots, but AtRabD2b was ubi- are localized to the ER, ER-Golgi intermediate compart- quitously expressed throughout the root, while ment or Golgi compartments. In plants, AtRabD2a and AtRabD2c expression was excluded from root hairs and AtRabD2b are associated with the Golgi apparatus and root tips (Figure 9A-D). trans-Golgi network [15,22,23,27], and we predicted that Peng et al. BMC Plant Biology 2011, 11:25 Page 12 of 16 http://www.biomedcentral.com/1471-2229/11/25 AtRabD2c will share this localization. To determine the and Rab2 (AtRabB orthologs), have been reported to func- subcellular localization of AtRabD2c, GFP-AtRabD2b tion in membrane trafficking between the ER and Golgi in and GFP-AtRabD2c constructs were introduced into mammalian cells [21,28-34]. The plant Rab1 and Rab2 Arabidopsis leaf protoplasts and the GFP signal was homologs AtRabD1, AtRabD2a and NtRab2 have also observed by confocal microscopy. For both constructs, been reported to function in ER to Golgi vesicle transport GFP localized to punctuate structures, reminiscent of the [15,16]. Here, we demonstrate a distinct physiological role Golgi apparatus. To verify the identity of these structures, for the Rab1 homologs AtRabD2b and AtRabD2c in pollen GFP-AtRabD2b and GFP-AtRabD2c were co-transfected development and pollen tube growth. Using the bioinformatics tool MetaOmGraph (http:// into Arabidopsis leaf protoplasts with the trans-Golgi marker ST-YFP [15], and YFP and GFP signals were www.metnetdb.org) [25,26] to determine the pairwise observed. Confocal results indicated that both AtRabD2b Pearsons correlation value between the expression pat- and AtRabD2c primarily colocalized with ST-YFP (Figure terns of all of the 57 AtRab genes (Additional file 1, 10) and are therefore associated with the Golgi, consis- Table S1), we found that among the four AtRabDs, only tent with a role in Golgi trafficking. Occasionally, the expression of AtRabD2b and AtRabD2c are highly AtRabD2b or AtRabD2c-labeled structures were seen correlated. From this data, we hypothesized that that did not contain ST-YFP; these could be the post- AtRabD2b and AtRabD2c have partially redundant func- Golgi compartments described previously [22,23,27]. tions that are not shared by the remaining AtRabD Cells expressing a single GFP or YFP fusion demon- family members. To test our hypothesis, we used T- strated the absence of cross talk between GFP and YFP DNA insertion single and double mutants to confirm signals (Additional file 3, Figure S2). that AtRabD2b and AtRabD2c have functional overlap and show that they are both required for normal pollen Discussion development and tip growth of pollen tubes. We also Rab GTPases are critical players in the transport of materi- showed that they both co-localize with the trans-Golgi als through the endomembrane system, controlling exocy- marker ST-YFP, consistent with their proposed role in tosisofproteinsand cell wall materials, endocytosis of Golgi trafficking. receptors and transporters, and membrane recycling pro- The conclusion that AtRabD2b and AtRabD2c are cesses. Together with SNARE proteins, they promote spe- partially functionally redundant is based on several lines cificity of vesicle transport to target compartments, of evidence. First, although single mutant plants con- ensuring that vesicles fuse only with their appropriate tar- taining AtrabD2b or AtrabD2c mutant alleles are indis- get and thus maintaining the distinct identity of individual tinguishable morphologically from their wild-type counterparts, even when grown under a variety of organelles. Two Rab subfamilies, Rab1 (AtRabD orthologs) Figure 10 AtRabD2b and AtRabD2c are both Golgi resident proteins. Arabidopsis leaf protoplasts were co-transformed with GFP-AtRabD2b or GFP-AtRabD2c and the Golgi marker ST-YFP. GFP, YFP and chlorophyll autofluorescence were detected by confocal microcopy. At least 10 transformed protoplasts were observed for each construct. Upper panel, GFP-AtRabD2b; lower panel, GFP-AtRabD2c. Scale bar = 10 μm. Peng et al. BMC Plant Biology 2011, 11:25 Page 13 of 16 http://www.biomedcentral.com/1471-2229/11/25 growth conditions, the AtrabD2b/2c double mutant has Second, pollen tubes grow very rapidly compared with a short-silique phenotype. Second, both AtrabD2b and many other cell types. Pollen tubes elongate by tip AtrabD2c mutant plants produce a small percentage of growth, whereby the pollen cytoplasm is confined to the deformed and collapsed pollen grains, while AtrabD2b/ most proximal region of the tube, and growth is 2c lines produce a higher percentage of deformed pollen restricted to the tube apex [35]. In vitro,lilypollen grains, many of which are severely deformed, some lack- tubes grow at about 150 nm/sec [35] and Arabidopsis pollen tubes at 37 nm/sec [36]; in vivo, tobacco pollen ing nuclei. It is probable that such aberrant pollen can grow at 42 nm/sec [37]. This fast growth is contin- would give rise to defects in pollen germination, and gent on rapid vesicle trafficking to deliver large amounts indeed, though the germination rate is similar between AtrabD2b or AtrabD2c single mutants and wild-type of membrane and cell wall components to the apical plants, about 10% of the pollen grains from AtrabD2b/ region of the tubes. This extensive trafficking require- 2c double mutant plants are unable to geminate. ment may preclude the remaining Rabs from completely Furthermore, AtrabD2b and AtrabD2c mutant pollen compensating for loss of AtRabD2b and AtRabD2c. tubes do not grow apically as well as do wild-type pollen Third, computational analysis of public microarray tubes and tend to have swollen tips and a shorter data, together with studies of the expression pattern length; this phenotype is substantially more severe in directed by the AtRabD2b and AtRabD2c promoters, AtrabD2b/2c double mutants. In addition, some pollen indicated that both are widely expressed in most organs tubes from AtrabD2b/2c double mutants branch or and several cell types, with high expression in pollen. burst, which is not seen in pollen tubes of wild-type Root hairs also showed expression of AtRabD2b, and, plants or either single mutant. These data also indicate like pollen tubes, root hairs elongate by tip growth. that the loss of function of the AtRabD2b/2c genes can- However, root hair growth in the AtrabD2b/2c double not be compensated for by the AtRabD1 or AtRabD2a mutant is indistinguishable from that of wild-type genes, suggesting that either some function(s) of the plants. This is consistent with the idea that AtRabD2b AtRabD2b and AtRabD2c proteins are distinct from and AtRabD2c are required for vesicle trafficking in those of AtRabD1 or AtRabD2a, or that they are not multiple cell types, and that the highest demand for this expressed in the same cell types. process may be in pollen and pollen tubes, in order to Both AtRabD2b and AtRabD2c co-localize with the optimally support the large secretory requirement of Golgi marker ST-YFP upon transient expression in Ara- these very rapidly elongating cells. In combination, these data indicate that the high expression of AtRabD2b and bidopsis leaf protoplasts, as was reported also for AtRabD2c in pollen may be important to facilitate AtRabD2a (formerly called AtRab1b) [15]. It is therefore possible that AtRabD2b and AtRabD2c function in vesi- membrane trafficking needed for pollen tube growth. cle trafficking between the ER and Golgi apparatus, as does AtRabD2a [15,22]. Complete disruption of Rab Conclusions function in ER-to-Golgi trafficking is expected to be In summary, we used a T-DNA insertion mutant lethal, due to loss of plasma membrane, vacuole and cell approach to demonstrate the function of AtRabD2b and wall assembly and integrity. However, the AtrabD2b/2c AtRabD2c. Our data indicated that both are Golgi resi- double mutant is indistinguishable from wild-type plants, dents; they have similar but not identical expression except for shorter siliques due to the pollen and pollen patterns, but are both highly expressed in pollen; they tube defects. There are several possible explanations for are both involved in tip growth of pollen tubes; and this. First, other AtRabs must perform the same function they are at least partially functionally redundant. Future in vegetative tissue. The most likely candidates are work will focus on elucidating the molecular basis for AtRabD2a and AtRabD1, which could compensate for the pollen phenotype in the AtrabD2b/2c double thelossoffunctioninthe AtrabD2b/2c mutant in most mutant. cell types [22]. Moreover, other Rab families, such as tobacco RabBs (NtRab2s) have also been shown to be Methods regulators of membrane trafficking between the ER and Plant Materials and Growth Conditions Golgi apparatus [16]. AtRabBs (AtRab2s) may have the Wild-type Arabidopsis (Arabidopsis thaliana) ecotype same function, such that they also participate in ER to Columbia (Col-0), AtrabD2c-1, AtrabD2b-1 and Golgi vesicle trafficking. The pattern of AtRabB1b RNA AtrabD2b/2c (crosses of AtrabD2b-1 and AtrabD2c-1) accumulation is most highly correlated with that of mutants in the same genetic background were used. AtRabD2b (67%) and AtRabD2c (68%) (Additional file 1, Seeds were sown in Sunshine Soil mix, incubated at 4°C Table S1). These genes might compensate in part for the for 2 to 3 days, then grown at 22°C, 70% relative humid- loss of function of AtrabD2b/2c. ity, in a 16-h light/8-h dark photoperiod [26]. Peng et al. BMC Plant Biology 2011, 11:25 Page 14 of 16 http://www.biomedcentral.com/1471-2229/11/25 Screening for T-DNA insertion mutants In vitro pollen germination and growth measurement T-DNA insertion mutants of AtRabD2b and AtRabD2c Pollen was obtained from flowers collected from Arabi- (Salk_045030 (AtrabD2b-1), Salk_117532 (AtrabD2b-2) dopsis plants (ten plant lines per genotype) 1 to 2 weeks and Salk_120116 (AtrabD2b-3) for AtrabD2b; after bolting. Pollen from AtrabD2b/2c, AtrabD2b and Salk_054626 (AtrabD2c-3) for AtrabD2c) were obtained AtrabD2c mutants, along with pollen from wild-type from ABRC [38]. Homozygous lines for T-DNA inser- plants, was germinated on agar medium containing 18% tions were identified by PCR genotyping. For each (w/v) sucrose, 0.01% (w/v) boric acid, 1mM MgSO , T-DNA insertion mutant, two sets of PCRs were per- 1mM CaCl ,1mM Ca(NO ) , and 0.5% (w/v) agar, pH 2 3 2 formed using genomic DNA as a template: one with a 7.0[39] overnightatroomtemperature and examined gene-specific primer and a T-DNA left border primer and photographed under a Zeiss Axioplan II compound LBb1, the second with two gene-specific primers. The microscope equipped with an AxioCam color digital PCR products were sequenced to confirm the locations camera. Measurements were performed using SIS Pro of the T-DNA insertion sites for all of the mutants. The software (OSIS, Lakewood, CO) using the bars in the ori- gene specific primers used are listed in Table 2. ginal image. For pollen tube length measurements, 200 pollen tubes were chosen randomly for each genotype, Crossing and screening for double mutant and significance was assessed using Student’s t-test. Single mutant alleles (AtrabD2b-1 and AtrabD2c-1; For fluorescence microscopy, the germinated pollen Figure 1A) were crossed, the F1 generation of these was transferred onto a slide and two drops of aniline crosses was allowed to self fertilization and the blue solution (0.005% aniline blue solution in 0.1 M AtrabD2b/2c double mutant was identified from the F2 sodium phosphate, pH 7.0) were added for ten minutes. generation by PCR genotyping. To confirm the pollen tube growth defects, 20 open flowers per genotype were cut below the pistil and inserted Semi-quantitative reverse transcription PCR vertically into germination medium in a 9-cm Petri dish. Total RNA was extracted from leaves of 20 DAI (days Plates were sealed and incubated overnight at 22°C at after imbibition) plants using the TRIZOL reagent 100% humidity under continuous illumination. The paths (Invitrogen). RT-PCR was performed using Super- of pollen tubes inside the pistils were visualized by fixing whole pistils in 2% glutaraldehyde and 2% paraformalde- Script™ III One-Step RT-PCR System (Invitrogen,) as per the manufacturer’smanual. The b-tubulin gene, hyde in 0.1 M sodium cacodylate buffer, pH 7.2, under which is highly conserved and constitutively expressed low vacuum (18 psi Hg) for 2 h at room temperature. in all eukaryotes, was used as a standard. The primers Samples were washed three times in the same buffer and used are listed in Table 2. The RT-PCR products were stained with Aniline Blue and DAPI. The tissue was then sequenced to confirm the correct amplification cleared for 24 hours at room temperature with a drop of product. clearing solution (240 g of chloral hydrate and 30 g of gly- cerol in 90 ml water). Pollen was examined with a Zeiss Axioplan 2 light microscope (LM) and images were cap- tured with a Zeiss AxioCamHRc digital camera (Carl Zeiss, Inc., Thornwood, NY) using AxioVision 4.3 soft- Table 2 Primers used in this study ware. The microscope was equipped with a DAPI filter set LBb1 GCGTGGACCGCTTGCTGCAACT comprising an excitation filter (BP 365/12 nm), a beam AtrabD2b-LP1 CCCTTCGTTGGGCTAGTAAAG splitter (395 nm), and an emission filter (LP 397 nm). The AtrabD2b-RP1 TTCAACAACGTCAAACAATGG objectives used for imaging were a Neofluar 40× oil, an AtrabD2c-LP1 GCGCATTACTGAGAGAGAAGAG Apochromat 63× oil, and a Neofluar 100× oil. AtrabD2c-RP1 TCCCATTCTTGGAAACAAGTG AtRabD2b-F ATGAATCCTGAATATGACTAT Cloning AtRabD2b-R TCAAGAAGAACAACAGCCT Promoter::GFP/GUS fusion constructs were made for AtRabD2c-F ATGAATCCTGAATATGACTAT each gene by cloning the amplified promoter region AtRabD2c-R TTAAGAGGAGCAGCAGCCT (intergenic region; 964 bp for AtRabD2b and 558 bp for AtRabD2c) into the binary vector pBGWFS7 (GATE- AtRabD2b-g-F caccATCGCTTATCCGCTCCGTGTATTTC WAY; Invitrogen). AtRabD2b-g-R TAAAGACCCCTGGTCCTTCAGC The genomic fragments containing AtRabD2b or AtRabD2c-g-F caccCTATCTCACTAAGCTGAAGATAC AtRabD2c with their respective promoters for comple- AtRabD2c-g-R GGCAATCTCTCCGGTTTGGTCC mentation of the mutant phenotype were amplified b-Tubulin-F CGTGGATCACAGCAATACAGAGCC using AtRabD2b-g-F and R or AtRabD2c-g-F and R pri- b-Tubulin-R CCTCCTGCACTTCCACTTCGTCTTC mers (Table 2). Products were cloned into the pENTR/ Peng et al. BMC Plant Biology 2011, 11:25 Page 15 of 16 http://www.biomedcentral.com/1471-2229/11/25 D vector (Invitrogen), and then were transferred into the (18 psi Hg) for 5 h at room temperature. Samples were pMDC123 binary vector for plant transformation. washed three times in the same buffer, postfixed in 1% osmium tetroxide in the same buffer for 2 h and washed Plant transformation and selection twotimes in thesamebuffer, followed by deionized Arabidopsis plants were transformed using Agrobacter- water. Samples were dehydrated through a graded etha- ium tumefaciens by the floral dip method [40] and nol series (50, 70, 85, 95, and 100%; 30 min per step), selected for Basta resistance conferred by the T-DNA. followed by two changes of ultrapure 100% ethanol, all 30 min per step. Fresh pollen was also examined with- Transcriptomic analysis out fixing. Fixed samples were critical point-dried in a MetaOmGraph (MOG; http://www.metnetdb.org) [25] DCP-1 Denton critical-point-drying apparatus (http:// was used to analyze expression patterns of AtRabD1, www.dentonvacuum.com) using liquid carbon dioxide, AtRabD2a, AtRabD2b and AtRabD2c and derive the and mounted on aluminum stubs with double-sided correlation between them. sticky pads and silver cement. Samples were then sputter-coated with 15 nm gold GUS assay (20%) and palladium (80%) in a Denton Vacuum LLC Transgenic T2 seedlings were germinated in soil and har- Desk II Cold Sputter Unit (http://www.dentonvacuum. vested at various stages of development. Plants or organs com), and viewed with a JEOL 5800LV SEM (http://www. were stained at room temperature overnight as described jeol.com) at 10 kV. Alternatively, released fresh pollen [41], then destained in 70% (v/v) ethanol. For each con- grains were directly mounted on stubs and sputter-coated struct, at least 7 independently transformed lines, 7 plants with gold particles before SEM analysis. All digitally col- for each stage, were harvested for GUS screening. lected images including the LM and SEM images were processed in Adobe PhotoShop 7.0 and made into plates Transient expression in protoplasts using Adobe Illustrator 10. Over 20 samples from each Transient gene expression in Arabidopsis mesophyll plant line were used for SEM or LM analysis. protoplasts was carried out as described previously [42]. In brief, Arabidopsis protoplasts were isolated from the Additional material leaves of 3-4 week old plants. Leaf strips were digested in a buffer containing cellulose R-10 and macerozyme Additional file 1: Table S1. Expression pattern of AtRab genes. Pearson correlation between expression patterns of AtRab genes R-10. After adding 30 μg of plasmid DNA, an equal determined using MetaOmGraph (Excel file). volume of protoplasts was mixed with PEG buffer (40% Additional file 2: Figure S1. Seed number per silique in wild-type (w/v) PEG4000, 25% (v/v) 0.8M mannitol, 10% 1M and mutant plants. Seed number was counted for 15 siliques of 5 CaCl ) then incubated at room temperature for 25 min. individual plants for the indicated genotypes. Error bars indicate standard deviation (pdf file). After gentle washing, the protoplasts were kept in the Additional file 3: Figure S2. Controls for confocal microscopy. dark at room temperature overnight and then viewed by Arabidopsis leaf protoplasts were transformed with either GFP-AtRabD2b confocal laser scanning microscopy as described below. or ST-YFP and imaged in the green, yellow and red channels as shown in Figure 10. No cross-talk between channels could be seen using these settings. Upper panel, GFP-AtRabD2b; lower panel, ST-YFP. Scale bar = 10 Confocal laser scanning microscopy μm (pdf file). Colocalization of GFP-RabD2b and GFP-RabD2c with ST-YFP was performed using a Leica TCS SP10 confo- cal microscope, which allows flexible selection of emis- Acknowledgements sion bandwidths to minimize bleed-through. We are grateful to Ian Moore, University of Oxford, United Kingdom for Transformed cells were excited with a 488 nm laser kindly providing the N-ST-YFP construct and for helpful suggestions about (power 20%) and 514 nm laser (50% power), and GFP the Rab genes. We also thank the Arabidopsis Biological Resource Center and the Salk Institute Genomic Analysis Laboratory for providing T-DNA and YFP signals were collected using 495-510 nm and insertion mutants. This research was supported in part by grant MCB- 560-640 nm bandwidths, respectively. Non-transformed 0951170 from the National Science Foundation to ESW and grant no. cells and cells expressing asingleGFP or YFPfusion NNX09AK78G from the National Aeronautics and Space Administration to DCB. were used as controls to confirm the absence of cross talk between GFP, YFP and autofluorescence signals. Authors’ contributions JP carried out the experimental analyses described and drafted the manuscript. HI helped with the microscopy and figures. ESW conceived of Scanning electron microscopy the study, participated in its design and analysis of the data and helped to Pollen that had been germinated in vitro was placed in draft the manuscript. DCB participated in the design of the study, analysis of 2% glutaraldehyde and 2% paraformaldehyde in 0.1 M the data and helped to draft the manuscript. All authors read and approved the final manuscript. sodium cacodylate buffer, pH 7.2, under low vacuum Peng et al. BMC Plant Biology 2011, 11:25 Page 16 of 16 http://www.biomedcentral.com/1471-2229/11/25 Received: 3 September 2010 Accepted: 26 January 2011 25. Mentzen WI, Peng J, Ransom N, Nikolau BJ, Wurtele ES: Articulation of Published: 26 January 2011 three core metabolic processes in Arabidopsis: fatty acid biosynthesis, leucine catabolism and starch metabolism. BMC Plant Biol 2008, 8:76. 26. Li L, Ilarslan H, James MG, Myers AM, Wurtele ES: Genome wide co- References expression among the starch debranching enzyme genes AtISA1, 1. Boguski MS, McCormick F: Proteins regulating Ras and its relatives. Nature AtISA2, and AtISA3 in Arabidopsis thaliana. J Exp Bot 2007, 1993, 366(6456):643-654. 58(12):3323-3342. 2. Takai Y, Sasaki T, Matozaki T: Small GTP-binding proteins. Physiol Rev 2001, 27. Zheng H, Camacho L, Wee E, Batoko H, Legen J, Leaver C, Malhó R, 81(1):153-208. Hussey P, Moore I: A Rab-E GTPase mutant acts downstream of the Rab- 3. Geldner N, Friml J, Stierhof YD, Jurgens G, Palme K: Auxin transport D subclass in biosynthetic membrane traffic to the plasma membrane in inhibitors block PIN1 cycling and vesicle trafficking. Nature 2001, tobacco leaf epidermis. Plant Cell 2005, 17(7):2020-2036. 413(6854):425-428. 28. Plutner H, Cox AD, Pind S, Khosravi-Far R, Bourne JR, Schwaninger R, Der CJ, 4. Molendijk AJ, Ruperti B, Palme K: Small GTPases in vesicle trafficking. Curr Balch WE: Rab1b regulates vesicular transport between the endoplasmic Opin Plant Biol 2004, 7(6):694-700. reticulum and successive Golgi compartments. J Cell Biol 1991, 5. Surpin M, Raikhel N: Traffic jams affect plant development and signal 115(1):31-43. transduction. Nat Rev Mol Cell Biol 2004, 5(2):100-109. 29. Nuoffer C, Davidson HW, Matteson J, Meinkoth J, Balch WE: A GDP-bound 6. Waizenegger I, Lukowitz W, Assaad F, Schwarz H, Jurgens G, Mayer U: The of rab1 inhibits protein export from the endoplasmic reticulum and Arabidopsis KNOLLE and KEULE genes interact to promote vesicle fusion transport between Golgi compartments. J Cell Biol 1994, 125(2):225-237. during cytokinesis. Curr Biol 2000, 10(21):1371-1374. 30. Tisdale EJ, Balch WE: Rab2 is essential for the maturation of pre-Golgi 7. Schimmoller F, Simon I, Pfeffer SR: Rab GTPases, directors of vesicle intermediates. J Biol Chem 1996, 271(46):29372-29379. docking. J Biol Chem 1998, 273(35):22161-22164. 31. Tisdale EJ, Jackson MR: Rab2 protein enhances coatomer recruitment to 8. Sanderfoot AA, Raikhel NV: The specificity of vesicle trafficking: coat pre-Golgi intermediates. J Biol Chem 1998, 273(27):17269-17277. proteins and SNAREs. Plant Cell 1999, 11(4):629-642. 32. Tisdale EJ: A Rab2 mutant with impaired GTPase activity stimulates 9. Armstrong J: How do Rab proteins function in membrane traffic? Int J vesicle formation from pre-Golgi intermediates. Mol Biol Cell 1999, Biochem Cell Biol 2000, 32(3):303-307. 10(6):1837-1849. 10. Tamm LK, Crane J, Kiessling V: Membrane fusion: a structural perspective 33. Allan BB, Moyer BD, Balch WE: Rab1 recruitment of p115 into a cis-SNARE on the interplay of lipids and proteins. Curr Opin Struct Biol 2003, complex: programming budding COPII vesicles for fusion. Science 2000, 13(4):453-466. 289(5478):444-448. 11. Pfeffer S, Aivazian D: Targeting Rab GTPases to distinct membrane 34. Moyer BD, Allan BB, Balch WE: Rab1 interaction with a GM130 effector compartments. Nat Rev Mol Cell Biol 2004, 5(11):886-896. complex regulates COPII vesicle cis–Golgi tethering. Traffic 2001, 12. Seabra MC, Wasmeier C: Controlling the location and activation of Rab 2(4):268-276. GTPases. Curr Opin Cell Biol 2004, 16(4):451-457. 35. Hepler PK, Vidali L, Cheung AY: Polarized cell growth in higher plants. 13. Vernoud V, Horton AC, Yang Z, Nielsen E: Analysis of the small GTPase Annu Rev Cell Dev Biol 2001, 17:159-187. gene superfamily of Arabidopsis. Plant Physiol 2003, 131(3):1191-1208. 36. Schiott M, Romanowsky SM, Baekgaard L, Jakobsen MK, Palmgren MG, 14. Rutherford S, Moore I: The Arabidopsis Rab GTPase family: another Harper JF: A plant plasma membrane Ca2+ pump is required for normal enigma variation. Curr Opin Plant Biol 2002, 5(6):518-528. pollen tube growth and fertilization. Proc Natl Acad Sci USA 2004, 15. Batoko H, Zheng HQ, Hawes C, Moore I: A rab1 GTPase is required for 101(25):9502-9507. transport between the endoplasmic reticulum and golgi apparatus and 37. Cheung AY, Wang H, Wu HM: A floral transmitting tissue-specific for normal golgi movement in plants. Plant Cell 2000, 12(11):2201-2218. glycoprotein attracts pollen tubes and stimulates their growth. Cell 1995, 16. Cheung AY, Chen CY, Glaven RH, de Graaf BH, Vidali L, Hepler PK, Wu HM: 82(3):383-393. Rab2 GTPase regulates vesicle trafficking between the endoplasmic 38. Alonso JM, Stepanova AN, Leisse TJ, Kim CJ, Chen H, Shinn P, reticulum and the Golgi bodies and is important to pollen tube growth. Stevenson DK, Zimmerman J, Barajas P, Cheuk R, Gadrinab C, Heller C, Plant Cell 2002, 14(4):945-962. Jeske A, Koesema E, Meyers CC, Parker H, Prednis L, Ansari Y, Choy N, 17. Grebe M, Xu J, Mobius W, Ueda T, Nakano A, Geuze HJ, Rook MB, Deen H, Geralt M, Hazari N, Hom E, Karnes M, Mulholland C, Ndubaku R, Scheres B: Arabidopsis sterol endocytosis involves actin-mediated Schmidt I, Guzman P, Aguilar-Henonin L, Schmid M, Weigel D, Carter DE, trafficking via ARA6-positive early endosomes. Curr Biol 2003, Marchand T, Risseeuw E, Brogden D, Zeko A, Crosby WL, Berry CC, Ecker JR: 13(16):1378-1387. Genome-wide insertional mutagenesis of Arabidopsis thaliana. Science 18. Preuss ML, Serna J, Falbel TG, Bednarek SY, Nielsen E: The Arabidopsis Rab 2003, 301(5633):653-657. GTPase RabA4b localizes to the tips of growing root hair cells. Plant Cell 39. Li H, Lin Y, Heath RM, Zhu MX, Yang Z: Control of pollen tube tip growth 2004, 16(6):1589-1603. by a Rop GTPase-dependent pathway that leads to tip-localized calcium 19. Ueda T, Uemura T, Sato MH, Nakano A: Functional differentiation of influx. Plant Cell 1999, 11(9):1731-1742. endosomes in Arabidopsis cells. Plant J 2004, 40(5):783-789. 40. Clough SJ, Bent AF: Floral dip: a simplified method for Agrobacterium- 20. de Graaf BH, Cheung AY, Andreyeva T, Levasseur K, Kieliszewski M, Wu HM: mediated transformation of Arabidopsis thaliana. Plant J 1998, Rab11 GTPase-regulated membrane trafficking is crucial for tip-focused 16(6):735-743. pollen tube growth in tobacco. Plant Cell 2005, 17(9):2564-2579. 41. Jefferson RA, Kavanagh TA, Bevan MW: GUS fusions: beta-glucuronidase as 21. Tisdale EJ, Bourne JR, Khosravi-Far R, Der CJ, Balch WE: GTP-binding a sensitive and versatile gene fusion marker in higher plants. EMBO J mutants of rab1 and rab2 are potent inhibitors of vesicular transport 1987, 6(13):3901-3907. from the endoplasmic reticulum to the Golgi complex. J Cell Biol 1992, 42. Sheen J: Signal transduction in maize and Arabidopsis mesophyll 119(4):749-761. protoplasts. Plant Physiol 2001, 127(4):1466-1475. 22. Pinheiro H, Samalova M, Geldner N, Chory J, Martinez A, Moore I: Genetic evidence that the higher plant Rab-D1 and Rab-D2 GTPases exhibit doi:10.1186/1471-2229-11-25 distinct but overlapping interactions in the early secretory pathway. Cite this article as: Peng et al.: AtRabD2b and AtRabD2c have J Cell Sci 2009, 122(Pt 20):3749-3758. overlapping functions in pollen development and pollen tube growth. 23. Geldner N, Denervaud-Tendon V, Hyman DL, Mayer U, Stierhof YD, Chory J: BMC Plant Biology 2011 11:25. Rapid, combinatorial analysis of membrane compartments in intact plants with a multicolor marker set. Plant J 2009, 59(1):169-178. 24. Saint-Jore C, Evins J, Batoko H, Brandizzi F, Moore I, Hawes C: Redistribution of membrane proteins between the Golgi apparatus and endoplasmic reticulum in plants is reversible and not dependent on cytoskeletal networks. Plant J 2002, 29(5):661-678. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png BMC Plant Biology Springer Journals

AtRabD2b and AtRabD2c have overlapping functions in pollen development and pollen tube growth

Download PDF
16 pages

Loading...

Page 2

Loading...

Page 3

Loading...

Page 4

Loading...

Page 5

Loading...

Page 6

Loading...

Page 7

Loading...

Page 8

Loading...

Page 9

Loading...

Page 10

Loading...

Page 11

Loading...

Page 12

Loading...

Page 13

Loading...

Page 14

Loading...

Page 15

Loading...

Page 16

 
/lp/springer-journals/atrabd2b-and-atrabd2c-have-overlapping-functions-in-pollen-development-NokJm00DFT

References (88)

Publisher
Springer Journals
Copyright
Copyright © 2011 by Peng et al; licensee BioMed Central Ltd.
Subject
Life Sciences; Plant Sciences; Agriculture; Tree Biology
eISSN
1471-2229
DOI
10.1186/1471-2229-11-25
pmid
21269510
Publisher site
See Article on Publisher Site

Abstract

Background: Rab GTPases are important regulators of endomembrane trafficking, regulating exocytosis, endocytosis and membrane recycling. Many Rab-like proteins exist in plants, but only a subset have been functionally characterized. Results: Here we report that AtRabD2b and AtRabD2c play important roles in pollen development, germination and tube elongation. AtrabD2b and AtrabD2c single mutants have no obvious morphological changes compared with wild-type plants across a variety of growth conditions. An AtrabD2b/2c double mutant is also indistinguishable from wild-type plants during vegetative growth; however its siliques are shorter than those in wild-type plants. Compared with wild-type plants, AtrabD2b/2c mutants produce deformed pollen with swollen and branched pollen tube tips. The shorter siliques in the AtrabD2b/2c double mutant were found to be primarily due to the pollen defects. AtRabD2b and AtRabD2c have different but overlapping expression patterns, and they are both highly expressed in pollen. Both AtRabD2b and AtRabD2c protein localize to Golgi bodies. Conclusions: These findings support a partially redundant role for AtRabD2b and AtRabD2c in vesicle trafficking during pollen tube growth that cannot be fulfilled by the remaining AtRabD family members. Background factor attachment protein receptor (SNARE) proteins to Ras-like small GTP-binding proteins (GTPases) regulate promote specificity of vesicle transport to target com- diverse processes in eukaryotic cells including signal partments and facilitate vesicle and target membrane transduction, cell proliferation, cytoskeletal organization fusion [7-13]. They are therefore essential for the trans- and intracellular membrane trafficking. GTPases are port of proteins and membrane through the endomem- activated by GTP binding and inactivated by subsequent brane system to their destination. hydrolysis of bound GTP to GDP, thus acting as mole- The Arabidopsis thaliana genome encodes 93 putative cular switches in these processes [1,2]. The Rab GTPase Ras superfamily proteins. Fifty-seven of these are Rab family is the largest and most complex within the Ras GTPases, more than in yeast but similar to the number protein superfamily. Rab GTPases are important regula- in humans [13,14]. According to their sequence similar- tors of endomembrane trafficking, regulating exocytosis, ity and phylogenetic clustering with yeast and mamma- endocytosis and membrane recycling processes in eukar- lian orthologs, these Rab proteins were assigned to eight yotic cells [3-6]. Rab GTPase functions have been exten- subfamilies, AtRabA to AtRabH, which can be further sively studied in yeast and mammalian systems. Both in divided into 18 subclasses [13]. Relatively few of the vivo and in vitro experiments have demonstrated that plant Rab orthologs have been investigated functionally. different Rab proteins function in distinct intracellular Most of these studies have used constitutively active membrane trafficking steps and they are hypothesized to (CA) and/or dominant negative (DN) mutations, gener- work together with soluble N-ethylmaleimide-sensitive ated by direct mutation of the conserved domain to restrict mutant GTPase proteins to the active GTP- * Correspondence: bassham@iastate.edu bound form (constitutively active) or inactive GDP- Department of Genetics, Development and Cell Biology, Iowa State bound form (dominant negative). Expression of CA or University, Ames, IA 50010, USA © 2011 Peng et al; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Peng et al. BMC Plant Biology 2011, 11:25 Page 2 of 16 http://www.biomedcentral.com/1471-2229/11/25 DN Rab GTPases can perturb the activity of the endo- RNA accumulation level across a variety of experimen- genous Rab, revealing their functional significance. For a tal conditions. Phenotypic analysis of knockout number of plant Rab GTPases, expression of their CA mutants indicates that they are at least partially func- and DN mutants in transformed plants, together with tionally redundant, and are important in pollen devel- protein localization information, has shown that these opment and pollen tube growth. The proteins both Rabs perform functions similar to those of their yeast localize to the trans-Golgi, consistent with their pro- posed role in trafficking from the ER to the Golgi and mammalian orthologs [15-19]. apparatus. Several reports indicate that Rab proteins are impor- tant for elongation of tip-growing cells in plants. For example, AtRabA4b is reported to localize to the tips of Results root hair cells and was proposed to regulate membrane The expression patterns of AtRabD2b and AtRabD2c are trafficking through a compartment involved in the closely correlated polarized secretion of cell wall components [18]. NtRab2 The four RabD family members in Arabidopsis share GTPase is important for trafficking between the endo- about 88% identity at the amino acid level. The accu- plasmic reticulum and Golgi bodies in tobacco pollen mulation pattern of the associated transcripts is quite tubes and may be specialized to optimally support the distinct across a wide variety of experimental condi- high secretory demands in these tip growing cells [16]. tions and developmental stages (MetaOmGraph, http:// NtRabA (Rab11) in tobacco is predominantly localized www.metnetdb.org/MetNet_MetaOmGraph.htm; [26]) to an inverted cone-shaped region at the pollen tube tip, (Table 1; Additional file 1, Table S1). AtRabD2b and and both constitutively active and dominant negative AtRabD2c expression patterns are correlated (at a mutants resulted in reduced tube growth rate, meander- Pearson correlation value of 0.72), whereas AtRabD1 ing pollen tubes, and reduced male fertility [20]. and AtRabD2a show very low correlation with the There are four genes in the Arabidopsis RabD subfam- others (Pearson correlation value of < 0.20). Based on ily, AtRabD1 (At3g11730), AtRabD2a (At1g02130, their high sequence similarity (99% amino acid iden- AtRab1b), AtRabD2b (At5g47200, AtRab1a) and tity) and the correlation between their mRNA accumu- AtRabD2c (At4g17530, AtRab1c) [13]. In mammals, the lation patterns, we hypothesized that AtRabD2b and orthologs of AtRabD, Rab1 isoforms, physically associate AtRabD2c might have some functional overlap that is with the ER, ER-Golgi intermediate compartment and not shared by AtRabD1 and AtRabD2a. Golgi and regulate membrane trafficking between the ER and Golgi complex [21]. Fluorescent protein fusions Identification of Null Mutations in the Genes AtRabD2b with AtRabD1, AtRabD2a and AtRabD2b localize to the and AtRabD2c Golgi and trans-Golgi network [22,23], and transient It was reported previously that an AtrabD2b AtrabD2c expression in plant cells of dominant negative mutants double mutant has no phenotype [22]. Based on our of rabD2a or rabD1 resulted in the inhibition of ER-to- correlation analysis above, we hypothesized that this Golgi trafficking [15,22,24], suggesting a related function mutant may have some more subtle defects that cannot for the plant Rab1 homologs. Pinheiro et al. [22] isolated be compensated for by the remaining family members. T-DNA insertion mutants in each of the AtRabD family To investigate this further, we identified T-DNA inser- genes and reported that each of the single and double tion mutants (Figure 1A) in AtrabD2b (3 alleles) and mutants lacked a detectable phenotype. By contrast, a AtrabD2c (1 allele). Homozygous lines for the T-DNA rabD2a rabD2b rabD2c triple mutant was lethal and a insertions were identified by PCR, using primers rabD1 rabD2b rabD2c triple mutant had stunted growth selected by iSct primers (http://signal.salk.edu/tdnapri- and low fertility, indicating that these gene family mem- mers.2.html), and the insertion sites were determined by bers perform important and overlapping functions. sequencing the PCR products (Figure 1B). Analysis of We previously hypothesized that closely related mRNA levels by RT-PCR indicated that AtrabD2b-1, genes with a high Pearson correlation in their RNA AtrabD2b-2 and AtrabD2c-1 are null mutants. However, accumulation level are functionally redundant, and showed that expression patterns of both the AtRabD2b Table 1 Pearson correlation between expression patterns and AtRabD2c genes are negatively correlated with the of AtRabD family members process of starch synthesis [25], whereas the expression AtRabD1 AtRabD2a AtRabD2b AtRabD2c patterns of the remaining RabD genes are not. We AtRabD1 100% therefore predicted that these two Rab proteins may AtRabD2a 17.08% 100% have redundant functions that are not shared by the AtRabD2b -3.4% 22.19% 100% other two AtRabD family members. Here we show that AtRabD2c 15.92% 25.23% 77.81% 100% AtRabD2b and AtRabD2c are highly correlated in their Peng et al. BMC Plant Biology 2011, 11:25 Page 3 of 16 http://www.biomedcentral.com/1471-2229/11/25 single mutants were crossed, F1 plants were allowed to AtrabD2b-1 self fertilize and the AtrabD2b-2/AtrabD2c-1 double ATG TGA mutant was identified from the F2 population by PCR using the primers for both AtrabD2b-2 and AtrabD2c-1. 100bp Hereafter, the AtrabD2b-2/AtrabD2c-1 double mutant A tRabD2b AtrabD2b-3 AtrabD2b-2 2055bp will be referred to as AtrabD2b/2c, and AtrabD2b-2 and ATG TAA AtrabD2c-1 single mutants will be referred to as AtrabD2b and AtrabD2c respectively. A tRabD2c 100bp 2060bp Siliques Are Shorter in the AtrabD2b/2c Double Mutant AtrabD2c-1 than in Either Single Mutant or in Wild-Type Lines Col-0 AtrabD2c AtrabD2b AtrabD2b/2c To evaluate phenotypes associated with the AtrabD2b M M 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 and AtrabD2c mutants, homozygous AtrabD2b (three alleles, AtrabD2b-1, AtrabD2b-2 and AtrabD2b-3), AtrabD2c and AtrabD2b/2c mutants, along with wild- type siblings, were grown on agar plates with or without various hormone, nutrient and light treatments. We tested over 50 of the conditions described in the Gantlet C website (http://www.gantlet.org); however, no significant phenotypic differences were observed in the seedlings for any of the mutant alleles (data not shown). In addition, we tested the seedling phenotype on media with or with- AtRabD2c out sucrose or vitamin B5 and, consistent with previous reports [22], no obvious phenotypes were observed. By contrast, AtrabD2b/2c double mutant lines AtRabD2b showed a phenotype associated with reproduction. In these lines, siliques were shorter when grown either under continuous light or long day (16h light/8h dark) Tubulin conditions. Neither the AtrabD2b nor the AtrabD2c single mutant alleles displayed a short silique pheno- Figure 1 Characterization of AtrabD2b and AtrabD2c mutations. type. The length of AtrabD2b/2c siliques was 70% of A, Gene map. The scaled linear map depicts the 8 exons as boxes that of wild-type, AtrabD2b or AtrabD2c single mutant and the 7 introns as lines between the boxes for both the lines (Figure 2; P < 0.01 by Student’s t-test). To evalu- AtRabD2b and AtRabD2c genes. The positions of the translational ate whether this reduced silique size is associated with start and stop codons in exon 1 and exon 8, respectively, are noted. The locations of the T-DNA insertions (not drawn to scale) in the a seed defect, siliques from AtrabD2b/2c, wild-type, genes are indicated. B, Genotypes of T-DNA insertion mutants. AtrabD2b and AtrabD2c mutant lines were opened at Genomic DNA was isolated from the indicated single and double 10 DAF (days after flowering). Consistently, no defects mutants and amplified by PCR. Primer pairs used were as following: in the seeds of either AtrabD2b or AtrabD2c single lane 1, AtrabD2c-LP1 and AtrabD2c-RP1; lane 2, AtrabD2c-RP1 and LBb1; lane 3, AtrabD2b-LP1 and AtrabD2b-RP1; lane 4, AtrabD2b-RP1 mutants were observed. However, approximately half and LBb1. C, Analysis of transcripts from AtrabD2b-1, AtrabD2c-1 and of the ovules in the AtrabD2b/2c double mutant were AtrabD2b/2c mutants. Total RNA from leaves of wild-type plants, not fertilized (Figure 3). Consistent with this observa- AtrabD2c-1, AtrabD2b-1 and AtrabD2b/2c was amplified by RT-PCR. tion, the AtrabD2b/2c mutant plants produced a smal- Primer pairs for AtRabD2c were AtRabD2c-F and AtRabD2c-R, primer ler quantity of seeds than wild-type plants or single pairs for AtRabD2b were AtRabD2b-F and AtRabD2b-R. Tubulin was used as control. mutants (Figure 3; Additional file 2, Figure S1). These results are consistent with a functional overlap between AtRabD2b and AtRabD2c that cannot be ful- the AtrabD2b-3 mutation had no effect on AtRabD2b filled by AtRabD1 or AtRabD2a. RNA accumulation (Figure 1C and data not shown). Progeny from AtrabD2b-1 and AtrabD2c-1 heterozy- Complementation of AtrabD2b/2c Mutant Phenotype gotes showed a T-DNA segregation ratio of approxi- To demonstrate that the AtrabD2b/2c mutant phenotype mately 3:1 based on kanamycin resistance, consistent is due to the mutations in the AtRabD2b and AtRabD2c with a single insertion. AtrabD2b-2 wassuppliedasa genes, constructs containing either AtRabD2b or homozygous line. To generate AtrabD2b AtrabD2c dou- AtRabD2c, each expressed from their native promoter, ble mutants, AtrabD2b-2 and AtrabD2c-1 homozygous were introduced into the AtrabD2b/2c double mutant. AtrabD2c Col-0 AtrabD2b/2c AtrabD2b Peng et al. BMC Plant Biology 2011, 11:25 Page 4 of 16 http://www.biomedcentral.com/1471-2229/11/25 Figure 2 The AtrabD2b/2c double mutant shows a striking shorter silique phenotype. A, Vegetative growth of AtrabD2b, AtrabD2c and AtrabD2b/2c plants. B, Inflorescence of AtrabD2b/2c and wild-type plants. Scale bars = 850 μm. C, Siliques from the AtrabD2b/2c mutant and wild-type plants; arrows indicate the sequence of siliques from the oldest to the youngest. Scale bars = 850 μm. D, Siliques (from 6 to 14 ) of the first inflorescence for wild type, single and double mutants were measured for each plant, with 10 plants measured for each genotype. Error bars indicate standard deviation. Peng et al. BMC Plant Biology 2011, 11:25 Page 5 of 16 http://www.biomedcentral.com/1471-2229/11/25 Figure 3 There are many non-fertilized ovaries in the AtrabD2b/2c double mutant. Individual siliques of wild type, single and double mutant plants were dissected and examined under the microscope. Arrows heads indicate unfertilized ovaries. Inset, seeds produced by a single plant. Scale bars = 200 μm. Both constructs were able to rescue the silique length pollen visualized under the SEM. Surprisingly, even the phenotype of the mutant (Figure 4A, C) and restored the AtrabD2b and AtrabD2c single mutant lines produce seed fertilization defect (Figure 4B) and seed number aberrant pollen at a level of about 10%. This is unex- (Additional file 2, Figure S1), confirming that the loss of pected, as the AtrabD2b and AtrabD2c single mutants AtRabD2b and AtRabD2c is responsible for these have normal-appearing siliques and seed quantities simi- phenotypes. lar to the wild-type plants. A likely explanation is that there are sufficient normal pollen grains in the single AtrabD2b/2c, AtrabD2b and AtrabD2c Pollen Have Defects mutants to efficiently fertilize the ovaries in the in Morphology and Pollen Tube Elongation AtrabD2b and AtrabD2c single mutants. Two possibilities could explain the unfertilized embryos We originally identified AtRabD2b and AtRabD2c seen in the AtrabD2b/2c double mutants. One possibi- because the transcript accumulation patterns of these lity is that the pollen bears a defect that leads to pollen two genes correlate with those of many genes associated sterility and inability to fertilize the embryos. Alterna- with starch metabolism. Indeed, the AtrabD2b/2c double tively, ovules may bear an abnormality such that their mutant pollen stained less intensely with IKI than wild- fertilization is reduced. To distinguish between these type pollen (Figure 5C), suggesting a decreased starch two possibilities, we observed the pollen by scanning content in the AtrabD2b/2c mutant pollen. This is con- sistent with the expression correlation, although the rea- electron microscopy (SEM). All of the pollen from wild- son for this phenotype is unclear. type plants looked normal, whereas more than 20% of the AtrabD2b/2c pollen exhibited an irregular, collapsed A single flower of Arabidopsis produces thousands of morphology (Figure 5A). We also observed that some pollen grains, but usually there are less than 100 abnormal pollen grains from the AtrabD2b/2c double embryos in one silique. If only 20% of the pollen grains mutant were devoid of nuclei, as indicated by DAPI are abnormal, we would not expect the strikingly staining, whereas all pollen from wild-type (Figure 5B) reduced fertility seen in the AtrabD2b/2c double andsinglemutantplants(data notshown)havenuclei. mutant. We therefore looked for additional explanations This defective pollen may be the severely collapsed for the reduced fertility. To evaluate germination and Peng et al. BMC Plant Biology 2011, 11:25 Page 6 of 16 http://www.biomedcentral.com/1471-2229/11/25 and 6B). Moreover, the pollen tubes of the AtrabD2b/2c double mutant were much shorter than those of wild- type plants or either single mutant (P < 0.01), and the single mutants had shorter pollen tubes than wild-type plants (Figure 6E; P < 0.01 for both mutants). Even though the AtrabD2b and AtrabD2c single mutants had collapsed pollen, shorter pollen tubes and swollen tips, their siliques were normal compared with wild-type plants. We hypothesize that the single mutants may still have sufficient normal pollen to enable all embryos to be fertilized. The in vitro pollen germination phenotypes were confirmed by analyzing pollen tube growth after in vivo pollination (Figure 6C). Open flowers from wild- type or AtrabD2b/2c mutant plants were incubated overnight on agar medium. The AtrabD2b/2c mutant flowers had reduced pollen germination and decreased pollen tube length compared with wild-type plants, sug- gesting that pollen germination and pollen tube growth may also be defective in vivo. Pollen and Pollen Tube Defects Cause the Shorter Siliques in the AtrabD2b/2c Mutant To investigate whether the unfertilized seeds are due to the observed pollen abnormality, or whether the ovary also has defects that might contribute to the reduced rate of fertilization, we crossed wild-type and AtrabD2b/ 2c mutant plants. If the shorter silique phenotype is borne only by the abnormal pollen, wild-type plant pol- len should rescue the AtrabD2b/2c mutant silique phe- Figure 4 Complementation of the double mutant phenotype. notype to a normal length (AtrabD2b/2c mutant female A, Siliques are shown from wild-type plants, AtrabD2b and AtrabD2c flower crossed with wild-type plant pollen). In contrast, single mutants, the AtrabD2b/2c double mutant and the AtrabD2b/ the AtrabD2b/2c mutant plant pollen crossed with a 2c double mutant complemented with either AtRabD2b or wild-type female would mimic the mutant phenotype of AtRabD2c. Scale bars = 0.5 cm. B, Individual siliques of rescued lines were dissected and examined under the microscope. Scale bars = decreased fertilization (wild-type female flower crossed 600 μm. C, Siliques (from 6 to 14 ) of the first inflorescence for the with AtrabD2b/2c mutant pollen). Alternatively, if the indicated genotypes were measured for each plant, with 10 plants ovary also has some abnormality, wild-type pollen measured for each genotype. Error bars indicate standard deviation. would not completely rescue the mutant phenotype, and AtrabD2b/2c mutant pollen would not mimic the tube growth of the pollen grains, pollen was germinated mutant phenotype. The results of these crosses indicated in vitro. After overnight incubation, almost all of the that pollen from wild-type plants can rescue the pollen from wild-type plants germinated and showed a AtrabD2b/2c short silique phenotype, and the pollen typical tip growth. However, about 10% of the pollen from AtrabD2b/2c can bestow the shorter silique phe- from the AtrabD2b/2c mutant did not germinate at all notype on wild-type plants (Figures 7A and 7C). Specifi- and 50% of the pollen germinated but did not grow api- cally, about half of the seeds were not fertilized in the cally as did pollen of wild-type plants (Figure 6A and siliques that developed from wild-type pistils fertilized 6B). Instead, these pollen tubes were shorter and had by AtrabD2b/2c pollen (Figure 7B). In contrast, the sili- swollen tips, some burst (≈5%), and others branched ques from AtrabD2b/2c usually had about 50% unferti- (≈2%; Figure 6A,B and 6D). The germination rate of the lized ovules, but when these pistils were fertilized by pollen from the single mutants was similar to the wild- wild-type pollen, all seeds looked normal, and the sili- type pollen. However, approximately 20% of the germi- ques were longer than those siliques in the same inflor- nating pollen also had swollentips(Figure 6Aand6B), escence which were self-fertilized (Figures 7). These although the phenotype was not as severe as the results confirm that the unfertilized ovaries are mostly, AtrabD2b/2c double mutant; burst or branched tubes if not exclusively, caused by pollen defects in the were never observed in either single mutant (Figure 6A AtrabD2b/2c mutant. Peng et al. BMC Plant Biology 2011, 11:25 Page 7 of 16 http://www.biomedcentral.com/1471-2229/11/25 Figure 5 Pollen defects in AtrabD2b, AtrabD2c and AtrabD2b/D2c mutants. A, Fresh pollen was examined by SEM. B, DAPI staining of pollen. Fresh pollen grains were stained with DAPI and photographed under the fluorescence microscope. Arrow indicates a pollen grain from the AtrabD2b/2c mutant that lacks a nucleus. C, IKI staining of pollen, demonstrating reduced staining of the AtrabD2b/2c double mutant pollen compared with wild-type pollen. Scale bars = 10 μm (A); 50 μm (B,C). In silico and GUS Analysis of AtRabD2b and AtRabD2c including high expression in floral organs and particu- Expression larly in the stamen (Figure 8; [25,26]). If AtRabD2b and AtRabD2c are involved in pollen To directly examine the spatial expression pattern of development and pollen tube growth, they are expected the AtRabD2b and AtRabD2c genes, transgenic lines to be co-expressed in pollen and pollen tubes. Public containing promoter:GFP/GUS constructs for each gene microarray data indicates that both AtRabD2b and were analyzed for GUS activity at various stages of AtRabD2c are expressed throughout development, development from germination to senescence. As Peng et al. BMC Plant Biology 2011, 11:25 Page 8 of 16 http://www.biomedcentral.com/1471-2229/11/25 Figure 6 Pollen tube elongation defects in AtrabD2b, AtrabD2c and AtrabD2b/2c mutants. A, Pollen was germinated in vitro for 6 hours and examined by SEM. B, Germinated pollen was stained with aniline blue then observed under an epifluorescence microscope. C. Open flowers from an AtrabD2b/2c mutant plant, along with a wild-type plant, were incubated overnight on medium then examined by fluorescence microscopy. D. Close up view of pollen tubes in the AtrabD2b/2c mutant. E. Pollen was germinated in vitro and pollen tube length measured after an overnight incubation using SIS Pro software (OSIS, Lakewood, CO) (n > 200). Error bars indicate standard deviation. Scale bars = 10 μm (A); 50 μm (B, C); 20 μm (D). indicated by the in silico analyses, both AtRabD2b and only expressed in the trichomes, while AtRabD2b was AtRabD2c were expressed widely during development. not expressed in these cells (Figure 9A). In flowers, GUS staining further indicated that in cotyledons, AtRabD2b was expressed in sepals, stamen and stigma, rosette leaves and cauline leaves, AtRabD2b expression while AtRabD2c was expressed in sepal, stamen, stigma was localized predominantly in vascular tissues (Figure and style (Figure 9E, F). This dichotomy of expression 9B), whereas AtRabD2c was expressed ubiquitously in suggests that AtRabD2b and AtRabD2c may function cotyledons and in mature leaves throughout the entire independently of each other in certain cells. Both genes leaf. Interestingly, in emerging leaves, AtRabD2c was were expressed in pollen grains and germinating pollen Peng et al. BMC Plant Biology 2011, 11:25 Page 9 of 16 http://www.biomedcentral.com/1471-2229/11/25 Figure 7 Wild-type pollen can restore the shorter siliques of the AtrabD2b/2c mutant to normal length.Wild-typeand AtrabD2b/2c double mutant plants were crossed and silique length measured after 10 days. A. Inflorescences from a cross between a wild-type plant and AtrabD2b/2c mutant. The blue arrow indicates a silique in which a wild-type pistil was fertilized with AtrabD2b;AtrabD2c pollen. The red arrow indicates a silique in which the AtrabD2b/2c mutant pistil was fertilized with wild-type pollen. B. Siliques from the crosses at 10 DAP (days after pollination) were dissected and examined under a stereo microscope. White arrowheads indicate unfertilized embryos found upon pollination of wild-type plants with AtrabD2b/2c pollen. C, More than 20 siliques were measured for each plant. Error bars indicate standard deviation. Scale bars= 850 μm (A); 500 μm (B). Peng et al. BMC Plant Biology 2011, 11:25 Page 10 of 16 http://www.biomedcentral.com/1471-2229/11/25 Signal Intensity Figure 8 In silico expression analysis of AtRabD2b and AtRabD2c. The spatial and temporal expression profiles of AtRabD2b and AtRabD2c were analyzed using Genevestigator anatomy (A) and development (B) tools, respectively. Numbers along the X axis represent the developmental stage: 1, germinated seed; 2, seedlings; 3, young rosette; 4, developed rosette; 5, bolting; 6, young flower; 7, developed flower; 8, flowers and siliques; 9, mature siliques. Signal Intensity Peng et al. BMC Plant Biology 2011, 11:25 Page 11 of 16 http://www.biomedcentral.com/1471-2229/11/25 Figure 9 Temporal and spatial expression pattern of AtRabD2b and AtRabD2c. Transgenic plants were generated that express the GUS gene driven by a 954bp or 558bp fragment upstream of the AtRabD2b or AtRabD2c start codon, respectively. GUS activity (blue color) was analyzed in cotyledons, young leaves, (A, B), old leaves, roots (C, D), flowers, pistils and germinated pollen (E, F). (Figure 9E, F), which is consistent with their role in pol- Subcellular localization of AtRabD2b and AtRabD2c len development and pollen tube growth. They both In mammals, different AtRabD orthologs (Rab1 isoforms) also showed expression in roots, but AtRabD2b was ubi- are localized to the ER, ER-Golgi intermediate compart- quitously expressed throughout the root, while ment or Golgi compartments. In plants, AtRabD2a and AtRabD2c expression was excluded from root hairs and AtRabD2b are associated with the Golgi apparatus and root tips (Figure 9A-D). trans-Golgi network [15,22,23,27], and we predicted that Peng et al. BMC Plant Biology 2011, 11:25 Page 12 of 16 http://www.biomedcentral.com/1471-2229/11/25 AtRabD2c will share this localization. To determine the and Rab2 (AtRabB orthologs), have been reported to func- subcellular localization of AtRabD2c, GFP-AtRabD2b tion in membrane trafficking between the ER and Golgi in and GFP-AtRabD2c constructs were introduced into mammalian cells [21,28-34]. The plant Rab1 and Rab2 Arabidopsis leaf protoplasts and the GFP signal was homologs AtRabD1, AtRabD2a and NtRab2 have also observed by confocal microscopy. For both constructs, been reported to function in ER to Golgi vesicle transport GFP localized to punctuate structures, reminiscent of the [15,16]. Here, we demonstrate a distinct physiological role Golgi apparatus. To verify the identity of these structures, for the Rab1 homologs AtRabD2b and AtRabD2c in pollen GFP-AtRabD2b and GFP-AtRabD2c were co-transfected development and pollen tube growth. Using the bioinformatics tool MetaOmGraph (http:// into Arabidopsis leaf protoplasts with the trans-Golgi marker ST-YFP [15], and YFP and GFP signals were www.metnetdb.org) [25,26] to determine the pairwise observed. Confocal results indicated that both AtRabD2b Pearsons correlation value between the expression pat- and AtRabD2c primarily colocalized with ST-YFP (Figure terns of all of the 57 AtRab genes (Additional file 1, 10) and are therefore associated with the Golgi, consis- Table S1), we found that among the four AtRabDs, only tent with a role in Golgi trafficking. Occasionally, the expression of AtRabD2b and AtRabD2c are highly AtRabD2b or AtRabD2c-labeled structures were seen correlated. From this data, we hypothesized that that did not contain ST-YFP; these could be the post- AtRabD2b and AtRabD2c have partially redundant func- Golgi compartments described previously [22,23,27]. tions that are not shared by the remaining AtRabD Cells expressing a single GFP or YFP fusion demon- family members. To test our hypothesis, we used T- strated the absence of cross talk between GFP and YFP DNA insertion single and double mutants to confirm signals (Additional file 3, Figure S2). that AtRabD2b and AtRabD2c have functional overlap and show that they are both required for normal pollen Discussion development and tip growth of pollen tubes. We also Rab GTPases are critical players in the transport of materi- showed that they both co-localize with the trans-Golgi als through the endomembrane system, controlling exocy- marker ST-YFP, consistent with their proposed role in tosisofproteinsand cell wall materials, endocytosis of Golgi trafficking. receptors and transporters, and membrane recycling pro- The conclusion that AtRabD2b and AtRabD2c are cesses. Together with SNARE proteins, they promote spe- partially functionally redundant is based on several lines cificity of vesicle transport to target compartments, of evidence. First, although single mutant plants con- ensuring that vesicles fuse only with their appropriate tar- taining AtrabD2b or AtrabD2c mutant alleles are indis- get and thus maintaining the distinct identity of individual tinguishable morphologically from their wild-type counterparts, even when grown under a variety of organelles. Two Rab subfamilies, Rab1 (AtRabD orthologs) Figure 10 AtRabD2b and AtRabD2c are both Golgi resident proteins. Arabidopsis leaf protoplasts were co-transformed with GFP-AtRabD2b or GFP-AtRabD2c and the Golgi marker ST-YFP. GFP, YFP and chlorophyll autofluorescence were detected by confocal microcopy. At least 10 transformed protoplasts were observed for each construct. Upper panel, GFP-AtRabD2b; lower panel, GFP-AtRabD2c. Scale bar = 10 μm. Peng et al. BMC Plant Biology 2011, 11:25 Page 13 of 16 http://www.biomedcentral.com/1471-2229/11/25 growth conditions, the AtrabD2b/2c double mutant has Second, pollen tubes grow very rapidly compared with a short-silique phenotype. Second, both AtrabD2b and many other cell types. Pollen tubes elongate by tip AtrabD2c mutant plants produce a small percentage of growth, whereby the pollen cytoplasm is confined to the deformed and collapsed pollen grains, while AtrabD2b/ most proximal region of the tube, and growth is 2c lines produce a higher percentage of deformed pollen restricted to the tube apex [35]. In vitro,lilypollen grains, many of which are severely deformed, some lack- tubes grow at about 150 nm/sec [35] and Arabidopsis pollen tubes at 37 nm/sec [36]; in vivo, tobacco pollen ing nuclei. It is probable that such aberrant pollen can grow at 42 nm/sec [37]. This fast growth is contin- would give rise to defects in pollen germination, and gent on rapid vesicle trafficking to deliver large amounts indeed, though the germination rate is similar between AtrabD2b or AtrabD2c single mutants and wild-type of membrane and cell wall components to the apical plants, about 10% of the pollen grains from AtrabD2b/ region of the tubes. This extensive trafficking require- 2c double mutant plants are unable to geminate. ment may preclude the remaining Rabs from completely Furthermore, AtrabD2b and AtrabD2c mutant pollen compensating for loss of AtRabD2b and AtRabD2c. tubes do not grow apically as well as do wild-type pollen Third, computational analysis of public microarray tubes and tend to have swollen tips and a shorter data, together with studies of the expression pattern length; this phenotype is substantially more severe in directed by the AtRabD2b and AtRabD2c promoters, AtrabD2b/2c double mutants. In addition, some pollen indicated that both are widely expressed in most organs tubes from AtrabD2b/2c double mutants branch or and several cell types, with high expression in pollen. burst, which is not seen in pollen tubes of wild-type Root hairs also showed expression of AtRabD2b, and, plants or either single mutant. These data also indicate like pollen tubes, root hairs elongate by tip growth. that the loss of function of the AtRabD2b/2c genes can- However, root hair growth in the AtrabD2b/2c double not be compensated for by the AtRabD1 or AtRabD2a mutant is indistinguishable from that of wild-type genes, suggesting that either some function(s) of the plants. This is consistent with the idea that AtRabD2b AtRabD2b and AtRabD2c proteins are distinct from and AtRabD2c are required for vesicle trafficking in those of AtRabD1 or AtRabD2a, or that they are not multiple cell types, and that the highest demand for this expressed in the same cell types. process may be in pollen and pollen tubes, in order to Both AtRabD2b and AtRabD2c co-localize with the optimally support the large secretory requirement of Golgi marker ST-YFP upon transient expression in Ara- these very rapidly elongating cells. In combination, these data indicate that the high expression of AtRabD2b and bidopsis leaf protoplasts, as was reported also for AtRabD2c in pollen may be important to facilitate AtRabD2a (formerly called AtRab1b) [15]. It is therefore possible that AtRabD2b and AtRabD2c function in vesi- membrane trafficking needed for pollen tube growth. cle trafficking between the ER and Golgi apparatus, as does AtRabD2a [15,22]. Complete disruption of Rab Conclusions function in ER-to-Golgi trafficking is expected to be In summary, we used a T-DNA insertion mutant lethal, due to loss of plasma membrane, vacuole and cell approach to demonstrate the function of AtRabD2b and wall assembly and integrity. However, the AtrabD2b/2c AtRabD2c. Our data indicated that both are Golgi resi- double mutant is indistinguishable from wild-type plants, dents; they have similar but not identical expression except for shorter siliques due to the pollen and pollen patterns, but are both highly expressed in pollen; they tube defects. There are several possible explanations for are both involved in tip growth of pollen tubes; and this. First, other AtRabs must perform the same function they are at least partially functionally redundant. Future in vegetative tissue. The most likely candidates are work will focus on elucidating the molecular basis for AtRabD2a and AtRabD1, which could compensate for the pollen phenotype in the AtrabD2b/2c double thelossoffunctioninthe AtrabD2b/2c mutant in most mutant. cell types [22]. Moreover, other Rab families, such as tobacco RabBs (NtRab2s) have also been shown to be Methods regulators of membrane trafficking between the ER and Plant Materials and Growth Conditions Golgi apparatus [16]. AtRabBs (AtRab2s) may have the Wild-type Arabidopsis (Arabidopsis thaliana) ecotype same function, such that they also participate in ER to Columbia (Col-0), AtrabD2c-1, AtrabD2b-1 and Golgi vesicle trafficking. The pattern of AtRabB1b RNA AtrabD2b/2c (crosses of AtrabD2b-1 and AtrabD2c-1) accumulation is most highly correlated with that of mutants in the same genetic background were used. AtRabD2b (67%) and AtRabD2c (68%) (Additional file 1, Seeds were sown in Sunshine Soil mix, incubated at 4°C Table S1). These genes might compensate in part for the for 2 to 3 days, then grown at 22°C, 70% relative humid- loss of function of AtrabD2b/2c. ity, in a 16-h light/8-h dark photoperiod [26]. Peng et al. BMC Plant Biology 2011, 11:25 Page 14 of 16 http://www.biomedcentral.com/1471-2229/11/25 Screening for T-DNA insertion mutants In vitro pollen germination and growth measurement T-DNA insertion mutants of AtRabD2b and AtRabD2c Pollen was obtained from flowers collected from Arabi- (Salk_045030 (AtrabD2b-1), Salk_117532 (AtrabD2b-2) dopsis plants (ten plant lines per genotype) 1 to 2 weeks and Salk_120116 (AtrabD2b-3) for AtrabD2b; after bolting. Pollen from AtrabD2b/2c, AtrabD2b and Salk_054626 (AtrabD2c-3) for AtrabD2c) were obtained AtrabD2c mutants, along with pollen from wild-type from ABRC [38]. Homozygous lines for T-DNA inser- plants, was germinated on agar medium containing 18% tions were identified by PCR genotyping. For each (w/v) sucrose, 0.01% (w/v) boric acid, 1mM MgSO , T-DNA insertion mutant, two sets of PCRs were per- 1mM CaCl ,1mM Ca(NO ) , and 0.5% (w/v) agar, pH 2 3 2 formed using genomic DNA as a template: one with a 7.0[39] overnightatroomtemperature and examined gene-specific primer and a T-DNA left border primer and photographed under a Zeiss Axioplan II compound LBb1, the second with two gene-specific primers. The microscope equipped with an AxioCam color digital PCR products were sequenced to confirm the locations camera. Measurements were performed using SIS Pro of the T-DNA insertion sites for all of the mutants. The software (OSIS, Lakewood, CO) using the bars in the ori- gene specific primers used are listed in Table 2. ginal image. For pollen tube length measurements, 200 pollen tubes were chosen randomly for each genotype, Crossing and screening for double mutant and significance was assessed using Student’s t-test. Single mutant alleles (AtrabD2b-1 and AtrabD2c-1; For fluorescence microscopy, the germinated pollen Figure 1A) were crossed, the F1 generation of these was transferred onto a slide and two drops of aniline crosses was allowed to self fertilization and the blue solution (0.005% aniline blue solution in 0.1 M AtrabD2b/2c double mutant was identified from the F2 sodium phosphate, pH 7.0) were added for ten minutes. generation by PCR genotyping. To confirm the pollen tube growth defects, 20 open flowers per genotype were cut below the pistil and inserted Semi-quantitative reverse transcription PCR vertically into germination medium in a 9-cm Petri dish. Total RNA was extracted from leaves of 20 DAI (days Plates were sealed and incubated overnight at 22°C at after imbibition) plants using the TRIZOL reagent 100% humidity under continuous illumination. The paths (Invitrogen). RT-PCR was performed using Super- of pollen tubes inside the pistils were visualized by fixing whole pistils in 2% glutaraldehyde and 2% paraformalde- Script™ III One-Step RT-PCR System (Invitrogen,) as per the manufacturer’smanual. The b-tubulin gene, hyde in 0.1 M sodium cacodylate buffer, pH 7.2, under which is highly conserved and constitutively expressed low vacuum (18 psi Hg) for 2 h at room temperature. in all eukaryotes, was used as a standard. The primers Samples were washed three times in the same buffer and used are listed in Table 2. The RT-PCR products were stained with Aniline Blue and DAPI. The tissue was then sequenced to confirm the correct amplification cleared for 24 hours at room temperature with a drop of product. clearing solution (240 g of chloral hydrate and 30 g of gly- cerol in 90 ml water). Pollen was examined with a Zeiss Axioplan 2 light microscope (LM) and images were cap- tured with a Zeiss AxioCamHRc digital camera (Carl Zeiss, Inc., Thornwood, NY) using AxioVision 4.3 soft- Table 2 Primers used in this study ware. The microscope was equipped with a DAPI filter set LBb1 GCGTGGACCGCTTGCTGCAACT comprising an excitation filter (BP 365/12 nm), a beam AtrabD2b-LP1 CCCTTCGTTGGGCTAGTAAAG splitter (395 nm), and an emission filter (LP 397 nm). The AtrabD2b-RP1 TTCAACAACGTCAAACAATGG objectives used for imaging were a Neofluar 40× oil, an AtrabD2c-LP1 GCGCATTACTGAGAGAGAAGAG Apochromat 63× oil, and a Neofluar 100× oil. AtrabD2c-RP1 TCCCATTCTTGGAAACAAGTG AtRabD2b-F ATGAATCCTGAATATGACTAT Cloning AtRabD2b-R TCAAGAAGAACAACAGCCT Promoter::GFP/GUS fusion constructs were made for AtRabD2c-F ATGAATCCTGAATATGACTAT each gene by cloning the amplified promoter region AtRabD2c-R TTAAGAGGAGCAGCAGCCT (intergenic region; 964 bp for AtRabD2b and 558 bp for AtRabD2c) into the binary vector pBGWFS7 (GATE- AtRabD2b-g-F caccATCGCTTATCCGCTCCGTGTATTTC WAY; Invitrogen). AtRabD2b-g-R TAAAGACCCCTGGTCCTTCAGC The genomic fragments containing AtRabD2b or AtRabD2c-g-F caccCTATCTCACTAAGCTGAAGATAC AtRabD2c with their respective promoters for comple- AtRabD2c-g-R GGCAATCTCTCCGGTTTGGTCC mentation of the mutant phenotype were amplified b-Tubulin-F CGTGGATCACAGCAATACAGAGCC using AtRabD2b-g-F and R or AtRabD2c-g-F and R pri- b-Tubulin-R CCTCCTGCACTTCCACTTCGTCTTC mers (Table 2). Products were cloned into the pENTR/ Peng et al. BMC Plant Biology 2011, 11:25 Page 15 of 16 http://www.biomedcentral.com/1471-2229/11/25 D vector (Invitrogen), and then were transferred into the (18 psi Hg) for 5 h at room temperature. Samples were pMDC123 binary vector for plant transformation. washed three times in the same buffer, postfixed in 1% osmium tetroxide in the same buffer for 2 h and washed Plant transformation and selection twotimes in thesamebuffer, followed by deionized Arabidopsis plants were transformed using Agrobacter- water. Samples were dehydrated through a graded etha- ium tumefaciens by the floral dip method [40] and nol series (50, 70, 85, 95, and 100%; 30 min per step), selected for Basta resistance conferred by the T-DNA. followed by two changes of ultrapure 100% ethanol, all 30 min per step. Fresh pollen was also examined with- Transcriptomic analysis out fixing. Fixed samples were critical point-dried in a MetaOmGraph (MOG; http://www.metnetdb.org) [25] DCP-1 Denton critical-point-drying apparatus (http:// was used to analyze expression patterns of AtRabD1, www.dentonvacuum.com) using liquid carbon dioxide, AtRabD2a, AtRabD2b and AtRabD2c and derive the and mounted on aluminum stubs with double-sided correlation between them. sticky pads and silver cement. Samples were then sputter-coated with 15 nm gold GUS assay (20%) and palladium (80%) in a Denton Vacuum LLC Transgenic T2 seedlings were germinated in soil and har- Desk II Cold Sputter Unit (http://www.dentonvacuum. vested at various stages of development. Plants or organs com), and viewed with a JEOL 5800LV SEM (http://www. were stained at room temperature overnight as described jeol.com) at 10 kV. Alternatively, released fresh pollen [41], then destained in 70% (v/v) ethanol. For each con- grains were directly mounted on stubs and sputter-coated struct, at least 7 independently transformed lines, 7 plants with gold particles before SEM analysis. All digitally col- for each stage, were harvested for GUS screening. lected images including the LM and SEM images were processed in Adobe PhotoShop 7.0 and made into plates Transient expression in protoplasts using Adobe Illustrator 10. Over 20 samples from each Transient gene expression in Arabidopsis mesophyll plant line were used for SEM or LM analysis. protoplasts was carried out as described previously [42]. In brief, Arabidopsis protoplasts were isolated from the Additional material leaves of 3-4 week old plants. Leaf strips were digested in a buffer containing cellulose R-10 and macerozyme Additional file 1: Table S1. Expression pattern of AtRab genes. Pearson correlation between expression patterns of AtRab genes R-10. After adding 30 μg of plasmid DNA, an equal determined using MetaOmGraph (Excel file). volume of protoplasts was mixed with PEG buffer (40% Additional file 2: Figure S1. Seed number per silique in wild-type (w/v) PEG4000, 25% (v/v) 0.8M mannitol, 10% 1M and mutant plants. Seed number was counted for 15 siliques of 5 CaCl ) then incubated at room temperature for 25 min. individual plants for the indicated genotypes. Error bars indicate standard deviation (pdf file). After gentle washing, the protoplasts were kept in the Additional file 3: Figure S2. Controls for confocal microscopy. dark at room temperature overnight and then viewed by Arabidopsis leaf protoplasts were transformed with either GFP-AtRabD2b confocal laser scanning microscopy as described below. or ST-YFP and imaged in the green, yellow and red channels as shown in Figure 10. No cross-talk between channels could be seen using these settings. Upper panel, GFP-AtRabD2b; lower panel, ST-YFP. Scale bar = 10 Confocal laser scanning microscopy μm (pdf file). Colocalization of GFP-RabD2b and GFP-RabD2c with ST-YFP was performed using a Leica TCS SP10 confo- cal microscope, which allows flexible selection of emis- Acknowledgements sion bandwidths to minimize bleed-through. We are grateful to Ian Moore, University of Oxford, United Kingdom for Transformed cells were excited with a 488 nm laser kindly providing the N-ST-YFP construct and for helpful suggestions about (power 20%) and 514 nm laser (50% power), and GFP the Rab genes. We also thank the Arabidopsis Biological Resource Center and the Salk Institute Genomic Analysis Laboratory for providing T-DNA and YFP signals were collected using 495-510 nm and insertion mutants. This research was supported in part by grant MCB- 560-640 nm bandwidths, respectively. Non-transformed 0951170 from the National Science Foundation to ESW and grant no. cells and cells expressing asingleGFP or YFPfusion NNX09AK78G from the National Aeronautics and Space Administration to DCB. were used as controls to confirm the absence of cross talk between GFP, YFP and autofluorescence signals. Authors’ contributions JP carried out the experimental analyses described and drafted the manuscript. HI helped with the microscopy and figures. ESW conceived of Scanning electron microscopy the study, participated in its design and analysis of the data and helped to Pollen that had been germinated in vitro was placed in draft the manuscript. DCB participated in the design of the study, analysis of 2% glutaraldehyde and 2% paraformaldehyde in 0.1 M the data and helped to draft the manuscript. All authors read and approved the final manuscript. sodium cacodylate buffer, pH 7.2, under low vacuum Peng et al. BMC Plant Biology 2011, 11:25 Page 16 of 16 http://www.biomedcentral.com/1471-2229/11/25 Received: 3 September 2010 Accepted: 26 January 2011 25. Mentzen WI, Peng J, Ransom N, Nikolau BJ, Wurtele ES: Articulation of Published: 26 January 2011 three core metabolic processes in Arabidopsis: fatty acid biosynthesis, leucine catabolism and starch metabolism. BMC Plant Biol 2008, 8:76. 26. Li L, Ilarslan H, James MG, Myers AM, Wurtele ES: Genome wide co- References expression among the starch debranching enzyme genes AtISA1, 1. Boguski MS, McCormick F: Proteins regulating Ras and its relatives. Nature AtISA2, and AtISA3 in Arabidopsis thaliana. J Exp Bot 2007, 1993, 366(6456):643-654. 58(12):3323-3342. 2. Takai Y, Sasaki T, Matozaki T: Small GTP-binding proteins. Physiol Rev 2001, 27. Zheng H, Camacho L, Wee E, Batoko H, Legen J, Leaver C, Malhó R, 81(1):153-208. Hussey P, Moore I: A Rab-E GTPase mutant acts downstream of the Rab- 3. Geldner N, Friml J, Stierhof YD, Jurgens G, Palme K: Auxin transport D subclass in biosynthetic membrane traffic to the plasma membrane in inhibitors block PIN1 cycling and vesicle trafficking. Nature 2001, tobacco leaf epidermis. Plant Cell 2005, 17(7):2020-2036. 413(6854):425-428. 28. Plutner H, Cox AD, Pind S, Khosravi-Far R, Bourne JR, Schwaninger R, Der CJ, 4. Molendijk AJ, Ruperti B, Palme K: Small GTPases in vesicle trafficking. Curr Balch WE: Rab1b regulates vesicular transport between the endoplasmic Opin Plant Biol 2004, 7(6):694-700. reticulum and successive Golgi compartments. J Cell Biol 1991, 5. Surpin M, Raikhel N: Traffic jams affect plant development and signal 115(1):31-43. transduction. Nat Rev Mol Cell Biol 2004, 5(2):100-109. 29. Nuoffer C, Davidson HW, Matteson J, Meinkoth J, Balch WE: A GDP-bound 6. Waizenegger I, Lukowitz W, Assaad F, Schwarz H, Jurgens G, Mayer U: The of rab1 inhibits protein export from the endoplasmic reticulum and Arabidopsis KNOLLE and KEULE genes interact to promote vesicle fusion transport between Golgi compartments. J Cell Biol 1994, 125(2):225-237. during cytokinesis. Curr Biol 2000, 10(21):1371-1374. 30. Tisdale EJ, Balch WE: Rab2 is essential for the maturation of pre-Golgi 7. Schimmoller F, Simon I, Pfeffer SR: Rab GTPases, directors of vesicle intermediates. J Biol Chem 1996, 271(46):29372-29379. docking. J Biol Chem 1998, 273(35):22161-22164. 31. Tisdale EJ, Jackson MR: Rab2 protein enhances coatomer recruitment to 8. Sanderfoot AA, Raikhel NV: The specificity of vesicle trafficking: coat pre-Golgi intermediates. J Biol Chem 1998, 273(27):17269-17277. proteins and SNAREs. Plant Cell 1999, 11(4):629-642. 32. Tisdale EJ: A Rab2 mutant with impaired GTPase activity stimulates 9. Armstrong J: How do Rab proteins function in membrane traffic? Int J vesicle formation from pre-Golgi intermediates. Mol Biol Cell 1999, Biochem Cell Biol 2000, 32(3):303-307. 10(6):1837-1849. 10. Tamm LK, Crane J, Kiessling V: Membrane fusion: a structural perspective 33. Allan BB, Moyer BD, Balch WE: Rab1 recruitment of p115 into a cis-SNARE on the interplay of lipids and proteins. Curr Opin Struct Biol 2003, complex: programming budding COPII vesicles for fusion. Science 2000, 13(4):453-466. 289(5478):444-448. 11. Pfeffer S, Aivazian D: Targeting Rab GTPases to distinct membrane 34. Moyer BD, Allan BB, Balch WE: Rab1 interaction with a GM130 effector compartments. Nat Rev Mol Cell Biol 2004, 5(11):886-896. complex regulates COPII vesicle cis–Golgi tethering. Traffic 2001, 12. Seabra MC, Wasmeier C: Controlling the location and activation of Rab 2(4):268-276. GTPases. Curr Opin Cell Biol 2004, 16(4):451-457. 35. Hepler PK, Vidali L, Cheung AY: Polarized cell growth in higher plants. 13. Vernoud V, Horton AC, Yang Z, Nielsen E: Analysis of the small GTPase Annu Rev Cell Dev Biol 2001, 17:159-187. gene superfamily of Arabidopsis. Plant Physiol 2003, 131(3):1191-1208. 36. Schiott M, Romanowsky SM, Baekgaard L, Jakobsen MK, Palmgren MG, 14. Rutherford S, Moore I: The Arabidopsis Rab GTPase family: another Harper JF: A plant plasma membrane Ca2+ pump is required for normal enigma variation. Curr Opin Plant Biol 2002, 5(6):518-528. pollen tube growth and fertilization. Proc Natl Acad Sci USA 2004, 15. Batoko H, Zheng HQ, Hawes C, Moore I: A rab1 GTPase is required for 101(25):9502-9507. transport between the endoplasmic reticulum and golgi apparatus and 37. Cheung AY, Wang H, Wu HM: A floral transmitting tissue-specific for normal golgi movement in plants. Plant Cell 2000, 12(11):2201-2218. glycoprotein attracts pollen tubes and stimulates their growth. Cell 1995, 16. Cheung AY, Chen CY, Glaven RH, de Graaf BH, Vidali L, Hepler PK, Wu HM: 82(3):383-393. Rab2 GTPase regulates vesicle trafficking between the endoplasmic 38. Alonso JM, Stepanova AN, Leisse TJ, Kim CJ, Chen H, Shinn P, reticulum and the Golgi bodies and is important to pollen tube growth. Stevenson DK, Zimmerman J, Barajas P, Cheuk R, Gadrinab C, Heller C, Plant Cell 2002, 14(4):945-962. Jeske A, Koesema E, Meyers CC, Parker H, Prednis L, Ansari Y, Choy N, 17. Grebe M, Xu J, Mobius W, Ueda T, Nakano A, Geuze HJ, Rook MB, Deen H, Geralt M, Hazari N, Hom E, Karnes M, Mulholland C, Ndubaku R, Scheres B: Arabidopsis sterol endocytosis involves actin-mediated Schmidt I, Guzman P, Aguilar-Henonin L, Schmid M, Weigel D, Carter DE, trafficking via ARA6-positive early endosomes. Curr Biol 2003, Marchand T, Risseeuw E, Brogden D, Zeko A, Crosby WL, Berry CC, Ecker JR: 13(16):1378-1387. Genome-wide insertional mutagenesis of Arabidopsis thaliana. Science 18. Preuss ML, Serna J, Falbel TG, Bednarek SY, Nielsen E: The Arabidopsis Rab 2003, 301(5633):653-657. GTPase RabA4b localizes to the tips of growing root hair cells. Plant Cell 39. Li H, Lin Y, Heath RM, Zhu MX, Yang Z: Control of pollen tube tip growth 2004, 16(6):1589-1603. by a Rop GTPase-dependent pathway that leads to tip-localized calcium 19. Ueda T, Uemura T, Sato MH, Nakano A: Functional differentiation of influx. Plant Cell 1999, 11(9):1731-1742. endosomes in Arabidopsis cells. Plant J 2004, 40(5):783-789. 40. Clough SJ, Bent AF: Floral dip: a simplified method for Agrobacterium- 20. de Graaf BH, Cheung AY, Andreyeva T, Levasseur K, Kieliszewski M, Wu HM: mediated transformation of Arabidopsis thaliana. Plant J 1998, Rab11 GTPase-regulated membrane trafficking is crucial for tip-focused 16(6):735-743. pollen tube growth in tobacco. Plant Cell 2005, 17(9):2564-2579. 41. Jefferson RA, Kavanagh TA, Bevan MW: GUS fusions: beta-glucuronidase as 21. Tisdale EJ, Bourne JR, Khosravi-Far R, Der CJ, Balch WE: GTP-binding a sensitive and versatile gene fusion marker in higher plants. EMBO J mutants of rab1 and rab2 are potent inhibitors of vesicular transport 1987, 6(13):3901-3907. from the endoplasmic reticulum to the Golgi complex. J Cell Biol 1992, 42. Sheen J: Signal transduction in maize and Arabidopsis mesophyll 119(4):749-761. protoplasts. Plant Physiol 2001, 127(4):1466-1475. 22. Pinheiro H, Samalova M, Geldner N, Chory J, Martinez A, Moore I: Genetic evidence that the higher plant Rab-D1 and Rab-D2 GTPases exhibit doi:10.1186/1471-2229-11-25 distinct but overlapping interactions in the early secretory pathway. Cite this article as: Peng et al.: AtRabD2b and AtRabD2c have J Cell Sci 2009, 122(Pt 20):3749-3758. overlapping functions in pollen development and pollen tube growth. 23. Geldner N, Denervaud-Tendon V, Hyman DL, Mayer U, Stierhof YD, Chory J: BMC Plant Biology 2011 11:25. Rapid, combinatorial analysis of membrane compartments in intact plants with a multicolor marker set. Plant J 2009, 59(1):169-178. 24. Saint-Jore C, Evins J, Batoko H, Brandizzi F, Moore I, Hawes C: Redistribution of membrane proteins between the Golgi apparatus and endoplasmic reticulum in plants is reversible and not dependent on cytoskeletal networks. Plant J 2002, 29(5):661-678.

Journal

BMC Plant Biology – Springer Journals

Published: Jan 26, 2011

There are no references for this article.