Plant Physiol.
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

First published online October 27, 2006; 10.1104/pp.106.086363

Plant Physiology 142:1480-1492 (2006)
© 2006 American Society of Plant Biologists

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Supplemental Data
Right arrow All Versions of this Article:
142/4/1480    most recent
pp.106.086363v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Related articles in Plant Physiol.
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (3)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Drakakaki, G.
Right arrow Articles by Raikhel, N.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Drakakaki, G.
Right arrow Articles by Raikhel, N.
Agricola
Right arrow Articles by Drakakaki, G.
Right arrow Articles by Raikhel, N.
CELL BIOLOGY AND SIGNAL TRANSDUCTION

Arabidopsis Reversibly Glycosylated Polypeptides 1 and 2 Are Essential for Pollen Development1,[W]

Georgia Drakakaki, Olga Zabotina, Ivan Delgado2, Stéphanie Robert, Kenneth Keegstra and Natasha Raikhel*

The Center for Plant Cell Biology and Department of Botany and Plant Sciences, University of California, Riverside, California 92521 (G.D., O.Z., I.D., S.R., N.R.); and Department of Energy Plant Research Laboratory, Michigan State University, East Lansing, Michigan 48824 (K.K.)


   ABSTRACT
TOP
 ABSTRACT
RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 LITERATURE CITED
 
Reversibly glycosylated polypeptides (RGPs) have been implicated in polysaccharide biosynthesis. To date, to our knowledge, no direct evidence exists for the involvement of RGPs in a particular biochemical process. The Arabidopsis (Arabidopsis thaliana) genome contains five RGP genes out of which RGP1 and RGP2 share the highest sequence identity. We characterized the native expression pattern of Arabidopsis RGP1 and RGP2 and used reverse genetics to investigate their respective functions. Although both genes are ubiquitously expressed, the highest levels are observed in actively growing tissues and in mature pollen, in particular. RGPs showed cytoplasmic and transient association with Golgi. In addition, both proteins colocalized in the same compartments and coimmunoprecipitated from plant cell extracts. Single-gene disruptions did not show any obvious morphological defects under greenhouse conditions, whereas the double-insertion mutant could not be recovered. We present evidence that the double mutant is lethal and demonstrate the critical role of RGPs, particularly in pollen development. Detailed analysis demonstrated that mutant pollen development is associated with abnormally enlarged vacuoles and a poorly defined inner cell wall layer, which consequently results in disintegration of the pollen structure during pollen mitosis I. Taken together, our results indicate that RGP1 and RGP2 are required during microspore development and pollen mitosis, either affecting cell division and/or vacuolar integrity.

Reversibly glycosylated polypeptides (RGPs) have been implicated in polysaccharide biosynthesis (Dhugga et al., 1991Go, 1997Go) because they are localized to the Golgi apparatus (Dhugga et al., 1997Go) and are able to react with UDP-Glc, UDP-Xyl, and UDP-Gal. The RGPs of a number of species have been studied: Arabidopsis (Arabidopsis thaliana; Delgado et al., 1998Go), maize (Zea mays; Rothschild and Tandecarz, 1994Go; Sagi et al., 2005Go), potato (Solanum tuberosum; Bocca et al., 1999Go), wheat (Triticum aestivum) and rice (Orzya sativa; Langeveld et al., 2002Go), cotton (Gossypium hirsutum; Zhao et al., 2001Go), and tomato (Lycopersicon esculentum; Selth et al., 2006Go). The RGPs of wheat, rice, and potato have been implicated in the initiation of starch biosynthesis (Bocca et al., 1997Go; Langeveld et al., 2002Go), although there is no direct evidence that confirms this hypothesis. However, the fact that none of the RGP sequences contain a predicted plastid-targeting signal renders this hypothesis unlikely. The existence of RGPs in both monocots and dicots and their absence in nonplant species indicates a plant-specific function.

RGPs have been detected in both the membrane and soluble fractions of most of the above-mentioned species. Furthermore, experiments conducted in maize and transgenic tobacco (Nicotiana tabacum) overexpressing Arabidopsis RGPs have shown them to be associated with the plasmodesmata in a Golgi-dependent manner (Sagi et al., 2005Go).

Although RGP was initially purified from pea (Pisum sativum) hypocotyls (Dhugga et al., 1991Go) and maize upper mesocotyls (Sagi et al., 2005Go), RGP1 was also found to be highly expressed in roots and in cell cultures in Arabidopsis (Delgado et al., 1998Go). Stably transformed tobacco plants containing a cotton RGP promoter::beta-glucuronidase fusion construct showed high expression in the roots, styles, and stigmas (Wu et al., 2006Go).

Data from transcriptome and proteomic analyses have shown that genes related to cell wall biosynthesis and regulation are highly expressed in Arabidopsis pollen (Becker et al., 2003Go; Honys and Twell, 2003Go; Holmes-Davis et al., 2005Go; Noir et al., 2005Go; Pina et al., 2005Go). This finding underscores the importance of the cell wall in the growth process of the pollen tube. Recent genetic evidence emphasizes the role of callose synthases and pectin methyl esterases (PMEs) for pollen plant development and fertility (Bosch et al., 2005Go; Bosch and Hepler, 2005Go; Enns et al., 2005Go; Jiang et al., 2005Go; Nishikawa et al., 2005Go). Other cell wall-related enzymes that have been shown to be important for pollen development include AtCSLA7 and UDP-sugar pyrophosphorylase (Goubet et al., 2003Go; Schnurr et al., 2006Go). Several monosaccharide transporters exhibiting high expression during various stages of pollen development have been identified (Truernit et al., 1996Go, 1999Go; Schneidereit et al., 2003Go, 2005Go; Scholz-Starke et al., 2003Go). Interestingly, RGP1 and RGP2 were also detected in Arabidopsis and rice pollen proteomes (Holmes-Davis et al., 2005Go; Noir et al., 2005Go; Dai et al., 2006Go).

The Arabidopsis genome contains five RGP genes (Girke et al., 2004Go). RGP1 (At3g02230) and RGP2 (At5g15650) have high sequence identity and similar gene structures and thus may have arisen through a gene duplication event (Blanc et al., 2000Go). To begin to understand the biological roles of RGPs, we created translational fusions of Arabidopsis RGP1 and RGP2 and determined their expression patterns and subcellular localization. We also undertook a genetic approach to determine the respective functions of AtRGP1 and AtRGP2. We demonstrated that a double-mutant cross between rgp1-1 and rgp2-1 resulted in lethality. Further analysis demonstrated the critical role of RGPs in pollen development.


   RESULTS
TOP
 ABSTRACT
 RESULTS
DISCUSSION
 MATERIALS AND METHODS
 LITERATURE CITED
 

The Arabidopsis RGP Gene Family

The Arabidopsis genome encodes five RGPs (Fig. 1 ). AtRGP1 and AtRGP2 share 93% identity at the amino acid level. RGP1 is 80% identical to RGP3 and 74% identical to RGP4. All RGPs are at least 43% identical to each other. Arabidopsis RGP1 was previously characterized and shown to reversibly autoglycosylate with UDP-Glc, UDP-Xyl, or UDP-Gal as substrates (Delgado et al., 1998Go). Because we have previously studied RGP1 (Delgado et al., 1998Go), and because among the Arabidopsis RGP genes it has the highest similarity to RGP2 (Girke et al., 2004Go), we decided to continue our studies on these two genes.


Figure 1
View larger version (13K):
[in this window]
[in a new window]
[as a PowerPoint slide]
  		Figure 1. Phylogenetic analysis of the RGP family in Arabidopsis. The alignments are used to calculate phylogenetic trees with the PHYLIP package (Felsenstein, 2005Go), as described in the Cell Wall Navigator database (Girke et al., 2004Go). Scale represents 10 substitutions per 100 amino acids. Bootstrap values for 100 replications are given at the branch points.

 

RGP1 and RGP2 Are Strongly Expressed in Actively Growing Tissues as Well as Pollen

To examine the potential tissue specificity of RGP1 and RGP2 expression, we utilized the fluorescent tagging of full-length proteins technique (Tian et al., 2004Go). In this approach, which has been shown to faithfully reproduce the native levels and expression patterns, as well as subcellular localizations of the fluorescently tagged gene products (Tian et al., 2004Go), RGP1 and RGP2 fused to yellow fluorescent protein (YFP) were expressed from their native promoters in both the presence and absence of 35S enhancers. The 35S enhancers have been shown to increase expression levels of YFP fusion proteins, but not to alter their subcellular localizations (Tian et al., 2004Go). Characterization of spatial expression patterns detected RGP1-YFP and RGP2-YFP in all tissues, but both were most strongly expressed in actively growing tissues. Figure 2, A and D , shows that RGP1-YFP was expressed predominantly in the root tip (Fig. 2A) and the apical meristem (Fig. 2D) of young seedlings. Strong expression of RGP1-YFP was also observed in lateral root primordia (Fig. 2C). RGP2-YFP, with expression augmented by the 35S enhancer, was expressed in the same tissues (Fig. 2B). More interestingly, during later developmental stages, both RGP fusions were highly expressed in pollen (Figs. 2E and 3, A and D ). These data are in agreement with in silico microarray analyses that show essentially the same expression patterns, with strong expression of AtRGP1 and AtRGP2 in roots, vegetative shoots, flowers, and mature pollen (http://csbdb.mpimp-golm.mpg.de). Microarray data from both the public databases and RGP1-YFP show that the highest expression levels are in tricellular pollen (Supplemental Fig. S1). The RGP1 transcript and protein have previously been detected at high levels in suspension cultures and roots (Delgado et al., 1998Go).


Figure 2
View larger version (62K):
[in this window]
[in a new window]
[as a PowerPoint slide]
  		Figure 2. Expression patterns of RGP1 and RGP2. Expression pattern of RGP1-YFP (A, C, D, and E) and RGP2-YFP (B) are primarily observed in the apical meristem and root tip in 6-d-old seedlings. C, High expression of RGP1-YFP occurred in lateral root primordia (C, inset). E, Expression pattern of RGP1-YFP in flowers is strongly localized in anthers (in mature pollen). The same pattern was observed for RGP2-YFP (data not shown). Bright field (A, B, and E, inset). YFP filter (A, B, C, and E). D, Overlay of single confocal images of YFP filter (yellow) and red autofluoresence (red). Bar = 400 µm (A, B, and D); 80 µm (C and E).

 

Figure 3
View larger version (67K):
[in this window]
[in a new window]
[as a PowerPoint slide]
  		Figure 3. Subcellular localization of RGP1-YFP and RGP2-YFP. A, RGP1-YFP in pollen tube, apical meristem (B), and root tip (C). Subcellular localization of RGP2-YFP (with 35S enhancers) in pollen tube (D), leaf (E), and root tip (F). All images are single confocal sections. Bar = 20 µm.

 
Confocal laser-scanning microscopy (CLSM) was used to examine the subcellular localization of the RGP-YFP fusions in various Arabidopsis organs. These studies revealed cytoplasmic (especially in roots) and punctate labeling for both the RGP1 and RGP2 fusion proteins (Fig. 3). The RGP2 fusion protein (with 35S enhancer) was observed in leaf epidermal cells as small fluorescent foci throughout the cytoplasm (Supplemental Video S1), indicating a characteristic pattern of association with Golgi. The punctate patterns were very pronounced and dynamic in growing pollen for both RGPs (Fig. 3, A and D; Supplemental Video S2). Occasionally, we observed larger fluorescing foci in RGP2-YFP leaves and hypocotyls, which was not indicative of a known pattern for a specific compartment (data not shown).


Arabidopsis RGPs Exhibit Cytoplasmic and Golgi-Associated Localization

Previous reports have shown that a significant amount of RGPs were present in soluble cellular fractions (Dhugga et al., 1991Go, 1997Go; Bocca et al., 1997Go, 1999Go), indicating that RGPs can be localized to the cytoplasm. However, it was assumed that most of the RGP present in the cytoplasm was released from the Golgi membranes or from plasmodesmata, respectively (Dhugga et al., 1997Go; Sagi et al., 2005Go). Consistent with the biochemical evidence (Dhugga et al., 1991Go; Delgado et al., 1998Go), our results clearly demonstrate that the fusion protein was present in the cytoplasm.

To validate the subcellular localization of Arabidopsis RGP1 and RGP2 in both the cytoplasm and Golgi, we performed electron microscopy immunolocalization on wild-type Arabidopsis roots. The pea RGP antibody was used for its high specificity and low background signal on western blots (Dhugga et al., 1997Go). Our analysis showed that Arabidopsis RGPs localized to both Golgi and cytoplasm (Fig. 4A ). These data are in agreement with the fluorescence localization of RGP1-YFP and RGP2-YFP.


Figure 4
View larger version (44K):
[in this window]
[in a new window]
[as a PowerPoint slide]
  		Figure 4. RGP1 and RGP2 are colocalized in the same compartment and interact in vivo. A, Immunolocalization of RGP in wild-type Arabidopsis roots. For the micrograph, ultrathin cryosections of wild-type Arabidopsis roots were incubated with pea RGP antibody (i). Control probed with 2% bovine serum albumin instead of antiserum (ii). Arrows indicate Golgi localization versus cytoplasm. B, RGP1 and RGP2 colocalize in Golgi-associated compartments. For the micrograph, ultrathin cryosections of Arabidopsis transgenic plants expressing T7-RGP1 and HA-RGP2 were incubated with T7 and HA antibodies (i and ii). Arrows indicate 15-nm gold particles (labeling T7-RGP1); arrowheads indicate 10-nm gold particles (labeling HA-RGP2). Bar = 100 nm (except in B [ii], 50 nm); G, Golgi apparatus. C, RGP2 immunoprecipitates with RGP1. EX, Soluble extracts (30%–50% AS) were prepared from whole Arabidopsis plants expressing T7-RGP1 and HA-RGP2. The extract was immunoprecipitated with agarose-immobilized T7 monoclonal antibody (EL) or agarose beads with no antibody (lane C). Soluble extract and eluents were separated by SDS-PAGE and probed with T7 or HA antibodies.

 
Subcellular fractionation of fluorescence-tagged plants was used to verify the Golgi association of the YFP fusion proteins. Both fusion proteins were detected in the Golgi/microsomal-enriched fraction (Supplemental Fig. S2).

Brefeldin A (BFA), a trafficking inhibitor that disrupts the Golgi apparatus and results in aggregates of Golgi membranes (for review, see Nebenfuhr et al., 2002Go), was used to further examine the association of the RGP fusions with this compartment. Germinated pollen (Supplemental Fig. S3) or root cells from the elongation zone (data not shown) that were treated with BFA displayed the characteristic fluorescent aggregates referred to as BFA bodies. The endocytic marker FM4-64 (Bolte et al., 2004Go) also colocalized with the BFA bodies in germinated pollen (Supplemental Fig. S3) or root cells (data not shown). These observations further demonstrate that the fluorescence pattern for RGPs was indicative of Golgi or other membrane association. Our results are consistent with previous observations that RGPs are soluble and peripherally associated with membranes (Delgado et al., 1998Go).

Because RGPs have been observed in plasmodesmata in maize and in transgenic tobacco overexpressing AtRGP2:GFP (Sagi et al., 2005Go), we attempted to determine whether Arabidopsis RGPs localized to plasmodesmata. We carried out plasmolysis treatments on roots in Arabidopsis plants expressing RGP1-YFP or RGP2-YFP and, despite an intensive search via CLSM, we observed no labeling in plasmodesmata (data not shown). This result could indicate species-specific localization or it could illustrate technical limitations that are unique to Arabidopsis. It will be interesting in the future to investigate whether RGPs are present in other endomembrane compartments.

The experiments described above show essentially identical expression patterns for RGP1 and RGP2. We next sought to determine whether they colocalize to the same compartments. To address this question, we created two epitope-tagged plant lines, T7-RGP1 and HA-RGP2. This approach allowed us to distinguish between the two nearly identical proteins. We performed double immunogold labeling using commercially available antibodies raised against the T7 and HA epitopes. Our results showed that HA (10 nm) and T7 (15 nm) labels were associated with the Golgi apparatus (Fig. 4B [i and ii]). We also observed cytoplasmic labeling with both sizes of gold particles (data not shown). Therefore, the RGP1 and RGP2 proteins both colocalized to Golgi-associated compartments and to the cytoplasm.

Because RGP1 and RGP2 colocalized to the same cellular compartments, we investigated whether they may be present in the same protein complex. For this analysis, we used the double-tagged plants from the previous experiment. Soluble extracts from the double-tagged plants were incubated with agarose-conjugated T7 antibody or agarose beads with no antibody as a control. The beads were then washed and the eluates were separated by SDS-PAGE and probed with HA or T7 antibodies. As shown in Figure 4C, HA-RGP2 was coprecipitated with T7-RGP1 in the eluate (lane EL). No significant amounts of HA-RGP2 or T7-RGP1 were eluted in the control (Fig. 4C, lane C). Thus, these data led to the conclusion that RGP1 and RGP2 interact in vivo.


Genetic Analysis of rgp1 and rgp2 and Double-Mutant Studies

We took a reverse-genetics approach to understand the functions of the RGP1 and RGP2 proteins. Mutant plant lines with T-DNA insertions in each of the genes were obtained from different collections. The rgp1-1 line (Gabi_652F12) contained an insert in the first intron of the RGP1 gene, approximately 763 bp downstream from the start codon (Fig. 5A [i] ). An rgp2-1 mutant was obtained from the Salk collection (SALK_132152); it contained a T-DNA insertion in the last exon of RGP2, approximately 150 nucleotides upstream of the stop codon (Fig. 5A [ii]). As shown in Figure 5B, reverse transcription (RT)-PCR of transcripts isolated from homozygous rgp1 or rgp2 plants showed no mRNAs for RGP1 or RGP2, respectively. We carried out RT-PCR of the Arabidopsis UBIQUITIN10 gene as a control (Fig. 5B). Under normal growth conditions, we did not observe any obvious phenotypes associated with the RGP1 or RGP2 T-DNA insertions (Fig. 5C).


Figure 5
View larger version (38K):
[in this window]
[in a new window]
[as a PowerPoint slide]
  		Figure 5. rgp1-1 and rgp2-1 mutants and their genetic cross. A, T-DNA insertion in RGP1 and RGP2. T-DNA insertion is located in the first intron in rgp1-1 (i). T-DNA insertion is located in the last exon in rgp2-1 (ii). Arrows indicate the position of the primers used for RT-PCR. B, RT-PCR of RGP1 and RGP2. Total RNAs from seedlings of UBIQUITIN 10 using the same templates with gene-specific primers are shown at the bottom. C, None of the mutant plants of RGP1 or RGP2 exhibited any obvious differences under normal growth conditions as compared to wild type. D, Effect of rgp1-1 and/or rgp2-1 mutation on seed development. Whole-mount images of the mature silique cleared by NaOH solution. i, rgp1/rgp1; ii, rgp2/rgp2; iii, RGP1/rgp1 RGP2/rgp2; iv, rgp1/rgp1 RGP2/rgp2. Arrowheads indicate unnatural intervals. Bar = 1 mm.

 
To address the possibility of genetic redundancy, we crossed the two single mutants to create a double mutant. We determined the zygosity of the RGP1 and RGP2 genes at their respective loci by genotyping 100 F2 progeny using PCR-based markers. The results of the genotyping experiments are reported in Table I . We found no homozygous double mutants among the 100 F2 progeny. Furthermore, we detected fewer rgp1/rgp1 RGP2/rgp2, and no RGP1/rgp1 rgp2/rgp2 plants. In an independent experiment using 303 F2 progeny, we came to the same conclusion, namely, no homozygous double mutant was identified (Supplemental Table S1). In this case, plants that contained only one functional copy of RGP1 (RGP1/rgp1 rgp2/rgp2) were identified; however, they did not survive after transfer to soil. We also examined the seed set from siliques removed from the F2 plants shown in Table I. Seeds from the selfing of wild type (data not shown) or the two single-mutant plants rarely aborted (Fig. 5D). However, siliques isolated from selfing of double heterozygous plants (RGP1/rgp1 RGP2/rgp2) or plants with only one functional copy of RGP2 (rgp1/rgp1 RGP2/rgp2) had vacant spaces (Fig. 5D). We did not observe aborted embryos or shriveled seeds. Therefore, analysis of the F2 population strongly supports the conclusion that the rgp1 rgp2 double mutant is lethal.


View this table:
[in this window]
[in a new window]

  		Table I. Summary of the lethality of rgp1-1 and rgp2-1 and mutant progeny

Percentage of genotypes (by PCR) observed in the F2 population following the genetic cross of rgp1-1 x rgp2-1. The percentage of the expected genotypes following the genetic cross of rgp1 x rgp2 in F2 is listed.

 

The Double-Mutant Defect Occurs in Pollen and Impacts Pollen Morphology

To examine more closely the nature of the lethal phenotype, we created reciprocal crosses of rgp1/rgp1 RGP2/rgp2 and the wild-type Columbia parent. When wild-type pollen was used to fertilize rgp1/rgp1 RGP2/rgp2 plants, we observed a reduction in the number of progeny that carried both mutant alleles, suggesting a reduction of rgp1 rgp2 transmission through the female gametophytes (Table II ). More interestingly, when pollen isolated from plants that carried one copy of RGP2 (rgp1/rgp1 RGP2/rgp2) was used to fertilize wild-type plants, we were able to detect only progeny with a T-DNA insert in RGP1. The RGP1/rgp1 RGP2/rgp2 genotype was not detected (Table II). These data indicate that pollen carrying both insertions (rgp1 rgp2) is not viable. Thus, failure during gametogenesis in rgp1 rgp2 pollen is the basis for the lethality of the double mutant.


View this table:
[in this window]
[in a new window]

  		Table II. Genetic analysis of rgp1/rgp1 RGP2/rgp2 gametophytes

Individuals inheriting the rgp1 or the rgp2 mutation were scored by PCR and then subjected to a X2 test. P, Statistical significance; n, sample size.

 
Because reciprocal crosses indicate that the male gametophyte is not viable in the homozygous rgp1 rgp2 double mutant, we closely examined the development of anthers in wild-type and rgp1/rgp1 RGP2/rgp2 mutants. We examined toluidine blue-stained sections of anthers at various stages of development using light microscopy. Anther development and microspore formation follow a well-established pattern in dicotyledonous plants (Sanders et al., 1999Go) and are separated into 14 stages. No morphological differences were observed between wild type and rgp1/rgp1 RGP2/rgp2 anthers at stages 8 to 10 (Fig. 6, A, B, C, and D , respectively). At this stage, meiosis was already complete, the callose walls surrounding the tetrads had degenerated, and microspores had been released. Following this stage, individual microspores further developed into pollen grains. During the later stages (stages 11–14), numerous pollen grains from the rgp1/rgp1 RGP2/rgp2 anthers were not completely developed (Figs. 6F and 7A ). Compared to mature wild-type pollen grains, many of those in mutant anthers were collapsed in appearance (Fig. 6, E and F, respectively). Taking into account the data from the reciprocal crosses, we concluded that the collapsed pollen grains (Figs. 6F and 7A) harbored mutations in both RGP1 and RGP2.


Figure 6
View larger version (109K):
[in this window]
[in a new window]
[as a PowerPoint slide]
  		Figure 6. Bright-field photographs of Arabidopsis wild-type and rgp1/rgp1 RGP2/rgp2 anthers in different developmental stages. Flower sections were stained with toluidine blue and photographed by bright-field microscopy. Locules from wild type (A, C, and E) and rgp1/rgp1 RGP2/rgp2 (B, D, and F) anthers at stages 8, 10, and 13, respectively. Arrows indicate collapsed mature pollen at stage 13. Bar = 50 µm.

 

Figure 7
View larger version (38K):
[in this window]
[in a new window]
[as a PowerPoint slide]
  		Figure 7. Loss of function of RGP1 and RGP2 confers aberrant pollen grain morphology and gametophytic lethality. A, Pollen grains, hand dissected from rgp1/rgp1 RGP2/rgp2 plants, contained round (arrowheads) and collapsed (arrows) pollen grains. Single confocal optical sections of pollen from rgp1/rgp1 RGP2/rgp2 anthers. B, FDA staining showed that collapsed pollen grains (B, inset) were nonviable. C, Staining with decolorized aniline blue did not clearly label the irregular-shaped pollen grains (arrows; C, inset) as compared to normal-looking pollen (arrowheads). D, Calcofluor white as well did not label the irregular-shaped pollen (D, arrows) as shown under light microscope (D, inset). All insets represent light microscope images of the corresponding confocal optical sections. E to G, SEMs of pollen grains from rgp1/rgp1 RGP2/rgp2 plants. E, There are two types of pollen grains: aborted (shrunken, indicated by arrow) and normal (arrowhead). F, Aborted pollen grains still had an intact exine layer, although in severe cases the spaces between reticulations were very small (G). Bar = 20 µm (A–D); 10 µm (E–G).

 
Mature pollen was released at various stages of development by hand dissection and further analyzed. We determined pollen viability by using fluorescence microscopy analysis after fluorescein diacetate (FDA) staining. Malformed pollen grains did not stain with FDA (Fig. 7B, arrows) and were incapable of germination in vitro, suggesting that they were not viable. We further investigated the small and flattened pollen grains for their nuclear constitution by staining with 4',6'-diamidino-2-phenylindole. In normal-looking pollen, the vegetative nucleus and the two generative nuclei were clearly distinguishable (Supplemental Fig. S4, arrowheads). In the malformed pollen, only diffuse staining was observed (Supplemental Fig. S4, arrows). Staining with calcofluor white (for cellulose) or aniline blue (for callose) failed to label the collapsed pollen grains (Fig. 7, C and D, arrows). Pollen released from single-mutant anthers did not show any differences in morphology or staining compared to that of wild-type ones (data not shown).

Using scanning electron microscopy (SEM), we examined more closely pollen from wild type or plants with only one functional copy of RGP2 (rgp1/rgp1 RGP2/rgp2). Although some pollen grains appeared normal (Fig. 7E, arrowheads), significant numbers of abnormal and collapsed pollen grains were detected in the rgp1/rgp1 RGP2/rgp2 line (Fig. 7, E [arrows]–G), which is consistent with light microscopic observations (Figs. 6F and 7A). Although mutant pollen grains were smaller in size and less robust, the basic structure of the exine layer was still present (Fig. 7, F and G).


Pollen Grains with One Functional Copy of RGP2 Are Arrested during Pollen Mitosis I

Cross sections of rgp1/rgp1 RGP2/rgp2 were further analyzed by transmission electron microscopy (TEM). Microspores did not show any structural abnormalities (Fig. 8, A–C ). Later in microspore development, the vacuoles fused into a single large vacuole during microspore expansion and nuclear migration (Fig. 8, D–F). During this stage, some of the spores had unusually large vacuoles (Fig. 8, E and F). Thereafter, the microspore underwent mitosis to produce a vegetative cell and a generative cell (GC). After nuclear division within the locule of a pollen sac, a mixture of pollen grains was found (Fig. 8, G–I). Wild-type-like pollen grains had undergone mitosis and the GC was attached at a peripheral position, surrounded by the hemispherical GC wall and fused with the intine layer (Fig. 8G). Significant numbers of pollen grains were arrested in the vacuolated stage (Fig. 8, H and I). These grains were not able to maintain turgor pressure and the vacuole eventually collapsed (Fig. 8I). Consistent with our observations using light and SEM during the later developmental stages, both wild-type-appearing (Fig. 8, J and L) and collapsed pollen grains were observed (Fig. 8, K and M).


Figure 8
View larger version (130K):
[in this window]
[in a new window]
[as a PowerPoint slide]
  		Figure 8. Development of rgp1/rgp1 RGP2/rgp2 plant spores. All spores of the same stage were observed in the same anther locule. A to C, Spores at the early microspore stages do not show any structural defects. D to F, Vacuolated spores at late microspore stages prior to PMI. Spores are dominated by large vacuoles. Some spores (E and F) contain unusually large vacuoles. G, Bicellular spore after PMI showing GC surrounded by cell wall. In the same locule, vacuolated microspores are arrested (H and I). Note that spore in I cannot maintain turgor. J and K, Spore at late bicellular stage. L and M, Tricellular mature pollen. Locules contain normally developed pollen (J and L) and collapsed pollen (K and M). Details of largely vacuolated pollen are indicated by an asterisk (N and P); they contain poorly defined intine (arrowheads) compared to the bicellular pollen (arrows; N and O). The exine layer is similar for both types of pollen. Bar = 2 µm (A–M); 0.5 µm (N–P). EX, Exine; IN, intine; VN, vegetative nucleus; V, vacuole.

 
A more detailed examination showed that the cell wall layer of the pectocellulosic intine in arrested pollen was not very well defined (Fig. 8, N and P, arrowheads). As we previously observed with SEM, despite the severe damage in the vegetative cells, the exine layers of these pollen grains were still normal in appearance (Fig. 8, N and P). Thus, our results suggested that RGP1 and RGP2 were essential during pollen development, specifically during the late microspore stages, and have impact on pollen cell wall deposition. At these stages, the highly vacuolated microspores were unable to further develop and undergo the first pollen mitosis, suggesting a possible effect on cell division.


   DISCUSSION
TOP
 ABSTRACT
 RESULTS
 DISCUSSION
MATERIALS AND METHODS
 LITERATURE CITED
 
We have demonstrated that the rgp1 rgp2 double mutant is lethal, underlying the critical role of RGPs, particularly in pollen development. We determined that the double mutant failed to complete the first pollen mitosis and suggested that RGP1 and RGP2 affect cell division and/or vacuolar integrity. We demonstrated in detail that RGPs are essential for pollen development and possibly for other stages in the life cycle of plants.


RGP1 and RGP2 Are Functionally Redundant

The Arabidopsis genome encodes five RGPs (Girke et al., 2004Go), one of which, RGP1, was previously characterized (Delgado et al., 1998Go). Plant RGPs are divided into two distinct phylogenetic classes (Langeveld et al., 2002Go); class 1 contains AtRGP1 to AtRGP4, as well as several other monocot and dicot RGPs, and class 2 is limited to RGPs from wheat and rice, as well as the remaining Arabidopsis paralog, AtRGP5. The highly conserved sequences of AtRGP1 and AtRGP2 suggest partial functional redundancy. Indeed, our genetic analysis confirms this hypothesis; single knockouts are viable and the double knockout is lethal.

Given that RGP1 to RGP4 are in the same class, it could still suggest that there may be some functional redundancy among the four family members. However, because the rgp1 rgp2 double mutant is lethal, we posit that the RGP3 and RGP4 proteins cannot compensate for their loss, at least during pollen development. If true, this deficiency may reflect true functional divergence within the family or it may be caused by different expression patterns or levels of expression of the other RGP genes. Ectopic expression of either RGP3 or RGP4 in the rgp1 rgp2 mutant background and determining whether they are sufficient to rescue the mutant phenotype may help distinguish between these possibilities. We propose that, in both single mutants (rgp1 or rgp2), sufficient RGP levels remain so that active RGP forms may function during pollen development.

Previous studies have shown that wheat RGP1 and RGP2 can assemble to form either homooligomers or heterooligomers with a predicted molecular mass of 230 kD (Langeveld et al., 2002Go). Potato RGP proteins are also predicted to form a complex, although, in this case, the architecture is thought to be pentameric or hexameric (Bocca et al., 1997Go). In Arabidopsis, the colocalization of RGP1 and RGP2 and coimmunoprecipitation provide evidence for their interaction. Because RGP proteins are thought to be more active as members of complexes (Langeveld et al., 2002Go), we anticipate that the Arabidopsis RGPs in a complex would show greater autoglycosylation efficiency than those present as monomers. It is also possible that the Arabidopsis RGP complex contains other RGP family members, a hypothesis that we are currently investigating.


RGP1 and RGP2 Expression Is Associated with Active Growth

Our results showed that RGP1 and RGP2 are highly expressed in actively growing tissues, which require synthesis of large amounts of cell wall components. This finding supports the hypothesis that RGPs are involved in cell wall biosynthesis and therefore in cell growth and division. In agreement with this result, RGPs have previously been found in growing tissues, such as the mesocotyl of pea and maize (Dhugga et al., 1991Go, 1997Go; Sagi et al., 2005Go), Arabidopsis (Delgado et al., 1998Go), and cotton (Zhao et al., 2001Go).

The high expression levels of RGP1 and RGP2 in pollen suggest an important function during pollen development. In agreement with the expression patterns, our genetic analysis showed that rgp1 rpg2 double-mutant pollen, derived from plants with one functional copy of RGP2, has severe defects in development.

The cotton RGP promoter::beta-glucuronidase fusion was expressed strongly in the roots, styles, and stigma of transgenic tobacco plants, but less strongly in pollen, indicating species-specific expression patterns (Wu et al., 2006Go). RGPs are also strongly expressed in pollinated melon (Cucumis melo; Nagasawa et al., 2005Go) and in Medicago seeds (stage 18 d after pollination; Gallardo et al., 2003Go), providing further evidence for a crucial role after pollination.

RGPs are present both in the cytoplasm (where many aspects of sugar nucleotide synthesis and metabolism occur) and at the surface of the Golgi compartments (where matrix polysaccharides are synthesized). As previously stated, phylogenetic analysis divides the RGPs into two classes. Langeveld et al. (2002)Go have suggested that the class 1 RGPs (AtRGPs 1–4) may play a role in cell wall synthesis and class 2 RGPs (AtRGP5) may be required for starch biosynthesis. Indeed, our localization results are in agreement with the possible role of RGP1 and RGP2 in cell wall polysaccharide biosynthesis. Sagi et al. (2005)Go proposed that the final destination of RGPs could be the cell wall or plasmodesmata, but we were unable to confirm this hypothesis for Arabidopsis. Continuing research on RGP3, RGP4, and RGP5 is necessary to test whether they might function in starch synthesis.


RGP1 and RGP2 Are Crucial for Normal Male Gametophyte Development

Our genetic analysis clearly demonstrates the importance of RGPs by illustrating that the rgp1 rgp2 double mutant is lethal and suggests that RGP1 and RGP2 are essential, at least for pollen development. Application of cell wall stains to wild type and pollen grains of double mutants with one copy of RGP showed that the stains failed to label the collapsed rpg1 rgp2 double-mutant pollen. Ultrastructural analysis showed that the double-mutant pollen had irregular organization with poorly defined intine (pectocellulosic cell wall layer). Taken together, these results suggested that the loss of RGP function affects cell wall formation in the pollen grains of the double mutant. As previously stated, RGPs have been implicated in noncellulosic polysaccharide biosynthesis (Dhugga et al., 1997Go; Delgado et al., 1998Go). Because the double mutant is male gametophytically lethal, it is expected that, in the double mutant pollen, cell wall formation could be affected. It would therefore be interesting in the future to find out which polysaccharides are affected in the double mutant.

There are numerous examples of male gametophytic mutants that have defects in cell wall formation, particularly in callose biosynthesis and pectin esterification. Mutants defective in callose biosynthesis exhibit several gametophytic defects. For example, the Arabidopsis callose synthase 5 (calS5) mutant has severe defects in the exine wall, although weaker alleles allow fertile pollen to be produced (Dong et al., 2005Go; Nishikawa et al., 2005Go). The CalS11 (gsl1) and CalS12 (gsl5) genes were shown to have partially overlapping roles in the formation of the callose wall that separates the microspores within the tetrad and, later, during pollen grain maturation (Enns et al., 2005Go). The rgp1 rpg2 double mutant formed microspores that are normal in appearance, but arrested at a postmeiotic stage, suggesting that tetrad formation and microsporogenesis were not affected in this background. Because there were no obvious defects during tetrad formation, or in the exine layer, and because RGP1-YFP and RGP2-YFP were not confined to the callose plug of the germinated pollen tubes (data not shown), we concluded that RGPs do not play a major or indispensable role in callose biosynthesis.

Pectin is another major component of pollen grain tube walls and is the major component of the outer fibril cell wall layer (Li et al., 1995Go). Recent evidence suggests that PMEs play an important role in pollen tube growth. Although there was no apparent phenotype in pollen grains, mutations in the Arabidopsis Vanguard1 gene, which encodes a pollen-specific PME, caused profound retardation of in vivo growth and structurally unstable pollen tubes when grown in vitro (Jiang et al., 2005Go). YFP tagging of VANGUARD1 demonstrated that the fusion protein was distributed throughout the pollen tube and was localized to the cell wall (Jiang et al., 2005Go). Similarly, tobacco plants in which the tobacco pollen-specific PME (NtPPME1) gene was silenced using an RNAi construct showed significantly retarded pollen tube growth in vivo (Bosch and Hepler, 2005Go). Mutations in the Arabidopsis pollen-specific PME1 (AtPPME1) gene did not affect the pollen grains themselves, but compromised the morphology and growth rate of pollen tubes (Tian et al., 2006Go). Although our double-mutant phenotype does not resemble those described above, it is possible that RGPs have an indirect effect on pectin biosynthesis in the pollen cell wall. Because pectin is the major component of the pollen grain, cell wall defects in pectin composition could have a severe impact on pollen development similar to that observed in the double mutant. Specific antibodies (for polysaccharides) will be used in the future to determine which polysaccharide, or cell wall component, is affected. Additionally, biochemical analysis of mutant pollen could determine the sugar or cell wall composition.

Detailed analysis of rgp1 rgp2 double-mutant pollen at various stages showed that pollen development was not affected until microspore release. However, aberrant pollen grains with enlarged vacuoles were observed in midmicrospore stage and they appeared to be collapsed at the bicellular stage, suggesting that the mutant was arrested before or during the first mitotic division. Several mutants in male gametophytic development, such as gem-1 (Park et al., 1998Go), stud (Hulskamp et al., 1997Go), and tio (Oh et al., 2005Go), have previously been described to have defects in cytokinesis. The gem-1 mutant produces a substantial proportion of microspores that either fail to establish a cell plate at pollen mitosis I (PMI) or produce partial and/or irregularly branching cell walls that alter division symmetry (Park et al., 1998Go; Park and Twell, 2001Go). TIO encodes a FUSED (FU) Ser-Thr protein kinase ortholog in Arabidopsis and tio mutants produce binucleate pollen grains that arise from the failure of cytokinesis at PMI (Oh et al., 2005Go). One of the characteristics of tio microspores is the incomplete formation of callose walls. In addition, Arabidopsis STUD/TETRASPORE/NACK2 encodes a kinesin-like protein; the corresponding mutants fail to undergo cytokinesis after meiosis and large multinucleate pollen grains are formed with aberrant cell walls (Hulskamp et al., 1997Go; Yang et al., 2003Go; Tanaka et al., 2004Go). The phenotypes of the tio and stud mutants underscore the importance of phosphorylation during cytokinesis in male gametophytes. Regulation of self glycosylation by RGPs may also be mediated by phosphorylation (Testasecca et al., 2004Go; G. Drakakaki and N. Raikhel, unpublished data). Because our double mutant is arrested during PMI, it is tempting to speculate that RGPs are involved in cell division and perhaps cell plate formation. Considering that RGPs are implicated in polysaccharide biosynthesis, cell plate formation during the first mitotic division could be affected. A more detailed analysis of rpg1 rpg2 mutant pollen in the early bicellular stage will enable us to explore this hypothesis.

Recent studies have shown that proteins involved in sugar nucleotide metabolism are important during male gametophyte development. The presence of a mutation in a UDP-sugar pyrophosphorylase (usp-2) resulted in lack of a pectocellulosic intine and degeneration of cytoplasm in mutant pollen (Schnurr et al., 2006Go). Like the usp mutant, our rgp1 rgp2 double mutant produced pollen that did not remain turgid; the abnormally enlarged vacuoles in this pollen collapsed during the early bicellular stage. The common feature of the two mutants may be defects in cell wall biosynthesis. In the usp mutant, the phenotype may arise because of defects in the sugar nucleotide supply. In the case of the rgp1 rgp2 double mutant, pollen defects may have arisen because of an inability to convert sugar nucleotides into either wall polysaccharides or other carbohydrates. If correct, this interpretation supports the hypothesis that RGPs are involved in cell wall polysaccharide biosynthesis and could explain the loss of turgor and the defects in the intine cell wall layer in mutant pollen.

Several monosaccharide transporters are also expressed at various stages of pollen maturation (Truernit et al., 1996Go, 1999Go; Schneidereit et al., 2003Go, 2005Go; Scholz-Starke et al., 2003Go). In particular, AtSTP2 transporter expression is confined to the early stages of pollen development. Both AtSTP2 mRNA and protein are first observed at the beginning of callose degradation and microspore release from the tetrads and are no longer detected after the first mitotic division and the formation of the trinuclear gametophyte (Truernit et al., 1999Go). Our rgp1 rgp2 double mutant arrested before or during the first mitotic division. It is therefore tempting to speculate that the loss of RGP1 and RGP2 prevented the downstream utilization of sugars derived from callose degradation. RGP1 and RGP2 may be involved in specific product formation (e.g. polysaccharide) that is necessary for cell/pollen viability. Alternatively, RGP1 and RGP2 may be responsible for maintaining and regulating a supply of sugars for wall biosynthesis that are important during this essential stage in plant reproduction.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
LITERATURE CITED
 

Plant Material

All Arabidopsis (Arabidopsis thaliana) plants used in this study were in the Columbia background. Plants were grown under standard long-day conditions at 22°C under a 16-h light/8-h dark cycle.


In Vitro Pollen Germination

Pollen was germinated on agar as previously described by Hicks et al. (2004)Go. Germination medium was applied to the surface of a microscope slide and allowed to solidify. Pollen from individual flowers was then dabbed onto the agar and further incubated 3 to 8 h at 100% humidity.


Preparation of Constructs

Transgenic Arabidopsis plants expressing YFP-tagged RGP1 and RGP2 from its native regulatory sequences were generated employing the fluorescence tagging of full-length proteins technique (Tian et al., 2004Go). First, AtRGP1 and AtRGP2 were amplified from genomic DNA as two fragments using two sets of primers: RGP1-P1, 5'-gctcgatccacctaggctttgccttgttgggtttatcctt-3', RGP1-P2, 5'-cacagctccacctccacctccaggccggccccaagcttcaatccaagtgac-3', RGP1-P3, 5'-tgctggtgctgctgcggccgctggggccgatgagcttaacccacccacta-3', RGP1-P4, 5'-cgtagcgagaccacaggatttgatttcggctcatttctca-3', RGP2-P1, 5'-gctcgatccacctaggctgccaaagctaacctctggtcat-3', RGP2-P2, 5'-cacagctccacctccacctccaggccggccctcatcccaagcttcaatccat-3', RGP2-P3, 5'-tgctggtgctgctgcggccgctggggcccttaacccaccagcagccagt-3', and RGP2-P4, 5'-cgtagcgagaccacaggacggactattttcccccatacct-3'.

For the second PCR reaction, a pair of gene nonspecific Gateway primers was used and triple template PCR was performed to produce the full-length RGP1 and RGP2 with YFP coding sequence inserted into their last exon approximately 30 bp upstream of the stop codon (Tian et al., 2004Go). The resulting DNA fragments were recombined into pDONR207 (Invitrogen), verified by DNA sequencing, and transferred by Gateway recombination into the binary destination vectors pBIN-GW and pMN-GW (Tian et al., 2004Go).

For the His-T7-AtRGP1, the EcoRI-NotI fragment of AtRGP1 cDNA (accession no. AF013627) was cloned into the EcoRI-NotI sites of the N-terminal His-T7-tag vector pET28c (Novagen). The resulting His-T7-AtRGP1 fusion was excised with XbaI-XhoI digestion and religated into the XbaI-SalI sites of a pCAMBIA 1300 modified vector.

Similarly, the NaeI-SalI fragment of AtRGP2 cDNA (accession no. AF013628) was cloned into the SmaI-XhoI sites of the N-terminal HA-tag vector pACT2 (Novagen). The resulting HA-AtRGP2 fusion was excised with BglII digestion and religated to a pCAMBIA 3300 modified vector.


Plant Transformation

Wild-type seedlings were transformed by an infiltration method using Agrobacterium tumefaciens strain GV3101 carrying the appropriate constructs (Clough and Bent, 1998Go). Five to ten independent lines were analyzed. The lines were amplified and the T3 plants were used for further experiments. The double-tagged T7-RGP1 x HARGP2 were obtained by genetic crosses of single-tagged homozygous lines.


Genomic DNA Extraction and Isolation of T-DNA Insertion lines

Seedlings of T-DNA insertion lines Salk _132152 (Alonso et al., 2003Go) and Gabi Kat_ 652F12 (Li et al., 2003Go) were grown individually. One fully expanded leaf of a 4-week-old plant was used to extract genomic DNA. We used two RGP1 gene-specific primers: forward (RGP1F), 5'-tcacttcctcaccataacaagg-3' and reverse (RGP1R), 5'-ttcacggtatgggtcgtaca-3', and the T-DNA primer, Gabi T-DNA, 5'-cccatttggacgtgaatgtagacac-3'. For RGP2, two gene-specific primers—forward (RGP2F), 5'-agcaacccttttgttaacctga-3' and reverse (RGP2R), 5'-aatatcgaccaaatgtgaaaagc-3' and the LBb, 5'-gcgtggaccgcttgctgcaact-3'—were used on genomic DNA.


RT-PCR Analysis

Total RNA was extracted from rosette leaves or Arabidopsis seedlings using the RNeasy plant mini kit (Qiagen) and treated with DNAse (Roche). RT-PCR was carried out using 500 ng total RNA with the One-Step RT-PCR kit (Qiagen) following the manufacturer's protocol. Primers used for RGP1 were forward (RGP1 RT-F), 5'-ctctccaaattctcttctctct-3' and reverse (RGP1 RT-R), 5'-caaaacggagaagatagatgatat-3', to generate a 1,176-bp product; for RGP2, forward (RGP2 RT-F), 5'-ataccctttcagtctccgtg-3' and reverse (RGP2 RT-R), 5'-gctcatactgctctcaagcttttgccac-3' to generate a 620-bp product. Ubiquitin UBQ-10 forward, 5'-gatctttgccggaaaacaattggaggatggt-3' and reverse 5'-cgacttgtcattagaaagaaagagataacagg-3' primers were used as control.


Immunoprecipitation

Whole 4-week-old plants were homogenized in extraction buffer (40 mM HEPES-KOH, pH 8; 0.45 M Suc; 1 mM EDTA; 1 mM MgCl2; 1 mM dithiothreitol) and protease inhibitors as complete mixture (Roche). Soluble extracts were separated by centrifugation at 100,000g for 40 min on a Ti70 rotor (Beckman) and further enriched by ammonium sulfate precipitation (30%–50%). Proteins were desalted using a PD10 (Amersham-Pharmacia Biotech) desalting column and immunoprecipitated using the column mode of a T7 tag affinity purification kit (Novagen) according to manufacturer's instructions. Control columns were prepared with unlinked agarose beads. Eluates from different fractions were separated on SDS-PAGE, followed by immunoblotting using various antibodies.


Genetic Crosses

To determine gametophytic transmission of the rgp1 and rgp2 insertions, reciprocal crosses were performed between wild type and plants with only one functional copy of RGP2 (rgp1/rgp1 RGP2/rgp2). Approximately 30 independent reciprocal crosses were done. Seeds were harvested and pooled for (rgp1/rgp1 RGP2/rgp2) male or female. After sterilization and germination on plates, DNA was extracted at the appearance of the fourth true leaf. Progeny were genotyped for the presence of rgp1 and rgp2 T-DNA insertions by PCR. X2 and other values are indicated. P values were calculated using GraphPad Quickcals (available online at http://www.GraphPad.com).


Light Microscopy

Plant tissues were fixed in formaldehyde acetic acid containing 37% (w/v) formaldehyde, glacial acetic acid, and 70% (v/v) ethanol (5:5:90; [v/v/v]), overnight at 4°C. Samples were then dehydrated through an ethanol series of up to 100% ethanol. The ethanol was replaced gradually by Technovit 7100 embedding solution (EMS). After polymerization, 3-µm sections were cut, mounted on glass slides, and stained with 0.5% toluidine blue for visualization on a Nikon Microphot FXA microscope. Anthers from stage 13 were squashed to release pollen and mounted in 50% (v/v) glycerol and viewed on a Nikon microscope.


TEM and SEM

Stamens for TEM were taken from different flower stages and fixed with 2% glutaraldehyde in 50 mM phosphate buffer, pH 7.4, overnight at 4°C. Samples were postfixed in 1% osmium tetroxide for 6 h on ice, dehydrated in graduated ethanol series, embedded in Spurr's resin, and polymerized at 50°C for approximately 72 h. Ultrathin sections (60–70 nm) were cut with a diamond knife and mounted on Formvar-coated copper grids. The sections were stained with lead citrate and uranyl acetate (Reynolds, 1963Go) and viewed using a FEI Tecnai 12-electron microscope (FEI).

Cryosections of Arabidopsis roots were used for all immunogold labeling experiments as described in Sanderfoot et al. (1998)Go. Controls were performed with the use of 2% bovine serum albumin substituted for the antisera. Many sections from independent plants were observed for each combination of antibodies. In all cases, the antisera demonstrated high specificity of the labeling.

For SEM, released pollen grains were mounted on stubs over double-sided tape and sputter coated with gold particles (Cressington 108 Auto). Specimens were examined with a scanning electron microscope (XL30 FEG SEM 30 ESEM; Philips) at an accelerating voltage of 10 kV.


CLSM

Live seedlings and plant tissue samples were mounted in water between 1.5 cover glasses, using silicon vacuum grease to create spaces between glass surfaces. Images were collected using Leica TCS SP2/UV. A 514-nm laser line from an argon ion was used to excite YFP. Released pollen grains were stained directly on slides with 0.1% aniline blue for callose, 0.01% calcofluor white for cellulose (as described by Nishikawa et al., 2005Go), and observed under UV light excitation. Similarly, FDA (Heslop-Harrison and Heslop-Harrison, 1970Go) was used to stain released pollen directly on slides viewed under standard fluorescein isothiocyanate conditions under CLSM (Leica TCS SP2/UV). In all cases, water immersion objectives 20x or 63x were employed. The optimal pinhole diameter was set at 1 Airy unit in all cases.

Material and methods used for supplementary data are described in the supplemental data.


Supplemental Data

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

Supplemental Materials and Methods S1.
Supplemental Figure S1. RGP1 expression in pollen grains.
Supplemental Figure S2. Subcellular fractionation of plants expressing RGP1-YFP and RGP2-YFP.
Supplemental Figure S3. Pollen expressing RGP1-YFP and RGP2-YFP are BFA sensitive.
Supplemental Figure S4. Fluorescent image of DAPI-stained tricellular pollen.
Supplemental Table S1. Supplemental summary of the lethality of the rgp1-1 and rgp2-1 and mutant progeny.
Supplemental Video S1. Z series of Arabidopsis leaf cells expressing RGP2-YFP.
Supplemental Video S2. Time series of germinating pollen expressing RGP1-YFP.


   ACKNOWLEDGMENTS
 
We thank Kanwarpal Dhugga for kindly providing the anti-pea RGP antibody. We also wish to thank Valentina Kovaleva for skillful immunolocalization; Marci Surpin for critical reading of this manuscript; Glenn Hicks for advice and useful comments on pollen confocal microscopy; Thomas Girke for advice on phylogenetic analysis; Elizabeth Lord and Juan Dong for useful suggestions; Jocelyn Brimo for assistance with graphic arts during submission of this manuscript; and Karen Bird for editorial assistance.

Received July 5, 2006; accepted October 17, 2006; published October 27, 2006.


   FOOTNOTES
 
1 This work was supported by the National Science Foundation (plant genome grant no. DBI–0211797). Back

2 Present address: Identigene, Inc., 5615 Kirby Dr., Ste. 800, Houston, TX 77005–2451. Back

The author responsible for distribution of materials integral to the findings presented in this article in accordance with the policy described in the Instructions for Authors (www.plantphysiol.org) is: Natasha Raikhel (nraikhel{at}ucr.edu).

[W] The online version of this article contains Web-only data. Back

www.plantphysiol.org/cgi/doi/10.1104/pp.106.086363

* Corresponding author; e-mail nraikhel{at}ucr.edu; fax 951–827–2155.


   LITERATURE CITED
TOP
 ABSTRACT
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 LITERATURE CITED
 
Alonso JM, Stepanova AN, Leisse TJ, Kim CJ, Chen HM, Shinn P, Stevenson DK, Zimmerman J, Barajas P, Cheuk R, et al (2003) Genome-wide insertional mutagenesis of Arabidopsis thaliana. Science 301: 653–657[Abstract/Free Full Text]

Becker JD, Boavida LC, Carneiro J, Haury M, Feijo JA (2003) Transcriptional profiling of Arabidopsis tissues reveals the unique characteristics of the pollen transcriptome. Plant Physiol 133: 713–725[Abstract/Free Full Text]

Blanc G, Barakat A, Guyot R, Cooke R, Delseny I (2000) Extensive duplication and reshuffling in the Arabidopsis genome. Plant Cell 12: 1093–1101[Abstract/Free Full Text]

Bocca SN, Kissen R, Rojas-Beltran JA, Noel F, Gebhardt C, Moreno S, du Jardin P, Tandecarz JS (1999) Molecular cloning and characterization of the enzyme UDP-glucose: protein transglucosylase from potato. Plant Physiol Biochem 37: 809–819[CrossRef][ISI][Medline]

Bocca SN, Rothschild A, Tandecarz JS (1997) Initiation of starch biosynthesis: purification and characterization of UDP-glucose:protein transglucosylase from potato tubers. Plant Physiol Biochem 35: 205–212

Bolte S, Talbot C, Boutte Y, Catrice O, Read ND, Satiat-Jeunemaitre B (2004) FM-dyes as experimental probes for dissecting vesicle trafficking in living plant cells. J Microsc (Oxf) 214: 159–173[CrossRef]

Bosch M, Cheung AY, Hepler PK (2005) Pectin methylesterase, a regulator of pollen tube growth. Plant Physiol 138: 1334–1346[Abstract/Free Full Text]

Bosch M, Hepler PK (2005) Pectin methylesterases and pectin dynamics in pollen tubes. Plant Cell 17: 3219–3226[Free Full Text]

Clough SJ, Bent AF (1998) Floral dip: a simplified method for Agrobacterium-mediated transformation of Arabidopsis thaliana. Plant J 16: 735–743[CrossRef][ISI][Medline]

Dai S, Li L, Chen T, Chong K, Xue Y, Wang T (2006) Proteomic analyses of Oryza sativa mature pollen reveal novel proteins associated with pollen germination and tube growth. Proteomics 6: 2504–2529[CrossRef][ISI][Medline]

Delgado IJ, Wang ZH, de Rocher A, Keegstra K, Raikhel NV (1998) Cloning and characterization of AtRGP1—a reversibly autoglycosylated Arabidopsis protein implicated in cell wall biosynthesis. Plant Physiol 116: 1339–1349[Abstract/Free Full Text]

Dhugga KS, Tiwari SC, Ray PM (1997) A reversibly glycosylated polypeptide (RGP1) possibly involved in plant cell wall synthesis: purification, gene cloning, and trans-Golgi localization. Proc Natl Acad Sci USA 94: 7679–7684[Abstract/Free Full Text]

Dhugga KS, Ulvskov P, Gallagher SR, Ray PM (1991) Plant polypeptides reversibly glycosylated by UDP-glucose—possible components of Golgi beta-glucan synthase in pea cells. J Biol Chem 266: 21977–21984[Abstract/Free Full Text]

Dong XY, Hong ZL, Sivaramakrishnan M, Mahfouz M, Verma DPS (2005) Callose synthase (CalS5) is required for exine formation during microgametogenesis and for pollen viability in Arabidopsis. Plant J 42: 315–328[CrossRef][ISI][Medline]

Enns LC, Kanaoka MM, Torii KU, Comai L, Okada K, Cleland RE (2005) Two callose synthases, GSL1 and GSL5, play an essential and redundant role in plant and pollen development and in fertility. Plant Mol Biol 58: 333–349[CrossRef][ISI][Medline]

Felsenstein J (2005) Using the quantitative genetic threshold model for inferences between and within species. Philos Trans R Soc Lond B Biol Sci 360: 1427–1434[CrossRef][Medline]

Gallardo K, Le Signor C, Vandekerckhove J, Thompson RD, Burstin J (2003) Proteomics of Medicago truncatula seed development establishes the time frame of diverse metabolic processes related to reserve accumulation. Plant Physiol 133: 664–682[Abstract/Free Full Text]

Girke T, Lauricha J, Tran H, Keegstra K, Raikhel N (2004) The cell wall navigator database: a systems-based approach to organism-unrestricted mining of protein families involved in cell wall metabolism. Plant Physiol 136: 3003–3008[Free Full Text]

Goubet F, Misrahi A, Park SK, Zhang ZN, Twell D, Dupree P (2003) AtCSLA7, a cellulose synthase-like putative glycosyltransferase, is important for pollen tube growth and embryogenesis in Arabidopsis. Plant Physiol 131: 547–557[Abstract/Free Full Text]

Heslop-Harrison J, Heslop-Harrison Y (1970) Evaluation of pollen viability by enzymatically induced fluorescence; intracellular hydrolysis of fluorescein diacetate. Stain Technol 45: 115–120[ISI][Medline]

Hicks GR, Rojo E, Hong SH, Carter DG, Raikhel NV (2004) Geminating pollen has tubular vacuoles, displays highly dynamic vacuole biogenesis, and requires VACUOLESS1 for proper function. Plant Physiol 134: 1227–1239[Abstract/Free Full Text]

Holmes-Davis R, Tanaka CK, Vensel WH, Hurkman WJ, McCormick S (2005) Proteome mapping of mature pollen of Arabidopsis thaliana. Proteomics 5: 4864–4884[CrossRef][ISI][Medline]

Honys D, Twell D (2003) Comparative analysis of the Arabidopsis pollen transcriptome. Plant Physiol 132: 640–652[Abstract/Free Full Text]

Hulskamp M, Parekh NS, Grini P, Schneitz K, Zimmermann I, Lolle SJ, Pruitt RE (1997) The STUD gene is required for male-specific cytokinesis after telophase II of meiosis in Arabidopsis thaliana. Dev Biol 187: 114–124[CrossRef][ISI][Medline]

Jiang LX, Yang SL, Xie LF, Puah CS, Zhang XQ, Yang WC, Sundaresan V, Ye D (2005) VANGUARD1 encodes a pectin methylesterase that enhances pollen tube growth in the Arabidopsis style and transmitting tract. Plant Cell 17: 584–596[Abstract/Free Full Text]

Langeveld SMJ, Vennik M, Kottenhagen M, van Wijk R, Buijk A, Kijne JW, de Pater S (2002) Glucosylation activity and complex formation of two classes of reversibly glycosylated polypeptides. Plant Physiol 129: 278–289[Abstract/Free Full Text]

Li Y, Rosso MG, Strizhov N, Viehoever P, Weisshaar B (2003) GABI-Kat SimpleSearch: a flanking sequence tag (FST) database for the identification of T-DNA insertion mutants in Arabidopsis thaliana. Bioinformatics 19: 1441–1442[Abstract/Free Full Text]

Li YQ, Faleri C, Geitmann A, Zhang HQ, Cresti M (1995) Immunogold localization of arabinogalactan proteins, unesterified and esterified pectins in pollen grains and pollen tubes of Nicotiana-tabacum-L. Protoplasma 189: 26–36[CrossRef]