SEARCH

SEARCH BY CITATION

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

(1[RIGHTWARDS ARROW]3)-β-d-Glucans are major components of the cell walls of Oomycetes and as such they play an essential role in the morphogenesis and growth of these microorganisms. Despite the biological importance of (1[RIGHTWARDS ARROW]3)-β-d-glucans, their mechanisms of biosynthesis are poorly understood. Previous studies on (1[RIGHTWARDS ARROW]3)-β-d-glucan synthases from Saprolegnia monoica have shown that three protein bands of an apparent molecular weight of 34, 48 and 50 kDa co-purify with enzyme activity. However, none of the corresponding proteins have been identified. Here we have identified, purified, sequenced and characterized a protein from the 34 kDa band and clearly shown that it has all the biochemical properties of proteins from the annexin family. In addition, we have unequivocally demonstrated that the purified protein is an activator of (1[RIGHTWARDS ARROW]3)-β-d-glucan synthase. This represents a new type of function for proteins belonging to the annexin family. Two other proteins from the 48 and 50 kDa bands were identified as ATP synthase subunits, which most likely arise from contaminations by mitochondria during membrane preparation. The results, which are discussed in relation with the possible regulation mechanisms of (1[RIGHTWARDS ARROW]3)-β-d-glucan synthases, represent a first step towards a better understanding of cell wall polysaccharide biosynthesis in Oomycetes.


Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

Oomycetes are ubiquitous microorganisms that have long been considered as a separate class within the kingdom Fungi. However, sequence comparisons and analyses of their phenotypic characteristics indicate that they are closer to brown algae and taxonomically unrelated to true fungi (Kumar and Rzhetsky, 1996; Paquin et al., 1997; Baldauf et al., 2000; Margulis and Schwartz, 2000). They are currently classified in the Stramenopile eukaryotic kingdom, which includes heterokont algae and water moulds (Margulis and Schwartz, 2000). The Oomycete family comprises plant and animal pathogens that can cause severe environmental and economical damages, as well as saprophytes that are beneficial to natural ecosystems by contributing to the recycling of nutrients from organic decayed matter (Margulis and Schwartz, 2000). In this family, the order Peronosporales includes the genus Phytophthora, which contains some of the most devastating plant pathogens worldwide. A typical example is the species Phytophthora infestans that is responsible for the potato and tomato late blight (Fry and Goodwin, 1997a,b). Another economically important order in the Oomycete family is the Saprolegniales, i.e. a group of water moulds which are common in aquatic environments both as saprophytes and parasites of fishes and crustaceans (Johnson et al., 2002). Some species in this order are responsible for fish infections called saprolegniosis, which cause damages in natural ecosystems as well as important economic losses for the aquaculture industry. The most severe losses are due to the species Saprolegnia ferax and Saprolegnia parasitica, which infect salmon species and sturgeon eggs in hatcheries (Neish and Hughes, 1980). Saprolegnia monoica is also responsible for losses in sturgeon hatcheries but the economic importance of this species is more limited (Lartseva, 1986). Owing to the recrudescence of saprolegniosis the development of specific and environmentally friendly inhibitors of Oomycete proliferation has become a priority for the aquaculture industry. Cell wall carbohydrate synthesizing enzymes, which play a central role in such vital processes as the morphogenesis and growth of Oomycetes are potential targets for novel specific inhibitors. In addition, Oomycete cell wall carbohydrates have been shown to have a substantive function in the interrelationship between the pathogens and the hosts (Johnson et al., 2002). But before novel inhibitors can be targeted to enzymes responsible for cell wall polysaccharide synthesis, it is crucial to understand the molecular mechanisms in which these enzymes are involved. Indeed, despite its economical and fundamental relevance cell wall biosynthesis in Oomycetes is one of the most important developmental processes that are still poorly understood.

One of the characteristics that distinguish Oomycetes from true fungi is their specific cell wall polysaccharide composition (Bartnicki-Garcia, 1968; Wessels and Sietsma, 1981). The latter consists essentially of (1[RIGHTWARDS ARROW]3)-β-d-glucans, (1[RIGHTWARDS ARROW]6)-β-d-glucans and cellulose whereas chitin, which is a major cell wall component of fungi, has been shown to occur in very small amounts in the walls of Oomycetes (Lin and Aronson, 1970; Aronson and Lin, 1978; Campos-Takaki et al., 1982; Bulone et al., 1992). The enzymes responsible for the synthesis of the major cell wall polysaccharides of Oomycetes, especially (1[RIGHTWARDS ARROW]3)-β-d-glucan and cellulose synthases, have not been identified and characterized yet. These enzymes have proved to be particularly difficult to study using biochemical approaches because of their location in plasma membranes and their inherent instability upon detergent extraction. By analogy with similar enzymes from plants (reviewed in Delmer, 1999) and fungi (Beauvais et al., 2001), they are probably occurring as multiprotein complexes. This evidence comes from previous work on S. monoica in which fractions enriched in glucan synthase activities were obtained after solubilization with the detergent CHAPS (Bulone et al., 1990; Bulone and Fèvre, 1996). In particular, the enriched fractions systematically contained several proteins that co-purify with enzyme activities. However, it is not known whether the various enriched proteins are actually all related to the biosynthesis of (1[RIGHTWARDS ARROW]3)-β-d-glucan and cellulose as none of them has been sequenced and identified. In addition, despite the isolation of a number of genes that have been shown to be involved in cellulose (reviewed in Doblin et al., 2002) and (1[RIGHTWARDS ARROW]3)-β-d-glucan (see for example Douglas et al., 1994; Li et al., 2003) synthesis in other organisms, similar genes in Oomycetes have still to be identified. For instance, if there is strong evidence that the yeast fks genes (Douglas et al., 1994) and their plant homologues (Li et al., 2003) are very likely coding for the catalytic subunits of (1[RIGHTWARDS ARROW]3)-β-d-glucan synthases, there is so far no report describing the isolation and function of similar genes in Oomycete species. This is also true for the genes encoding the catalytic subunits of the cellulose synthase complexes in Oomycetes. Such studies are underway in our laboratory where we are using S. monoica as a model organism to study cell wall polysaccharide biosynthesis and investigate the composition of the glucan synthase complexes. Interestingly, previous work on glucan synthases from S. monoica strongly suggested that this microorganism contains two (1[RIGHTWARDS ARROW]3)-β-d-glucan synthase activities with different biochemical properties (Billon-Grand et al., 1997). One of these activities has an optimal pH of action of 6 and it is inhibited in vitro by divalent cations such as Ca2+, Mn2+ and Mg2+ at a concentration of 2 mM. The other activity is maximal at an alkaline pH of 9 and it is stimulated by 2 mM of the above-mentioned divalent cations. It was shown that either activity can be specifically assayed in the presence of the other (Billon-Grand et al., 1997). Studies on the (1[RIGHTWARDS ARROW]3)-β-d-glucan synthase that has an optimal pH of action of 6 have shown that three protein bands of an apparent molecular weight (MW) of 34, 48 and 50 kDa are particularly enriched in fractions with high specific enzyme activity (Bulone et al., 1990; Bulone and Fèvre, 1996). In the present work we have taken advantage of the latter results to identify, purify and characterize in as much detail as possible a protein that co-purifies with (1[RIGHTWARDS ARROW]3)-β-d-glucan synthase. In particular, we demonstrate that one of the proteins in the 34 kDa band belongs to the annexin family and that it is an activator of the enzyme. Two other proteins from the 48 and 50 kDa bands were also identified.

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

Effect of EGTA on (1[RIGHTWARDS ARROW]3)-β-d -glucan synthase activity

Previous studies have shown that the solubilized (1[RIGHTWARDS ARROW]3)-β-d-glucan synthase from S. monoica can be rapidly enriched and recovered in an active form by glycerol gradient centrifugation and/or product entrapment (Bulone et al., 1990; Bulone and Fèvre, 1996). In addition, a 34 kDa protein was a major component in the enriched fractions, suggesting that it is associated to the enzyme complex and thus plays a role in the catalysis of (1[RIGHTWARDS ARROW]3)-β-d-glucan synthesis (Bulone et al., 1990; Bulone and Fèvre, 1996). However, no information is available to date on the identity and function of this protein. In order to address this question, fractions enriched in (1[RIGHTWARDS ARROW]3)-β-d-glucan synthase were prepared by ultracentrifugation of CHAPS-extracted membrane proteins on a linear glycerol gradient, and the major proteins in the peak of activity were subjected to protein sequencing. Even though density values are given as an estimate with an accuracy of no more than 0.01 units, repetition of experiments showed that protein and enzyme activity distribution in gradients were highly reproducible and shifts as small as one fraction were found to be significant and perfectly reliable. Hence, fraction numbers are used thereafter as the absolute reference for the description of the position of protein and enzyme activity in glycerol gradients. (1[RIGHTWARDS ARROW]3)-β-d-Glucan synthase activity was maximum in fraction 12 of the gradient and sedimented ahead of the peak of protein at a density of ∼1.045, with a purification factor of 14-fold with respect to the detergent extract (Fig. 1A). SDS-PAGE analysis of each fraction indicated that the intensity of 34, 48 and 55 kDa bands correlated with the distribution of the (1[RIGHTWARDS ARROW]3)-β-d-glucan synthase activity in the gradient (Fig. 1B). The 55 kDa band corresponds to the one that was originally detected at an apparent MW of 50 kDa by Bulone et al. (1990). Edman degradation of trypsic peptides from the 48 and 55 kDa bands excised from SDS-PAGE gels revealed that the corresponding proteins are, respectively, β and α subunits of the naturally abundant membrane-bound mitochondrial ATP synthases (Table S1 in Supplementary material). This type of enzymes has also been shown to co-purify with glucan synthases from plant species, certainly as a result of contaminations from mitochondria during membrane preparation (Bulone et al., 1995). Thus, it is unlikely that the 48 and 50 kDa ATP synthase subunits identified here have a role in (1[RIGHTWARDS ARROW]3)-β-d-glucan synthesis. Sequences obtained from the 34 kDa band (GIGTDEYGLSAAIVR and GAGTTEQLLYPVLGGR) showed a significantly high identity (53–62%) and similarity (60–73%) with proteins from various origins belonging to the annexin family (Table S1).

image

Figure 1. Separation of a 34 kDa annexin from the CHAPS-extracted (1[RIGHTWARDS ARROW]3)-β-d-glucan synthase by glycerol gradient centrifugation. Two millilitres of detergent extract were loaded on a linear glycerol gradient and centrifuged at 120 000 g for 3 h 45 min. The gradient presented in A–C was devoid of EGTA whereas D–F correspond to a gradient that contained 10 mM EGTA. A and D, density of the gradient fractions (squares) and levels of glucan synthase activity (circles) and protein content (triangles). B and E, SDS-PAGE analysis (10% acrylamide) of the CHAPS-extract that was loaded on the gradients (B, lane D) and of fractions 5–15 from each gradient. The gels were stained with silver. C and F, Western blot analysis of the fractions shown in B and E, using specific antibodies produced against a purified 34 kDa annexin from S. monoica (see Fig. 2). In B, C, E, F, a total amount of 10 μg protein was loaded in each lane, except for fractions 5–8 for which 3–9 μg was loaded because of lower protein concentrations in these fractions.

Download figure to PowerPoint

Annexins are known to interact in a calcium-dependent manner with plasma membranes from which they can be specifically released by incubating the membranes with calcium-chelating agents like EGTA (Boustead et al., 1989; Burgoyne and Geisow, 1989; Smallwood et al., 1990). In order to further investigate the possible involvement of an annexin in (1[RIGHTWARDS ARROW]3)-β-d-glucan synthase activity, CHAPS-extracted proteins were centrifuged on a glycerol gradient that contained 10 mM EGTA (Fig. 1D–F). In these conditions, the 34 kDa band was distributed over fractions 8–15, but its intensity was much higher in the low density fractions (fractions 14–15, density ∼1.01–1.02, Fig. 1E). This differs from the results obtained in the absence of EGTA in which the highest intensity of the 34 kDa band was observed in higher density fractions (fractions 10–13, density ∼1.03–1.06, Fig. 1B). This observation indicates that the addition of EGTA in the gradient provokes a shift of some proteins that are present in the 34 kDa band towards low density fractions. One of the proteins that were systematically shifted was subsequently identified as a 34 kDa annexin, by combining protein purification, sequencing and antibody production (see corresponding results in the following sections). The shift of this 34 kDa annexin was evidenced by Western blot analyses of each fraction of the gradients using the specific anti-annexin antibodies prepared as indicated in the next sections. In particular, the use of the antibodies revealed that the 34 kDa annexin is essentially present in fractions 8–12 in the gradient devoid of EGTA (Fig. 1C), whereas the protein is detected exclusively in fractions 13–15 in the gradient that contains EGTA (Fig. 1F).

Interestingly, when the position of the 34 kDa annexin in the gradient was shifted in the presence of EGTA the peak of (1[RIGHTWARDS ARROW]3)-β-d-glucan synthase activity was also slightly shifted towards low density fractions (maximum of activity in fraction 13 in the presence of EGTA as opposed to fraction 12 in the absence of chelator). In addition, depleting fractions with the highest activity in one (or several) 34 kDa protein – possibly the 34 kDa annexin – provokes a decrease in (1[RIGHTWARDS ARROW]3)-β-d-glucan synthase activity. This is illustrated by the situation in fraction 11 (density ∼1.05) in which the activity drops by 63% when EGTA is present in the gradient (Fig. 1A and D). Conversely, a higher activity is measured when the amount of one or several of the 34 kDa proteins is increased in a fraction that originally contains a rather low glucan synthase activity. It is the case for instance of activity in fractions 14 and 15, which significantly increases when EGTA is added in the gradient (Fig. 1A and D). Altogether, these results suggest that at least one 34 kDa protein activates the (1[RIGHTWARDS ARROW]3)-β-d-glucan synthase from S. monoica. In the next part of this work, we have purified the 34 kDa annexin as a preliminary step for further investigations on the possible involvement of this protein as a (1[RIGHTWARDS ARROW]3)-β-d-glucan synthase activator.

Purification of the 34 kDa annexin

Incubation of microsomal fractions in the presence of 10 mM EGTA provoked the release and enrichment of a 34 kDa band in the supernatant (SEGTA) recovered after centrifugation of the preparation at 150 000 g (Fig. 2A). No enrichment of 34 kDa proteins was observed in the control performed in the same conditions, but in the absence of EGTA (fraction SC, Fig. 2A), strongly suggesting that proteins in the 34 kDa band interact with the membranes in a calcium-dependent manner. Although the 34 kDa band was clearly enriched in the SEGTA supernatant, SDS-PAGE analysis of this fraction revealed that it contained numerous other proteins distributed over a wide range of MW (Fig. 2A). In order to further purify the 34 kDa proteins released by EGTA, fraction SEGTA was subjected to a calcium precipitation step as described in Experimental procedures. The resulting pellet was resuspended in the presence of an excess of EGTA (20 mM), which allowed the specific release of only one band of 34 kDa in a supernatant designated S1 (Fig. 2A). The pellet corresponding to supernatant S1 was further incubated in the presence of 20 mM EGTA in an attempt to sequentially extract more of the 34 kDa proteins. This additional extraction step followed by centrifugation yielded a new supernatant designated S2. Even though it was indeed possible to further extract 34 kDa proteins from the pellet corresponding to supernatant S1, supernatant S2 contained several additional bands of a rather weak intensity (Fig. 2A). Further, 2D-PAGE analysis of fraction S2 indicated that the 34 kDa band consisted of a major spot as well as of at least nine other minor proteins that could be resolved as single components only by 2D-PAGE (Fig. 2B). A single spot was observed when overloading a 2D-gel with fraction S1 and letting the developing reaction of the silver staining protocol proceed for 30 min instead of the usual 10–15 min (Fig. 2C). Based on the pattern obtained by combining the high-resolution 2D-PAGE technique with the very sensitive silver staining protocol, fraction S1 was considered as containing only one protein. Several of the sequences obtained from the corresponding spot were overlapping, making possible the reconstitution of a peptide of 35 amino acids, which showed significantly high similarities with proteins from the annexin family (Fig. 3). Sequencing of the major spot of 34 kDa in fraction S2 (Fig. 2B) indicated that this protein was identical to the one in fraction S1 (data not shown).

image

Figure 2. Purification of the 34 kDa annexin from S. monoica. All the gels presented were stained with silver. A. SDS-PAGE analysis (12.5% acrylamide) of the fractions recovered at different purification steps (theoretical amount of 3 μg total protein per lane). Mb, microsomal fraction; SEGTA, supernatant obtained after incubation of the microsomal fraction in the presence of Tris-KOH buffer containing 10 mM EGTA and centrifugation; Sc was obtained as SEGTA but in the absence of EGTA; S1, supernatant recovered after precipitation of the calcium-binding proteins in SEGTA with 15 mM CaCl2 and re-solubilization of the pelleted proteins in 20 mM EGTA; the proteins that remained insoluble in 20 mM EGTA were subjected to a second solubilization step in 20 mM EGTA to yield supernatant S2. B. 2D-PAGE analysis of fraction S2 (100 μg total protein loaded; 12.5% acrylamide for the second dimension). C. 2D-PAGE analysis of fraction S1. The gel was overloaded (100 μg protein for only one spot) and stained for a longer time than usual to evaluate the purity of the 34 kDa protein.

Download figure to PowerPoint

image

Figure 3. Overlapping amino acid sequences obtained from the protein spot in Fig. 2C after partial proteolysis with trypsin and V8 protease (endoproteinase Glu-C). The overlapping sequences allowed the reconstitution of a 35-amino-acid peptide (Sm) that can be aligned with putative annexin sequences from Ustilago maydis (Um, Accession No. Q4P8I3) and Tetraodon nigroviridis (Tn, Accession No. Q4T4P0) and with a sequence from an annexin from Brachydanio rerio (Br, annexin 1, Accession No. Q804H1). Strictly conserved amino acids are shaded.

Download figure to PowerPoint

From the results presented above, it was concluded that fraction S1 contained a purified annexin and was thus suitable for further functional and molecular characterization of the protein. Typically, a total amount of 120 μg purified annexin can be obtained in fraction S1 from 25 g of freshly harvested mycelium. Acrylamide pieces containing the 34 kDa annexin were excised from 2D-gels loaded with fraction S1 and used as a source of antigen for antibody production in rabbits. The polyclonal antibodies recognized a single band at 34 kDa in CHAPS-extracted membrane proteins separated by SDS-PAGE, and the major 34 kDa spot in a glycerol gradient fraction enriched in (1[RIGHTWARDS ARROW]3)-β-d-glucan synthase activity (data not shown). The latter spot was identified as the 34 kDa annexin by protein sequencing (not shown). These results show that the antibodies can be used to firmly demonstrate that the sequenced 34 kDa annexin is systematically shifted towards low density fractions in glycerol gradients that contain EGTA, as exemplified in Fig. 1C and F.

Effect of the purified annexin on (1[RIGHTWARDS ARROW]3)-β-d -glucan synthase activity

Immunoprecipitation experiments have previously provided strong evidence that the protein we have now identified as a 34 kDa annexin is part of the (1[RIGHTWARDS ARROW]3)-β-d-glucan synthase complex (Bulone and Fèvre, 1996). However, the new antibodies prepared in the present work against the denatured annexin were not able to immunoprecipitate the native form of the protein (data not shown). As a consequence, it was not possible to co-immunoprecipitate the annexin and (1[RIGHTWARDS ARROW]3)-β-d-glucan synthase activity or to use the antibodies to physically separate in native conditions the annexin from the catalytic subunit of the enzyme. As such experiments could not be performed, another approach based on add-back experiments was implemented to investigate the effect of the purified annexin on (1[RIGHTWARDS ARROW]3)-β-d-glucan synthase activity. In this approach fractions from glycerol gradients prepared in the presence or in the absence of EGTA were supplemented with increasing amounts of the purified annexin (fraction S1) and enzyme activity was measured in the resulting mixtures (Fig. 4A). Enzyme assays systematically showed that the purified annexin fraction used for these add-back experiments did not contain detectable (1[RIGHTWARDS ARROW]3)-β-d-glucan synthase activity (data not shown). Thus, fraction S1 could be used as a source of annexin for testing the effect of the purified protein on (1[RIGHTWARDS ARROW]3)-β-d-glucan synthase activity. In a first set of experiments, fraction 11 from the gradient that contained EGTA was used as a source of enzyme because it did not contain any annexin, as illustrated in Fig. 1F. Addition of annexin in this gradient fraction raised the level of (1[RIGHTWARDS ARROW]3)-β-d-glucan synthase activity in a dose-dependent manner by up to 100% compared with the control performed in the same conditions but in the absence of annexin (Fig. 4A, +EGTA). Amounts of added annexin as low as 0.1 μg were sufficient to provoke a clear activation of (1[RIGHTWARDS ARROW]3)-β-d-glucan synthase. Interestingly, the dose-dependent activating effect was reproducibly observed for added amounts of purified annexin of up to 0.6 μg while addition of higher amounts of annexin, e.g. 1.2 μg, did not provoke any significant further increase of activity (Fig. 4A). No activation was observed in the presence of boiled annexin. Thus, it can be concluded that the higher activity measured after addition of fraction S1 in the gradient fraction is due to the 34 kDa annexin. In addition, these results suggest that the activation of glucan synthase is certainly dependent on the annexin/enzyme ratio because the addition of annexin above a certain optimal ratio does not lead to a further increase of activity.

image

Figure 4. Effect of the purified annexin on (1[RIGHTWARDS ARROW]3)-β-d-glucan synthase activity. The purified annexin fraction (fraction S1, Fig. 2) was first shown to be devoid of any (1[RIGHTWARDS ARROW]3)-β-d-glucan synthase activity. A. The purified annexin was added to three different enzyme preparations: a detergent extract of a microsomal fraction (D), fraction 11 of the glycerol gradient devoid of EGTA (–EGTA, Fig. 1A–C) and fraction 11 of the gradient centrifuged in the presence of 10 mM EGTA (+EGTA, Fig. 1D–F). After 20 min incubation at room temperature (1[RIGHTWARDS ARROW]3)-β-d-glucan synthase activity was assayed in each mixture. Light grey, dark grey, dotted and black boxes, addition of 0.1, 0.3, 0.6 and 1.2 μg of annexin respectively; white boxes, controls performed in the absence of added annexin; striped boxes, addition of 1.2 μg of a boiled (100°C, 10 min) annexin fraction. B. The levels of (1[RIGHTWARDS ARROW]3)-β-d-glucan synthase activity measured after 20 min incubation at room temperature of mixtures consisting of a fraction devoid of annexin [fraction 11 of the gradient containing EGTA (Fig. 1D–F)], 0.6 μg of annexin and 10 μM−5 mM calcium. The activity measured in the absence of calcium (not shown) was identical to the one measured in the presence of 10 μM calcium. Standard deviations from three experiments performed in the same conditions are shown in A and B.

Download figure to PowerPoint

Annexins are well known to be calcium-binding proteins. Hence, add-back experiments were repeated with 0.6 μg of purified annexin in the presence of increasing concentrations of calcium in order to investigate the possible effect of the cation on the activating effect of the annexin (Fig. 4B). The simultaneous addition of annexin and calcium in the range 10–400 μM did not further stimulate (1[RIGHTWARDS ARROW]3)-β-d-glucan synthase activity, suggesting that the observed activating effect is not calcium-dependent. The use of higher concentrations of calcium provoked an inhibition of enzyme activity (Fig. 4B), consistent with the previously reported inhibitory effect of the cation in the range 500 μM−2 mM on the (1[RIGHTWARDS ARROW]3)-β-d-glucan synthase activity assayed at pH 6 (Billon-Grand et al., 1997).

It is noteworthy that the level of activity observed during the add-back experiments never reached the one measured in fraction 11 from the gradient that contained no EGTA (Fig. 4A). This suggests that reconstitution of a fully active enzyme requires other factors in addition to the 34 kDa annexin. It is possible that another compound that also activates (1[RIGHTWARDS ARROW]3)-β-d-glucan synthase is originally present in the detergent extract and separated from the enzyme complex during centrifugation on a glycerol gradient that contains EGTA.

Add-back experiments were also performed on fraction 11 from the glycerol gradient that contained no EGTA as well as on a CHAPS extract of membrane-bound proteins (Fig. 4A, –EGTA and D). These two fractions contained the 34 kDa annexin, as shown in Fig. 1C (lanes D and 11). The addition of 0.6 and 1.2 μg of purified annexin to these fractions did not significantly increase the level of (1[RIGHTWARDS ARROW]3)-β-d-glucan synthase activity, compared with the controls performed in the same conditions but in the absence of added annexin or in the presence of boiled annexin (Fig. 4A). These results indicate that the addition of an excess of annexin to the enzyme preparation does not provoke any further increase of activity, which is consistent with the observation made above that no significant further increase of activity is observed beyond a certain annexin/enzyme ratio.

ATP and GTP are known to affect glucan synthase activities in S. monoica (Fèvre, 1983), but the mode of action of these nucleotides has not been elucidated. Interestingly, the (1[RIGHTWARDS ARROW]3)-β-d-glucan synthase from yeast has been shown to be stimulated by GTP (Shematek and Cabib, 1980), and a GTP-binding protein with GTPase activity has been identified as a regulator of the enzyme activity (Mazur and Baginsky, 1996; Qadota et al., 1996). In addition, ATPase/GTPase activities have also been demonstrated to be associated to several plant annexins (McClung et al., 1994; Calvert et al., 1996; Shin and Brown, 1999). However, our labelling experiments conducted on enriched (fraction S2) and purified (fraction S1) annexin preparations in the presence of [α-32P]-GTP and [α-32P]-ATP clearly demonstrated that the 34 kDa annexin does not bind ATP and GTP, and hence that it is not involved in an ATP/GTP-dependent regulatory mechanism of the (1[RIGHTWARDS ARROW]3)-β-d-glucan synthase activity from S. monoica (data not shown).

Sequence analysis of the 34 kDa annexin

Overlapping partial cDNA sequences were obtained in several steps by a polymerase chain reaction (PCR)-based approach. Peptide sequences from the purified 34 kDa annexin were used to search the Oomycete Phytophthora Functional Genomics database (http://www.pfgd.orgGajendran et al., 2006). As detailed in Experimental procedures, specific primers were prepared for PCR amplification using DNA sequences corresponding to Phytophthora peptides that were nearly identical to those we sequenced from the purified 34 kDa annexin. A first DNA fragment was amplified and its sequence was used to design more primers for subsequent RACE-PCR amplification. Overlapping cDNAs corresponding to the 5′ and 3′ ends were cloned and sequenced until a full-length cDNA sequence could be reconstituted. The deduced amino acid sequence obtained from the full-length cDNA sequence has been submitted to the DDBJ/EMBL/GenBank databases under Accession Number DQ323662 and it is presented in Fig. S1 in Supplementary material. It contains all peptides experimentally sequenced from the purified annexin and corresponds to a protein of 328 amino acids, with a theoretical molecular mass and pI of 36 kDa and 5.41 respectively. The theoretical molecular mass is in good agreement with the one observed on both SDS-PAGE and 2D-PAGE gels, the difference of 1–2 kDa being within the experimental error. The theoretical pI of the protein is 0.4 unit lower than the one estimated from 2D-PAGE gels (Fig. 2B and C). This might be due to the inaccuracy of the experimental determination of the pI value and/or to the partial reliability and accuracy of the commercial linear pH gradients. The overall deduced amino acid sequence gave up to 34% identity and 53% similarity with proteins in sequence databases belonging to the annexin family. A motif scan performed in the PROSITE database (http://www.expasy.org/prosite/) using the complete sequence firmly identified the protein as a member of the annexin family. In particular, by analogy with a large number of annexin sequences from other species, the amino acid sequence can be divided into four repeat domains, each containing a consensus sequence designated endonexin fold (Geisow et al., 1986; Barton et al., 1991) as well as a type-II Ca2+-binding site (Fig. S1). The latter is identified as a GXGT loop with a D or E residue 41 amino acids downstream of the first G (Chen et al., 1993). As for animal annexins, the consensus sequence of the type-II Ca2+-binding site is well conserved in each domain. This contrasts with the situation in plants in which the Ca2+-binding sites are usually well conserved only in domains I and IV, as illustrated in Fig. S1 with the example of an annexin from cotton. In the third domain, the first G of the type-II Ca2+-binding site of the S. monoica annexin is substituted with a K residue, a common feature to some other annexins (Delmer and Potikha, 1997), as exemplified in Fig. S1 with a fungal annexin. In the third domain of the human annexin shown in Fig. S1, as well as for numerous other type-II Ca2+-binding sites encountered in many annexins (Delmer and Potikha, 1997), the first G is substituted by an R residue. The general occurrence in annexins of consensus sequences characteristic of Ca2+-binding sites is consistent with the experimental observation that all known members of the annexin family exhibit a calcium-binding activity, which is generally evidenced through the use of EGTA (see for instance Boustead et al., 1989; Burgoyne and Geisow, 1989; Smallwood et al., 1990). Our results on the 34 kDa protein from S. monoica comply with both the presence of several consensus sequences of Ca2+-binding sites (Fig. S1) and the experimental evidence that the protein is specifically solubilized by EGTA and precipitated by calcium (Fig. 2). Thus, it can be inferred that the 34 kDa protein identified in the present work is indeed a calcium-binding protein. The scores obtained with the ScanProsite tool for searches of post-translational modifications were all very low. In addition, as mentioned above, the apparent MW of the protein is close to the theoretical one. Altogether these results suggest that the protein does not contain any post-translational modification. Alternatively, if some modifications are present they do not affect the apparent molecular mass of the protein, as judged by its mobility on SDS-PAGE and 2D-PAGE gels.

Phylogenetic analysis of the 34 kDa annexin

Complete sequences of annexin genes from various fungal (Aspergillus and Neurospora), slime mould (Dictyostelium discoideum) and plant (Arabidopsis thaliana) species were used for phylogenetic analysis. Oomycete sequences were extracted from the Phytophthora Functional Genomics database (Gajendran et al., 2006). At the time the analysis was performed, the latter database contained EST sequences of different lengths among which only one encompassed the four-domain core region used for phylogenetic reconstruction. To maximize the sequences taken into account, the genetic distances for neighbour-joining were calculated among all pairwise comparisons with the Poisson-corrected proportion of amino acid differences (Fig. S2 in Supplementary material). It was verified that the same general topology was obtained after global gap removal, but on a reduced set of sequences. In all simulations, the annexin from S. monoica grouped with Phytophthora EST sequences and formed a clearly separate Oomycete clade with a good bootstrap support (99%). In keeping with a recent analysis (Khalaj et al., 2004), the annexin from Neurospora crassa grouped with other Ascomycete annexins to form a well separated clade (bootstrap 96%) and it did not segregate with the annexin from D. discoideum as reported earlier by Braun et al. (1998) (Fig. S2). Thus, the ANX14 name assigned by the latter authors to the N. crassa annexin is no longer supported. Because most of the Phytophthora EST sequences are not complete and need to be verified, the orthology between the sequence from S. monoica obtained in the present work and any of the sequences from Phytophthora is not reasonably predictable.

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

In the present work we have characterized a 34 kDa protein that was previously shown to systematically co-purify with (1[RIGHTWARDS ARROW]3)-β-d-glucan synthase activity, regardless of the method used for enzyme purification (Bulone et al., 1990; Bulone and Fèvre, 1996). Sequencing and purification of this protein revealed that it presents the general characteristics of proteins from the annexin family (Geisow et al., 1986; Barton et al., 1991). In particular it contains four so-called endonexin domains of 65–67 amino acids that each bears a Ca2+-binding site. Because the theoretical molecular mass of the protein is 36 kDa, and not 34 kDa as experimentally determined in the present and previous works (Bulone et al., 1990; Bulone and Fèvre, 1996), we propose to call this protein p36. Analysis of the sequence of p36 indicates that the protein likely contains no post-translational modifications, as opposed to many annexins from vertebrates that bear modifications like phosphorylation and myristoylation (Gerke and Moss, 2002 and references therein). Potential post-translational modifications including different types of phosphorylation sites have been reported for some plant annexins, and it has been proposed that these modifications may determine the specific function of the proteins (Shin and Brown, 1999). However, this has never been experimentally demonstrated and there is so far no evidence that such potential sites are actually phosphorylated.

Separation of p36 from the (1[RIGHTWARDS ARROW]3)-β-d-glucan synthase activity in glycerol gradients and add-back experiments showed that p36 is a (1[RIGHTWARDS ARROW]3)-β-d-glucan synthase activator. The (1[RIGHTWARDS ARROW]3)-β-d-glucan synthase assayed in the present work has an optimal pH of 6 and it is inhibited by mM concentrations of calcium. Thus, one may expect that addition in glycerol gradients of mM concentrations of a calcium chelator such as EGTA would provoke an increase of enzyme activity. This effect was not observed, most likely because the calcium concentration in glycerol gradients is much lower than the mM cation concentrations required for inhibition. In addition, the presence of EGTA in glycerol gradients shifts the annexin activator from the peak of activity, which results in an overall decrease of enzyme activity (Fig. 1). Apart from p36 identified in the present work, only one other annexin isolated from cotton fibre and called Anx(Gh1) has been shown to affect a glucan synthase activity (Andrawis et al., 1993). There has been some confusion in the literature on the possible function of Anx(Gh1) within (1[RIGHTWARDS ARROW]3)-β-d-glucan and cellulose synthase complexes. There is strong evidence that these complexes correspond to two different enzymes coded by different genes (see for instance Douglas et al., 1994; Pear et al., 1996; and Li et al., 2003), one enzyme forming (1[RIGHTWARDS ARROW]3)-β-glucosidic bonds while the other catalyses the formation of (1[RIGHTWARDS ARROW]4)-β-glucosidic linkages. Even though it has been proposed that Anx(Gh1) colocalizes with cellulose synthase and plays a role in the dimerization mechanisms of cellulose synthase catalytic subunits (Hofmann et al., 2003; Hofmann, 2004), the only experimental evidence available to date that links a glucan synthase to Anx(Gh1) concerns the cotton (1[RIGHTWARDS ARROW]3)-β-d-glucan synthase (Andrawis et al., 1993) and not cellulose synthase. Interestingly, Anx(Gh1) had the opposite effect as p36 as it inhibited the cotton (1[RIGHTWARDS ARROW]3)-β-d-glucan synthase activity in vitro by up to 15-fold. In addition, as opposed to the glucan synthase studied in the present work, the plant enzyme was stimulated by calcium. Thus, annexins from various species and calcium may have opposite regulatory effects on (1[RIGHTWARDS ARROW]3)-β-d-glucan synthases. It will be necessary to characterize more annexins from organisms that produce (1[RIGHTWARDS ARROW]3)-β-d-glucan as a cell wall component to determine whether the observed effects are species-specific. Another important aspect will be to show if annexins with opposite effects can occur in a given organism and participate in the spatiotemporal regulation of enzyme activities born by different isoforms. It is noteworthy that the inhibitory results obtained with the cotton annexin (Andrawis et al., 1993) were somewhat in contradiction with those obtained later by another group, which showed that a recombinant annexin from the same plant species has no effect on (1[RIGHTWARDS ARROW]3)-β-d-glucan synthase activity (Shin and Brown, 1999). In the latter case though, the protein used for add-back experiments was expressed in Escherichia coli and it is possible that some post-translational modifications are important for the inhibitory function of the corresponding annexin. In addition, the recombinant annexin used by Shin and Brown (1999) may not be the same as the one that was previously purified from cotton fibres and shown to inhibit (1[RIGHTWARDS ARROW]3)-β-d-glucan synthase activity (Andrawis et al., 1993), although this needs to be confirmed. Indeed, alignment of the complete sequences of both proteins shows 98% identity and 99% similarity with only four amino acid substitutions, but these slight differences could be due to sequencing errors. Since a given plant species contains several annexins, it is possible that only some of them are specifically involved in the regulation of (1[RIGHTWARDS ARROW]3)-β-d-glucan synthase activity. Interestingly, searches in the Phytophthora Functional Genomics database (Gajendran et al., 2006) reveal the presence of different putative annexin sequences in Phytophthora species. As for plant species, more annexins from Oomycetes need to be isolated and characterized to study in more detail their specific functions.

Because p36 was unable to bind ATP and GTP (data not shown), it can be inferred that the protein has no ATPase or GTPase activity, as opposed to some animal and plant annexins (McClung et al., 1994; Calvert et al., 1996; Shin and Brown, 1999; Bandorowicz-Pikula et al., 2001). This result is consistent with the fact that sequence analysis of p36 does not show the occurrence of any of the Walker nucleotide-binding consensus motifs (Walker et al., 1982).

It was proposed that the two (1[RIGHTWARDS ARROW]3)-β-d-glucan synthase activities from S. monoica with optimal pH of action of 6 and 9 correspond to two isoforms of the enzyme and that they are located in different areas of the hyphae (Billon-Grand et al., 1997). Their sensitivity to divalent cations would reflect their cellular location, with the alkaline enzyme located in the apical part of the hyphae where calcium concentration is higher, and the other enzyme located in subapical areas that contain lower concentrations of calcium. According to this hypothesis, the alkaline enzyme would be more specifically responsible for cell wall synthesis in the fast-growing areas of the hyphae, and calcium would play an important role in the control of the relative levels of the two differently distributed (1[RIGHTWARDS ARROW]3)-β-d-glucan synthase activities. Interestingly, no stimulation of activity was detected when the same add-back experiments as those described in the present work were performed on the alkaline enzyme (data not shown). Thus, it can be proposed that the stimulatory effect of p36 is specific for the enzyme that has an optimal pH of action of 6. Other yet unidentified factors may also be involved in the regulation of both or either isoforms. Interestingly, previous work showed the existence of a membrane-bound activator of the (1[RIGHTWARDS ARROW]3)-β-d-glucan synthase from S. monoica (Girard and Fèvre, 1991). This effector, which was heat-stable, non-dialysable and resistant to protease treatment was proposed to be a glycoprotein that modulates glucan synthase activity during vesicle trafficking and targeting of the enzyme to the plasma membrane. It is now important to identify all the components of the glucan synthase complexes at each step of maturation and after transfer to the plasma membrane to understand how the enzyme activities are regulated. This will provide further insights on the mechanisms of cell wall polysaccharide biosynthesis and, implicitly, on the processes of morphogenesis and growth of S. monoica and other closely related Oomycete species. In addition, many Oomycetes, including S. monoica, are pathogens of numerous animals and plants, and their cell wall polysaccharides play a central role in the relationship with the hosts (Johnson et al., 2002). Thus, deciphering the molecular mechanisms of cell wall polysaccharide biosynthesis in Oomycetes is an important step to better understand the biology of these microorganisms and, possibly, to develop new growth inhibitors targeted to cell wall synthesizing enzymes. Beyond the determination of the specific function of annexins in Oomycetes, the present work represents a first progress towards the achievement of these fundamental and economically important objectives. In addition to the characterization of p36 reported here, research on the identification and functional characterization of the other components of the (1[RIGHTWARDS ARROW]3)-β-d-glucan synthase complex, including the catalytic subunit, is ongoing in our laboratory.

Experimental procedures

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

Reagents

CHAPS, UDP-glucose, cellobiose, Pipes and Tris were purchased from Sigma and UDP-d-[U-14C]glucose (303 mCi mmol−1) was bought from Perkin Elmer. The liquid scintillation mixture was from Packard BioSciences (Groningen, the Netherlands). The reagents used for protein assay (Bradford, 1976), SDS-PAGE and 2D-PAGE analyses were from Bio-Rad. Kits were used for immunodetection of secondary antibodies on Western blots as well as for the extraction, amplification and cloning of nucleic acids, as indicated in the following sections. All other chemicals were bought from Sigma.

Strain and growth conditions

The strain S. monoica Pringsheim 53-967 Dick was obtained from the Centraal Bureau voor Schimmel Culture (CBS, Baarn, the Netherlands) and maintained on Potato Dextrose Agar (PDA) in Petri dishes. Proteins and nucleic acids were extracted from cells grown from 48 h to 96 h at 24°C in the liquid medium of Machlis (1953). The latter (∼100 ml) was inoculated with 30 agar plugs of ∼5 mm cut from the margins of colonies growing on PDA.

Cell fractionation and detergent extraction of membrane-bound proteins

Cells harvested after 3 or 4 days of growth in liquid medium (Machlis) were filtered through filter paper and disrupted at 4°C in extraction buffer (10 mM Tris-HCl pH 7.4) using a Waring Blender homogenizer for four periods of 30 s at maximal speed. Cell membranes were prepared by differential centrifugation (Bulone et al., 1990) and resuspended in extraction buffer. The protein content was determined by the Bradford (1976) assay (Bio-Rad) and the membrane suspension was diluted in extraction buffer to obtain a final protein concentration in the range 3–4 mg ml−1. Membrane-bound proteins were extracted in the presence of CHAPS, which was added to the suspension of membranes at a final concentration of 0.5% (critical micellar concentration). After 30 min at 4°C under continuous stirring, the preparation was centrifuged at 100 000 g for 1 h and the supernatant was used as a source of glucan synthase.

Rapid enrichment of the CHAPS-extracted (1[RIGHTWARDS ARROW]3)-β-d -glucan synthase and SDS-PAGE analysis

Fractions enriched in (1[RIGHTWARDS ARROW]3)-β-d-glucan synthase activity were obtained by centrifuging 2 ml of CHAPS extract on 28 ml linear glycerol gradients prepared using glycerol solutions in extraction buffer containing no EGTA or 10 mM EGTA (density = 1.025 and 1.110). Fractions of 2 ml were collected from the bottom of the centrifuge tubes after 3 h 45 min centrifugation at 120 000 g and 4°C in a swinging bucket rotor. Their protein content was determined as indicated earlier and (1[RIGHTWARDS ARROW]3)-β-d-glucan synthase activity was measured in each fraction as described in the next paragraph. Proteins in the gradient fractions were analysed by SDS-PAGE (Laemmli, 1970) and stained with silver (Blum et al., 1987).

(1[RIGHTWARDS ARROW]3)-β-d -Glucan synthase assay

(1[RIGHTWARDS ARROW]3)-β-d-Glucan synthase activity was determined in a total volume of 92.5 μl containing 25 μl of enzyme fraction, 8 mM Pipes-Tris pH 6.0, 10 mM cellobiose, 1.3 mM dithiothreitol, 1 mM UDP-glucose and 0.16 μM UDP-d-[U-14C]glucose. The mixture was incubated at 25°C for 1 h and the reaction was terminated by adding two volumes of ethanol. Reaction products synthesized in these conditions were characterized in great detail using a series of biochemical, chemical and physical techniques and unequivocally shown to be strictly linear (1[RIGHTWARDS ARROW]3)-β-d-glucans (Pelosi et al., 2003). After an overnight precipitation at −20°C, the 66% ethanol-insoluble polysaccharides were recovered on Whatman GF/C glass-fibre filters and successively washed with 4 ml of water and 66% ethanol. The radioactivity retained on the filters was measured by liquid scintillation (Wallac WinSpectral 1414 counter). Enzyme activity was expressed as the incorporation of glucose in ethanol-insoluble polysaccharides.

Purification of the 36 kDa annexin

Proteins from the annexin family are known to interact with plasma membranes in a calcium-dependent manner (Burgoyne and Geisow, 1989). They can be released from intact membranes upon incubation in the presence of EGTA (Boustead et al., 1989; Smallwood et al., 1990). For this purpose, the pelleted cell membranes prepared as described earlier were resuspended in Tris-KOH buffer pH 7.4 containing 10% glycerol and 10 mM EGTA in order to obtain a final protein concentration of 4 mg ml−1. After a 45 min incubation at 4°C, the preparation was centrifuged at 150 000 g for 1 h and the supernatant (SEGTA) containing the released annexins was recovered. In the next purification step, annexins were precipitated by adding CaCl2 to SEGTA to a final concentration of 15 mM and by incubating the preparation for 45 min at 4°C under continuous stirring. The annexins were then pelleted (1 h centrifugation at 150 000 g) and resuspended in 10 mM Tris-KOH buffer pH 7.4 containing 10% glycerol and 20 mM EGTA. The re-solubilized annexins were recovered in the supernatant designated as fraction S1 after a centrifugation at 10 000 g for 10 min. The corresponding pellet was resuspended once more in the presence of 20 mM EGTA to solubilize more annexins. A last centrifugation at 10 000 g for 10 min gave a supernatant that contained solubilized annexins and that was designated as fraction S2.

2D-PAGE analysis

The protein profiles of supernatants S1 and S2 were analysed by 2D-PAGE. A theoretical amount of 100 μg total protein was denatured for 1 h at 20°C in a solution containing a final concentration of 8 M urea, 4% CHAPS and 20 mM dithiothreitol. The samples were then loaded on Ready Strip IPG (Bio-Rad) strips (7 or 11 cm, pH gradient 5–8) and the gels were actively rehydrated in the sample at 50 V and 20°C for 10 h. Isoelectric focalization was performed using the Protean IEF cell from Bio-Rad and the following steps: 250 V for 15 min, from 250 to 6000 V in 150 min and 6000 V for 6 h. Strips were then transferred in a reducing buffer containing 6 M urea, 2% SDS, 0.375 M Tris-HCl pH 6.8, 20% glycerol and 130 mM dithiothreitol. After a 15 min incubation, the reduced proteins were alkylated by transferring the gels for 20 min in the same buffer as above except for the addition of 0.01% bromophenol blue and the replacement of dithiothreitol by 135 mM iodoacetamide. The strips were then placed on top of SDS-PAGE gels and overlaid with 1% agarose in SDS running buffer prior to electrophoresis. After separation in the second-dimension gel (12.5% acrylamide), proteins were stained with silver (Blum et al., 1987) or Coomassie blue (for antibody production).

Antibody production and Western blot analyses

Antibodies were produced against the purified annexin by using the corresponding Coomassie-blue-stained spot obtained by 2D-PAGE analysis of fraction S1 as a source of antigen. Spots from several gels were pooled in order to obtain the equivalent of 50 μg pure protein per injection. Specific polyclonal antibodies were produced in rabbit by the company CovalAb (Lyon, France) using standard protocols. For Western blot analyses, proteins separated by SDS-PAGE were transferred to nitrocellulose membranes (Millipore, 0.45 μm), which were subsequently stained with Ponceau red (0.2% in 3% trichloroacetic acid), destained in water and blocked in Tris Buffer Saline (TBS: 20 mM Tris-HCl pH 7.4, 150 mM NaCl) containing 5% non-fat milk. After washing in TBS containing 0.1% Tween 20, the membranes were probed for 4 h at room temperature with the anti-annexin antiserum diluted 1/1000 in TBS-Tween, washed in the latter buffer and incubated for 2 h at room temperature with goat anti-rabbit antibodies conjugated to peroxidase (Sigma, ref. A-8275, dilution 1/2000 in TBS-Tween containing 0.5% milk). Detection of antibody binding was performed using an electrochemiluminescence-based kit according to the manufacturer's instructions (Amersham, ref. RPN 2109).

Reconstitution of (1[RIGHTWARDS ARROW]3)-β-d -glucan synthase activity using add-back experiments

Typical add-back experiments consisted in mixing 25 μl of a given enzyme preparation (detergent extract or various glycerol gradient fractions) with increasing amounts of purified annexin (0–1.2 μg). In some experiments, calcium concentrations in the range 10 μM−5 mM were added in the mixtures together with 0.6 μg of purified annexin. In all experiments the volume was completed to 92.5 μl with Pipes-Tris buffer pH 6.0 containing dithiothreitol, cellobiose, UDP-glucose and UDP-d-[U-14C]glucose to reach the same final concentrations of reagents as for a typical (1[RIGHTWARDS ARROW]3)-β-d-glucan synthase assay.

Partial hydrolysis of proteins and peptide sequencing

The procedure described by Matsudaira (1993) for protein sequencing was modified as follows. Pieces of Coomassie-blue-stained SDS-PAGE or 2D-PAGE gels were destained in 95% acetone for 1 h at 4°C and dried under vacuum at 30°C prior to rehydration in 0.1 M (NH4)2CO3. Gel pieces were then washed three times for 2 h at 4°C in the latter solution, vacuum-dried, rehydrated for 12 h at 4°C in 50 mM Tris-HCl buffer pH 8.0 containing 6 M guanidine-HCl and 5 mM dithiothreitol, and washed four times for 2–16 h in 0.1 M (NH4)2CO3. Proteins were successively reduced for 20 min at 50°C by incubating the gel pieces in a 0.1 M (NH4)2CO3 solution containing 13.5 mM dithiothreitol, and alkylated for 30 min in the dark and at room temperature by transferring the gel pieces in 0.1 M (NH4)2CO3 supplemented with 33 mM iodoacetamide. The gel pieces were then washed four times for 12 h at 4°C in 0.1 M (NH4)2CO3 and dried under vacuum as indicated earlier. At this stage, they were rehydrated in 50 mM (NH4)2CO3 pH 7.8 prior to partial proteolysis of the corresponding proteins. The latter step was performed by adding 2 μg ml−1 (final concentration) of either sequencing grade modified trypsin (Promega, ref. V5111) or sequencing grade modified endoproteinase Glu-C (Sigma, ref. P-6181). After 24 h incubation at 37°C, peptides were extracted from the gel pieces four times for 4 h at 4°C in 60% acetonitrile and 0.1% trifluoroacetic acid. Extracts were pooled and concentrated by evaporation. Peptides were separated by reverse-phase HPLC at a flow rate of 50 μl min−1 over 150 min, using a C8 Vydac 208 TP microbore column (1 mm × 150 mm, 5 μm, 300 Å) and a linear gradient of acetonitrile (8.6–70%) containing 0.1% trifluoroacetic acid. Peptides from individual peaks detected at 214 nm were sequenced at the ‘Institut de Biologie Structurale’ (IBS, Grenoble, France) using the Edman degradation procedure.

Isolation and sequencing of the full-length annexin cDNA

DNA was extracted and purified from the mycelium of S. monoica using the Nucleospin Tissue kit from Macherey Nagel (Düren, Germany). The following specific primers were devised for PCR amplification using the amino acid sequences obtained from the purified annexin and the codons of a homologous nucleotide sequence extracted from the Phytophthora Functional Genomics database (http://www.pfgd.org, Gajendran et al., 2006): 5′-upstream primer: GGCAGGACCAACGAGGAGATCAACAT; 3′-downstream primer: CGCCGCACTCAGCCCGTACTCGTC. The amplification was performed using 35 PCR cycles with an annealing temperature of 60°C. More specific primers were synthesized from the sequence of the amplified fragment (Biofidal, Lyon, France) to isolate overlapping cDNA fragments from purified mRNA. The latter were prepared using the μMACS mRNA isolation kit from Miltenyi Biotec (Bergisch Gladbach, Germany), following the manufacturer's recommendations for the isolation of mRNA from plant tissues. Overlapping cDNA were obtained by RACE-PCR according to the instructions given by the manufacturer of the First Choice RLM-RACE kit (Ambion, Austin, USA), using the outer 3′-downstream primer CTTCACCGGCCTTGTAAATGACTTCC and the inner 3′-downstream primer TTGGTTGACGATGGCAAGGTAGAA for the 5′ end. For amplification of the 3′ end, the outer 5′-upstream primer was ATGACCTGAGCGGTGACTTG and the inner 5′-upstream primer was AGAACATGGAAGGGCATTGG. The other primers were from the kit. The amplified cDNA fragments were sequenced by the company Biofidal (Lyon, France).

Phylogenetic analysis of the 34 kDa annexin

Phylogenetic analysis was performed with sequences from Aspergillus fumigatus (ANXC3.1 and ANXC3.2; Khalaj et al., 2004), Aspergillus niger (Accession No. AY033935), D. discoideum (Accession No. M69022), N. crassa (Accession No. AF036871), P. infestans[Accession No. from the Phytophthora Functional Genomics database (http://www.pfgd.org): Pi_003_34757_Jun03 and Pi_002_33858_Jun03], Phytophthora sojae (Accession No. Pi_001_20934_Jun03, Pi_002_21817_Jun03, Pi_006_22647_Jun03 and Pi_014_22864_Jun03) and S. parasitica[Accession No. from the Oomycete Genomics database (http://www.oomycete.org/ogd/): Sp_002_00572_May04]. One sequence from A. thaliana (Accession No. AF083913) was used as an outgroup. Amino acid alignment was performed with clustal w (Thompson et al., 1994) from the SEAVIEW alignment editor (Galtier et al., 1996) used for manual refinement. The phylogenetic analysis was based on the annexin conserved core tetrad region. A phylogenetic tree was constructed with the PHYLO_WIN software (Galtier et al., 1996) according to the distance-based neighbour-joining method (Saitou and Nei, 1987), following the Poisson-corrected proportion of amino acid differences for all pairwise comparisons. As the available sequences were not complete (ESTs), different simulations were made to attest the robustness of the phylogenetic reconstruction. The statistical significance of the branching orders was assessed by the bootstrap test (Felsenstein, 1985).

Acknowledgements

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

The authors express their gratitude to Dr Jean Gagnon and Jean-Pierre Andrieu (‘Institut de Biologie Structurale’, Grenoble, France) for assistance with the Edman degradation procedure and for the use of their sequencing equipment.

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information
  • Andrawis, A., Solomon, M., and Delmer, D.P. (1993) Cotton fiber annexins: a potential role in the regulation of callose synthase. Plant J 3: 763772.
  • Aronson, J.M., and Lin, C.C. (1978) Hyphal cell wall chemistry of Leptomitus lacteus. Mycologia 70: 363369.
  • Baldauf, S.L., Roger, A.J., Wenk-Siefert, I., and Doolittle, W.F. (2000) A kingdom-level phylogeny of eukaryotes based on combined protein data. Science 290: 972977.
  • Bandorowicz-Pikula, J., Buchet, R., and Pikula, S. (2001) Annexins as nucleotide-binding proteins: facts and speculations. Bioessays 23: 170178.
  • Bartnicki-Garcia, S. (1968) Cell wall chemistry, morphogenesis and taxonomy of fungi. Annu Rev Microbiol 22: 87108.
  • Barton, G.J., Newman, R.H., Freemont, P.S., and Crumpton, M.J. (1991) Amino acid sequence analysis of the annexin super-gene family of proteins. Eur J Biochem 198: 749760.
  • Beauvais, A., Bruneau, J.M., Mol, P.C., Buitrago, M.J., Legrand, R., and Latgé, J.P. (2001) Glucan synthase complex of Aspergillus fumigatus. J Bacteriol 183: 22732279.
  • Billon-Grand, G., Marais, M.-F., Joseleau, J.-P., Girard, V., Gay, L., and Fèvre, M. (1997) A novel (1[RIGHTWARDS ARROW]3)-β-glucan synthase from the Oomycete Saprolegnia monoica. Microbiology 143: 31753183.
  • Blum, H., Beier, H., and Gross, H. (1987) Improved silver staining of plant proteins, RNA and DNA in polyacrylamide gels. Electrophoresis 8: 9399.
  • Boustead, C.M., Smallwood, M., Small, H., Bowles, D.J., and Walker, J.H. (1989) Identification of calcium-dependent phospholipid-binding proteins in higher plant cells. FEBS Lett 244: 456460.
  • Bradford, M.M. (1976) A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 72: 248254.
  • Braun, E.L., Kang, S., Nelson, M.A., and Natvig, D.O. (1998) Identification of the first fungal annexin: analysis of annexin gene duplications and implications for eukaryotic evolution. J Mol Evol 47: 531543.
  • Bulone, V., and Fèvre, M. (1996) A 34-kDa polypeptide is associated with (1[RIGHTWARDS ARROW]3)-β-glucan synthase activity from the fungus Saprolegnia monoica. FEMS Microbiol Lett 140: 145150.
  • Bulone, V., Girard, V., and Fèvre, M. (1990) Separation and partial purification of (1[RIGHTWARDS ARROW]3)-β-glucan and (1[RIGHTWARDS ARROW]4)-β-glucan synthases from Saprolegnia. Plant Physiol 94: 17481755.
  • Bulone, V., Chanzy, H., Gay, L., Girard, V., and Fèvre, M. (1992) Characterization of chitin and chitin synthase from the cellulosic cell wall fungus Saprolegnia monoica. Exp Mycol 16: 821.
  • Bulone, V., Fincher, G.B., and Stone, B.A. (1995) In vitro synthesis of a microfibrillar (1[RIGHTWARDS ARROW]3)-β-glucan by a ryegrass (Lolium multiflorum) endosperm (1[RIGHTWARDS ARROW]3)-β-glucan synthase enriched by product entrapment. Plant J 8: 213225.
  • Burgoyne, R.D., and Geisow, M.J. (1989) The annexin family of calcium-binding proteins. Cell Calcium 10: 110.
  • Calvert, C.M., Gant, S.J., and Bowles, D.J. (1996) Tomato annexins p34 and p35 bind to F-actin and display nucleotide phosphodiesterase activity inhibited by phospholipid binding. Plant Cell 8: 333342.
  • Campos-Takaki, G.M., Dietrich, S.M.C., and Mascarenhas, Y. (1982) Isolation and characterization of chitin from the cell walls of Achlya radiosa. J Gen Microbiol 128: 207209.
  • Chen, J.M., Sheldon, A., and Pincus, M.R. (1993) Structure-function correlations of calcium binding and calcium channel activities based on 3-dimensional models of human annexins I, II, III, V and VII. J Biomol Struct Dyn 10: 10671089.
  • Delmer, D.P. (1999) Cellulose biosynthesis: exciting times for a difficult field of study. Annu Rev Plant Physiol Plant Mol Biol 50: 245276.
  • Delmer, D.P., and Potikha, T.S. (1997) Structures and functions of annexins in plants. Cell Mol Life Sci 53: 546553.
  • Doblin, M.S., Kurek, I., Jacob-Wilk, D., and Delmer, D.P. (2002) Cellulose biosynthesis in plants: from genes to rosettes. Plant Cell Physiol 43: 14071420.
  • Douglas, C.M., Foor, F., Marrinan, J.A., Morin, N., Nielsen, J.B., Dahl, A.M., et al. (1994) The Saccharomyces cerevisiae FKS1 (ETG1) gene encodes an integral membrane protein which is a subunit of (1[RIGHTWARDS ARROW]3)-β-glucan synthase. Proc Natl Acad Sci USA 91: 1290712911.
  • Felsenstein, J. (1985) Confidence limits on phylogenies: an approach using the bootstrap. Evolution 39: 783791.
  • Fèvre, M. (1983) Nucleotide effects on glucan-synthesis activities of particulate enzymes from Saprolegnia. Planta 159: 130135.
  • Fry, W.E., and Goodwin, S.B. (1997a) Re-emergence of potato and tomato late blight in the United States. Plant Dis 81: 13491357.
  • Fry, W.E., and Goodwin, S.B. (1997b) Resurgence of the Irish potato famine fungus. Bioscience 47: 363371.
  • Gajendran, K., Gonzales, M.D., Farmer, A., Archuleta, E., Win, J., Waugh, M.E., and Kamoun, S. (2006) Phytophthora functional genomics database (PFGD): functional genomics of phytophthora–plant interactions. Nucleic Acids Res 34: D465D470.
  • Galtier, N., Gouy, M., and Gautier, C. (1996) SEAVIEW and PHYLO_WIN: two graphic tools for sequence alignment and molecular phylogeny. Comput Appl Biosci 12: 543548.
  • Geisow, M., Fritsche, U., Hexham, J., Dash, B., and Johnson, T.A. (1986) A consensus amino acid sequence repeat in Torpedo and mammalian calcium-dependent membrane binding proteins. Nature 320: 636638.
  • Gerke, V., and Moss, S.E. (2002) Annexins: from structure to function. Physiol Rev 82: 331371.
  • Girard, V., and Fèvre, M. (1991) Solubilization of a membrane-bound stimulator of (1[RIGHTWARDS ARROW]3)-β-glucan synthase from Saprolegnia. Plant Sci 76: 193200.
  • Hofmann, A. (2004) Annexins in the plant kingdoms. Annexins 1: 5161.
  • Hofmann, A., Delmer, D.P., and Wlodawer, A. (2003) The crystal structure of annexin Gh1 from Gossypium hirsutum reveals an unusual S3 cluster. Eur J Biochem 270: 25572564.
  • Johnson, T.W., Jr, Seymour, R.L., and Padgett, D.E. (2002) Biology and Systematics of the Saprolegniaceae. Book published on the World-Wide-Web: http://dl.uncw.edu
  • Khalaj, V., Smith, L., Brookman, J., and Tuckwell, D. (2004) Identification of a novel class of annexin genes. FEBS Lett 562: 7986.
  • Kumar, C., and Rzhetsky, A. (1996) Evolutionary relationships of eukaryotic kingdoms. J Mol Evol 42: 183193.
  • Laemmli, U.K. (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227: 680685.
  • Lartseva, L.V. (1986) Saprolegnia on the spawn of sturgeons and salmons. Hydrobiol J 22: 103107.
  • Li, J., Burton, R.A., Harvey, A.J., Hrmova, M., Wardak, A.Z., Stone, B.A., and Fincher, G.B. (2003) Biochemical evidence linking a putative callose synthase gene with (1[RIGHTWARDS ARROW]3)-β-d-glucan biosynthesis in barley. Plant Mol Biol 53: 213225.
  • Lin, C.C., and Aronson, J.M. (1970) Chitin and cellulose in the cell walls of the Oomycete Apodachlya sp. Arch Mikrobiol 72: 111114.
  • McClung, A.D., Carroll, A.D., and Battey, N.H. (1994) Identification and characterization of ATPase activity associated with maize (Zea mays) annexins. Biochem J 303: 709712.
  • Machlis, L. (1953) Growth and nutrition of watermolds in the subgenus Euallomyces. II – optimal composition of the minimal medium. Am J Bot 40: 449460.
  • Margulis, L., and Schwartz, K.V. (2000) Five Kingdoms: An Illustrated Guide to the Phyla of Life on Earth. New York, NY: Freeman and Co.
  • Matsudaira, P. (1993) A Practical Guide to Protein and Peptide Purification for Microsequencing. San Diego: Academic Press.
  • Mazur, P., and Baginsky, W. (1996) In vitro activity of (1[RIGHTWARDS ARROW]3)-β-glucan synthase requires the GTP-binding protein Rho1. J Biol Chem 271: 1460414609.
  • Neish, G.A., and Hughes, G.C. (1980) Fungal Diseases of Fishes, Book 6. Neptune, NJ: T.W.F. Publications.
  • Paquin, B., Laforest, M.J., Forget, L., Roewer, I., Wang, Z., Longcore, J., and Lang, B.F. (1997) The fungal mitochondrial genome project: evolution of fungal mitochondrial genomes and their gene expression. Curr Genet 31: 380395.
  • Pear, J.R., Kawagoe, Y., Schreckengost, W.E., Delmer, D.P., and Stalker, D.M. (1996) Higher plants contain homologs of the bacterial celA genes encoding the catalytic subunit of cellulose synthase. Proc Natl Acad Sci USA 93: 1263712642.
  • Pelosi, L., Imai, T., Chanzy, H., Heux, L., Buhler, E., and Bulone, V. (2003) Structural and morphological diversity of (1[RIGHTWARDS ARROW]3)-β-d-glucans synthesized in vitro by enzymes from Saprolegnia monoica. Comparison with a corresponding in vitro product from blackberry (Rubus fruticosus). Biochemistry 42: 62646274.
  • Qadota, H., Python, C.P., Inoue, S.B., Arisawa, M., Anraku, Y., Zheng, Y., et al. (1996) Identification of yeast Rho1p GTPase as a regulatory subunit of (1[RIGHTWARDS ARROW]3)-β-glucan synthase. Science 272: 279281.
  • Saitou, N., and Nei, M. (1987) The neighbor-joining method: a new method for reconstructing phylogenetic trees. Mol Biol Evol 4: 406425.
  • Shematek, E.M., and Cabib, E. (1980) Biosynthesis of the yeast cell wall. II. Regulation of (1[RIGHTWARDS ARROW]3)-β-glucan synthetase by ATP and GTP. J Biol Chem 255: 895902.
  • Shin, H., and Brown, R.M., Jr (1999) GTPase activity and biochemical characterization of a recombinant cotton fiber annexin. Plant Physiol 119: 925934.
  • Smallwood, M.F., Gurr, S.J., McPherson, M.J., Roberts, K., and Bowles, D.J. (1990) Purification and partial sequence analysis of plant annexins. Biochem J 281: 501505.
  • Thompson, J., Higgins, D., and Gibson, T. (1994) clustal w: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Res 22: 46734680.
  • Walker, J.E., Saraste, M., Runswick, M.J., and Gay, N.J. (1982) Distantly related sequences in the α- and β-subunits of ATP synthase, myosin, kinases, and other ATP-requiring enzymes and a common nucleotide-binding fold. EMBO J 1: 945951.
  • Wessels, J.G.H., and Sietsma, J.H. (1981) Fungal cell walls: a survey. In Encyclopedia of Plant Physiology. Plant Carbohydrates II. Tanner, W., and Loewus, F.A., (eds). Berlin: Springer Verlag, pp. 352394.

Supporting Information

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

Fig. S1. Amino acid sequence of the annexin from S. monoica and alignment against examples of fungal, plant and mammal annexin sequences Fig. S2. Phylogenetic position of the annexin sequence from S. monoica. Table S1. Identification of the three major proteins that co-purify with (1[RIGHTWARDS ARROW]3)-β-D-glucan synthase activity (see Fig. 1).

FilenameFormatSizeDescription
MMI_5389_sm_FigsS1-S2_TableS1.pdf36KSupporting info item

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.