Formation of the outer layer of the Dictyostelium spore coat depends on the inner-layer protein SP85/PsB

Talibah Metcalf1, Karen Kelley2, Gregory W. Erdos2, Lee Kaplan1 and Christopher M. West1

1 Dept of Anatomy and Cell Biology, ICBR, University of Florida, Gainesville, FL 32610-0235, USA
2 College of Medicine and Electron Microscopy Core Laboratory, ICBR, University of Florida, Gainesville, FL 32610-0235, USA

Correspondence
Christopher M. West
westcm{at}ufl.edu


   ABSTRACT
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
The Dictyostelium spore is surrounded by a 220 µm thick trilaminar coat that consists of inner and outer electron-dense layers surrounding a central region of cellulose microfibrils. In previous studies, a mutant strain (TL56) lacking three proteins associated with the outer layer exhibited increased permeability to macromolecular tracers, suggesting that this layer contributes to the coat permeability barrier. Electron microscopy now shows that the outer layer is incomplete in the coats of this mutant and consists of a residual regular array of punctate electron densities. The outer layer is also incomplete in a mutant lacking a cellulose-binding protein associated with the inner layer, and these coats are deficient in an outer-layer protein and another coat protein. To examine the mechanism by which this inner-layer protein, SP85, contributes to outer-layer formation, various domain fragments were overexpressed in forming spores. Most of these exert dominant negative effects similar to the deletion of outer-layer proteins, but one construct, consisting of a fusion of the N-terminal and Cys-rich C1 domain, induces a dense mat of novel filaments at the surface of the outer layer. Biochemical studies show that the C1 domain binds cellulose, and a combination of site-directed mutations that inhibits its cellulose-binding activity suppresses outer-layer filament induction. The results suggest that, in addition to a previously described early role in regulating cellulose synthesis, SP85 subsequently contributes a cross-bridging function between cellulose and other coat proteins to organize previously unrecognized structural elements in the outer layer of the coat.


Abbreviations: cA-ISM, extracellular (interspore matrix) fraction obtained from cells sporulated in the presence of 8-Br-cAMP; pAb, polyclonal antibody


   INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Cellulose is a major component of the cell walls of plants, algae and several free-living eukaryotic micro-organisms, including the algal plant pathogen Phytophthera parasiticans, the water mould Achlya and cysts of Acanthamoeba, the soil amoeba Hartmannella glebae, the green alga Acetabularia and the amoeba-flagellates Naegleria gruberi and Schizopyrenus russelli (reviewed by West, 2002). Most of these organisms are either economically important or pathogenic to plants or animals, either intrinsically or as carriers of pathogenic bacteria such as Legionella (Harb et al., 2000). Some of these walls contain proteins with carbohydrate-binding activity, yet the roles of these proteins are not understood. Cellulose-binding activities were first discovered in cellulases and other polysaccharidases, as discrete domains of 35–100 aa that help to target the hydrolase. In certain anaerobic bacteria and fungi, these enzymes form multiprotein complexes called cellulosomes (Bayer et al., 1998). More recently, cellulose-binding activities have been identified in non-hydrolases that appear to contribute to cell wall organization. A wall protein virulence factor in the algal plant pathogen Phytophthera parasiticans was observed to bind cellulose via a duplicated Cys-rich domain (Mateos et al., 1997), which may be important for organization of cell wall cellulose. Previous studies have shown that the structural protein SP85 from the Dictyostelium spore coat binds cellulose in vitro (Zhang et al., 1998, 1999), and additional cellulose-binding proteins have been discovered recently in the slime sheath and stalk tube of Dictyostelium (Wang et al., 2001). Cellulose-binding proteins, and other carbohydrate-binding proteins such as galectins in fungi (Boulianne et al., 2000) and a Cys-rich chitin-binding protein in Entamoeba (Frisardi et al., 2000), have high potential to regulate events at the molecular interface between polysaccharides and proteins in forming walls.

The spore coat of Dictyostelium is a cell wall specialized for long-term protection of the enclosed spore cell from environmental extremes. The spore coat consists of a thick middle layer composed primarily of cellulose fibrils sandwiched between an electron-dense inner layer next to the plasma membrane and an electron-dense outer layer that is exposed to the environment (West, 2002). Cell walls from many organisms, including yeast, fungi and epidermal cells of vascular plants, exhibit a similar trilaminar organization. The mechanism by which multiple layers are formed is poorly understood, but a recent study in Dictyostelium implicates the spore coat protein SP85 in a feedback loop that regulates cellulose synthesis (West et al., 2002). Absence of SP85 appeared to lead to precocious cellulose synthesis, whereas expression of a fragment of SP85 containing two of its domains inhibited cellulose synthesis. In addition, cellulose deposition appears to be required for subsequent condensation of the outer electron-dense layer (Zhang et al., 2001).

Wild-type spores show low permeability to a fluorescent lectin probe that binds to a polysaccharide rich in galactose and N-acetylgalactosamine associated with the inner layer of the coat (Erdos & West, 1989). The location and intense electron density of the outer layer suggests that it composes the permeability barrier to fluorescent lectins, which was supported by increased permeability of a glycosylation mutant that affects the fucosylation of coat proteins associated with the outer layer (Gonzalez-Yanes et al., 1989). Multiple proteins, including SP96, SP87, SP80, SP70 and SP60, are associated with the outer layer based on immunocytochemical localization (West & Erdos, 1988; Erdos & West, 1989) and protease accessibility studies (Orlowski & Loomis, 1979). Mutant spores lacking SP96, SP70 or SP60 as a result of gene targeting show increased labelling by a fluorescent lectin (Fosnaugh et al., 1995) and an antibody against the coat protein SP85, also associated with the inner layer (Srinivasan et al., 2000). Labelling is even greater in strains lacking two or three of the proteins, suggesting that they each contribute independently and additively to outer-layer function. These findings implicate the outer layer as the permeability barrier but morphological evidence to support this model has been lacking.

Spores lacking SP85, associated with the inner layer, also exhibit increased lectin labelling compared to normal spores, but the effect is more subtle as it requires pre-extraction with hot urea and 2-mercaptoethanol (Zhang et al., 1999). A similar effect is observed in a mutant targeting the O-glycosylation of mucin-like regions of SP85 (Aparacio et al., 1990). These studies suggested that SP85 contributes to outer-layer functionality by cross-bridging other molecules in the coat based on the similar effects of expressing fragments of SP85 hypothesized to have independent binding functions. For example, the C-terminal (C1C2) domain of SP85 binds cellulose and another coat protein, SP65, in vitro (Zhang et al., 1998, 1999), and other studies suggest that SP85 can be isolated as a multiprotein complex (McGuire & Alexander, 1996). Based on these binding activities, SP85 is hypothesized to contribute to the organization of protein-cellulose interfaces between the coat layers. New studies reported here document that deletion of SP85 or expression of SP85 fragments renders ultrastructural effects that are, for some constructs, similar to the deletion of outer-layer proteins. Biochemical studies show that SP85 is important for the incorporation of the outer-layer coat protein SP96, and mutational studies suggest that the cellulose-binding activity of SP85, which maps to its Cys-rich C1 domain, contributes to SP85 action. In addition, the appearance of novel, regularly spaced, punctate electron densities and curvilinear thin filaments in the spore coats of specific mutant strains suggests previously unrecognized units of organization of the outer layer of the coat.


   METHODS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Cell strains and culture.
All cell strains, listed in Table 1, were grown on HL-5 medium. Strains expressing recombinant SP85 fragments were grown in the presence of 90 µg G418 ml-1 to ensure maintenance of the expression construct (Zhang et al., 1999), except during the final passage prior to sporulation. Sporulation was induced at an air–water interface by washing growing cells in KP buffer (10 mM potassium phosphate, pH 6·5) and depositing on non-nutrient agar plates at 1·5x106 cells cm-2 (Zhang et al., 2001). After 48 h, sori were collected by scraping the cells into KP buffer and cells were separated from the soluble fraction, referred to as interspore matrix, by centrifugation at 13 000 g for 1 min. Cells were resuspended in KP buffer, filtered to remove stalks and cell aggregates, recovered by centrifugation at 2000 g for 1 min and resuspended in KP buffer. Spore concentration was determined by diluting an aliquot in 0·1 % Calcofluor White ST (American Cyanamid) in H2O and counting in a haemocytometer. The interspore matrix fraction was concentrated by vacuum centrifugation.


View this table:
[in this window]
[in a new window]
 
Table 1. Antibodies and cell strains employed

 
To induce sporulation in suspension, cells which had developed for 16–20 h (past the upright finger stage) were scraped from the surface of one to two 10 cm diameter non-nutrient agar plates and transferred to a 15 ml tube on ice. Cells were suspended in 2 ml 25 mM EDTA/KOH in KP (pH 6·5) using a long-stemmed Pasteur pipette at room temperature until dissociated to single cells. The tube was filled with ice-cold KP and an aliquot removed for counting in a haemocytometer. Cells were recovered by centrifugation at 2000 g for 2 min, the supernatant was removed completely by aspiration and cells were immediately resuspended in 1·5 ml cold KP buffer per original plate. To this 0·4 ml 100 mM 8-Br-cAMP-KOH in KP (pH 6·5) was added to a final concentration of 20 mM (Maeda, 1992). The final cell density was 5x107 ml-1. The suspension was incubated on a gyratory shaker for 6 h at 22 °C. The culture supernatant (cA-ISM) was recovered by centrifugation at 13 000 g for 15 min, supplemented with protease inhibitors (1 mM PMSF, 10 µg leupeptin ml-1, 10 µg aprotinin ml-1) and stored at -80 °C.

Spore coats were purified from intact spores as described by Zhang et al. (1998).

Transmission electron microscopy.
Spores were fixed in glutaraldehyde or a combination of glutaraldehyde and formalin at pH 7·2 (McDowell & Trump, 1976), collected on poly-L-lysine-treated Millipore membrane filters and covered with 3 % agarose. Agarose-embedded spores were peeled away from the filters, post-fixed in 1 % buffered osmium tetroxide, stained en bloc and dehydrated in a graded ethanol series and 100 % acetone. Spores were embedded in EmBed 812 epoxy resin and polymerized at 60 °C for 2 d. Thin sections of approximately 70 nm were collected on Butvar-coated 100-mesh copper grids and post-stained with 2 % aqueous uranyl acetate, followed by Reynold's lead citrate. Grids were examined on a Hitachi H-7000 transmission electron microscope (Hitachi Scientific Instruments). Digital micrographs were taken on a BioScan/Digital Micrograph 2·5 (Gatan) at an exposure level optimized for viewing the outer layer and processed with MEGA View III/AnalySIS 3.1 (Soft Imaging System).

For immunolocalization studies, spores were fixed in 1 % (v/v) glutaraldehyde, 4 % (w/v) paraformaldehyde in 15 mM sodium phosphate, 135 mM NaCl, pH 7·2 (PBS), for 1 h at 4 °C. Spores were recovered and dehydrated as above and embedded in Unicryl acrylic resin (British Biocell International) under UV light at -10 °C for 2 d. Thin sections (70 nm) were collected on Butvar-coated 400-mesh nickel grids. Sections were treated with 1 % (w/v) non-fat dry milk in 0·1 % Tween-20, 0·5 M NaCl, 0·05 M Tris/HCl, pH 7·2 (HST), incubated overnight in a humid chamber at 4 °C in mAb 83.5 (1 : 1000 ascites) in HST, washed in HST and incubated with a secondary antibody (goat anti-mouse IgG, 1 : 30 diluted in PBS) conjugated to 18 nm colloidal gold (Jackson ImmunoResearch Laboratories) for 1 h at room temperature. Sections were post-stained and observed as above.

Mutant and normal strain Ax3 spores were prepared in parallel to ensure comparable fixation and staining. NC1- and mutant NC1-expressing spores were also processed in parallel. Micrographs shown are representative of images from at least two independent clones.

Site-directed mutagenesis.
pVSBNC1 (West et al., 2002) was modified by site-directed mutagenesis using the QuikChange kit (Stratagene) as described by Sassi et al. (2001). The nucleotide sequence for W352EN was changed from TGGGAAAAT to TCAAAACAT; the sequence for R390GK was changed from CGTGGTAAG to AGTGGTGAG; the sequence for K417GQ was changed from AAAGGTCAA to CAAGGTGAA; and the sequence for R445RGE was changed from AGACGTGGTGAA to AGTGCTGCTCAA. For some of the constructs it was necessary to design the oppositely oriented mutagenic primers in staggered (3–5 nt), rather than perfectly overlapping fashion to amplify plasmid DNA. Simultaneous introduction of these mutations in pVSBNC1 was performed as described by Wang & Malcolm (1999).

Anti-SP85 antibody.
A frozen spleen from a mouse immunized with the bacterially expressed, T7-tagged N domain of SP85 (NT7) that produced polyclonal antibody (pAb) NT7 used in a previous study (Zhang et al., 1999) was thawed and cells dissociated from the tissue were fused with plasmacytoma SP2/0 cells to create hybridomas as described by Simrell & Klein (1979). Hybridomas were grown in HAT selective media supplemented with 20 % horse serum. Hybridoma clones were chosen based on the ability of antibodies in their culture supernatants to immunoprecipitate soluble SP85 from the cA-ISM fraction (see above) and recognize SP85 on Western blots. Clone 5F5 from this screen is used in this report.

Cellulose-binding assay.
Avicel powder (type PH-101; FMC corporation) was stored as a 50 mg ml-1 suspension in H2O, 0·02 % NaN3 at 4 °C. For each assay, a volume containing 9 mg Avicel was transferred with a wide-bore pipette tip to a 1·5 ml polypropylene tube, centrifuged at 10 000 g for 1 min and the supernatant removed by aspiration. cA-ISM (200 µl) was pre-centrifuged at 13 000 g for 5 min to remove any insoluble material and used to resuspend the Avicel pellet. The suspension was agitated at 22 °C for 13–16 h. Binding was incomplete at 2 h (data not shown). The bound and non-bound fractions were separated by centrifugation at 10 000 g for 1 min and the supernatant was transferred to a new tube (non-bound fraction). The pellet was resuspended in 0·5 ml 50 mM NH4Ac (pH 6·8), centrifuged again and the supernatant aspirated. In some experiments, 150 µl of the non-bound supernatant was reincubated with 9 mg Avicel and the process repeated. 1x, 2x times; or 4x times; Laemmli sample buffer containing 50 mM dithiothreitol was added to each sample, which was boiled, precleared by centrifugation and loaded onto individual lanes of a 7–20 % linear gradient SDS-PAGE gel. Gels were electroblotted onto 0·45 µm pore-size nitrocellulose film and probed with primary antibodies (Table 1). Primary antibody labelling was detected using an appropriate alkaline phosphatase-conjugated secondary antibody, which was developed colorimetrically with 5-bromo-4-chloro-3-indolylphosphate and nitro blue tetrazolium (Zhang et al., 1999).


   RESULTS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
The outer layer is structurally deficient in coats lacking outer-layer proteins
The outer-layer-associated proteins SP96, SP70 and SP60 are not expressed in strain TL56 due to targeted disruption of their genes (Fosnaugh et al., 1995). A side-by-side comparison of thin sections from spores of strain TL56 and the normal strain Ax3 by electron microscopy showed a difference in the outer layers of their coats. Instead of presenting as a continuous electron-dense layer as in Ax3 (Fig. 1a), the outer layer of TL56 displayed a markedly reduced electron density and in favourable sections, consisted of a series of electron densities, or dots, with a mean spacing of 34 nm (Fig. 1c, e). The dots were not always so apparent, presumably because multiple dots can be accommodated within the 70 nm thickness of the preparation. The diminished electron density of the outer layer was seen all around the spore and in all of the spores in the field. In tangential sections the electron densities appeared to form a regular two-dimensional array (Fig. 1d). Some images of normal Ax3 spores showed an unevenness or undulation in the outer layer that appeared to have a regularity similar to the spacing of the dots in TL56 coats (Fig. 1b). Though the example shown is likely to represent a distortion resulting from the preparation, the similar periodicity suggests that the residual structures seen in TL56 coats correspond to a normal underlying structure. Thus one or more of the coat proteins SP96, SP70 and SP60 is required for and may constitute the material that fills the gap between the dots, and their absence is likely to explain the increased permeability of the coat to macromolecular tracers (Fosnaugh et al., 1995; Srinivasan et al., 2000).



View larger version (145K):
[in this window]
[in a new window]
 
Fig. 1. Electron microscopy of normal and mutant spores deficient in outer-layer proteins. Spores from strains Ax3 (a, b) or TL56 (c–e) were fixed, embedded, sectioned and stained with uranyl acetate and lead citrate, in parallel, to ensure comparable treatment. All sections are oriented in the same direction with the outer layer to the right. Panels (a), (c) and (e) are approximately normal to the cell surface; panels (b) and (d) are partially tangential. The rulers in panels (b) and (c) correspond to periodicities in the patterns of electron density of the outer layer. Since the measured thickness of the coat in freeze-fracture replicas is 220 nm (West, 2002), the coats appear to have undergone shrinkage in these preparations.

 
The outer layer is also deficient in coats lacking the inner-layer protein SP85
The inner-layer protein SP85 is absent in strain HW70 (SP85-) due to targeted disruption of its gene (Zhang et al., 1999). A side-by-side comparison of strain SP85- and Ax3 spores showed that the outer layers of SP85-null coats have an overall diminished electron density (Fig. 2a–c). The material remaining exhibits, in some preparations, evidence of periodicity, with a spacing of 30–35 nm. In tangential sections these densities appeared to be more linear than those of strain TL56 (Fig. 2d), which may explain why they were less apparent in perpendicular sections. The low electron density of the outer layer was seen around the entire perimeter of the spore and in all of the spores in the field. In addition, SP85- spores appeared to have less dense packing of organelles in their cytoplasm and tended to have more open vacuoles and more irregular margins than Ax3 spores, features that were also apparent in living spores viewed in the light microscope using phase-contrast optics (data not shown). This may have been due to the previously described precocious coat formation in this strain (West et al., 2002).



View larger version (65K):
[in this window]
[in a new window]
 
Fig. 2. Electron microscopy of spores deficient in the inner-layer protein SP85. (a–d) Spores from strains Ax3 (a) and HW70 (SP85-; b–d) were prepared for electron microscopy in parallel as described in the legend to Fig. 1. Panels (a–c) were approximately normal to the plane of the outer layer, whereas panel (d) was partially tangential. Outer layers are to the right. (e) Spore coats were purified from the strains indicated and 1·4x106 coats per lane were analysed by SDS-PAGE and Western blotting, by immunoprobing with mAb 16.1 for SP85 and mAb 83.5 for SP96 and SP75. Based on comparisons with other samples, not all analysed in the same gel (e.g. West et al., 1996), lane D is underloaded and lane C is overloaded relative to lanes A and B.

 
The composition of SP85- spore coats was examined by SDS-PAGE and Western blotting (Fig. 2e). As expected, SP85- coats were lacking in SP85. In addition, these coats were missing SP96, but expressed SP75. As controls, TL56 coats contained SP85 and SP75, but were missing SP96 as expected, and DP87- coats were missing SP75 as expected, but contained SP96 and SP85. SP85- coats were previously shown to also be deficient in the coat protein SP65 (Zhang et al., 1999). The reduced level of SP96 was not noted in that study presumably because Coomassie blue staining rather than Western blot analysis had been used to examine coat protein composition. Thus the inner-layer protein SP85 appears to contribute to the organization of the outer layer, possibly by supporting the incorporation of SP96 and SP65.

Expression of SP85 fragments disrupts the outer layer
The sequence of SP85 suggests that it consists of a series of globular domains (West, 2002), named N (for NH2-terminal), M (for middle), C1 (for COOH-terminal 1) and C2 (for COOH-terminal 2), as depicted in Fig. 3(a). The M domain is flanked by mucin-like domains which consist of tetrapeptide repeats, named TR1 and TR2. Individual domains or domain combinations were previously expressed in prespore cells to investigate the role of SP85 in the regulation of coat formation and in coat permeability (Zhang et al., 1999; West et al., 2002). These included the N domain, the C1C2 domain and fusions of the N domain with either C1C2, C1 or C2. The latter constructs were designed to ensure delivery to the prespore vesicle as the N domain carries targeting information (Zhang et al., 1999) and to promote folding of the short C1 and C2 domains (C. M. West and others, unpublished data). Each construct was incorporated efficiently into the coat (or precoat in the case of NC1), except for N which was recovered in the interspore matrix (Zhang et al., 1999; West et al., 2002). Further analysis of NC1 and NC2 expression using domain-specific antibodies (see Table 1) showed partial cleavage of these constructs (Fig. 4a), releasing the N domain into the interspore matrix, and retention of free C1 with the cell (C2 was not detected). Cleavage appeared to occur at the internal myc epitope, which separates the N and C1 domains, based on absence of mAb 9E10 binding and correspondence of apparent Mr values of the N and C1 domains with their expected values (data not shown). Thus, expression of the NC1-construct resulted in the accumulation of NC1, N and C1; expression of NC2 resulted in the accumulation of NC2 and N.



View larger version (106K):
[in this window]
[in a new window]
 
Fig. 3. Electron microscopy of spores expressing fragments of SP85. (a) Schematic diagram of SP85 domains based on sequence analysis and domain expression studies. (b–g) Representative EM sections from the normal strain Ax3 (b), normal strains expressing the NC1C2 domain fusion (c), the N domain (d) or the C1C2 domain (e), and the NC2 domain fusion expressed in Ax3 (f) or HW70 (SP85-; g), are shown. The outer layer is to the right.

 


View larger version (100K):
[in this window]
[in a new window]
 
Fig. 4. Western blot analysis of mutant spores. (a) Fruiting bodies from strains expressing NC1 or NC2 were dispersed in buffer, filtered to remove stalks and centrifuged to separate cells (P, pellet) from the interspore matrix (S, supernatant). 106 cell equivalents of each were analysed by SDS-PAGE and Western blot analysis, using either mAb 9E10 against the internal c-myc epitope, pAb NT7 against the N domain or pAb CT7 against the C1 and C2 domains. The positions of NC1, NC2 and breakdown products (N and C1) are indicated in the margin. Lanes 1–4 had been simultaneously labelled with mAb 83.5, explaining the additional weakly labelled bands. (b, c) Strains expressing NC1 and mutant NC1s, and strains Ax3 and HW70 (SP85-null) were compared similarly. Blots were probed with mAb 83.5 against the Fuc-epitope on SP96 and SP75 (b) or mAb 5F5 against the N- domain of SP85 (c). Positions of proteins are shown in the margin. Letters B–E in strain names correspond to the C4C motifs shown in Table 2.

 
Most of the expression constructs caused loss of electron-dense material from the outer layer. A phenotype similar to that of strain TL56 was seen when NC2 was expressed, resulting in an incomplete outer layer containing a residual array of electron densities with a mean spacing of 36 nm (Fig. 3f). A similar effect occurred when NC2 was expressed in the SP85-null strain HW70 (Fig. 3g). The least effect was seen in spores expressing NC1C2 (Fig. 3c), which included the largest number of SP85 domains. A strong effect was observed when N was expressed, resulting in the appearance of dot-like structures and small patches where the outer layer appeared to be absent (Fig. 3d). Spores expressing C1C2 had mosaic coats where regions of the outer layer showed punctate electron densities and other regions exhibited filamentous projections as described below for another construct (Fig. 3e). These filament-like structures were also occasionally seen in NC1C2 spores. In general, the differences shown in Fig. 3 were observed throughout the population of spores examined and correlate with increased macromolecular permeabilities measured previously (Zhang et al., 1999). In summary, defects as severe as those seen in the absence of SP85 were observed when SP85 fragments were overexpressed. Though it is not known whether the effects can be attributed to the exact construct expressed or breakdown products, the results clearly implicate active roles for the N and C2 domains in SP85 function.

A second kind of defect was seen in spores expressing the NC1 domain fusion. Only about 15 % of the prespore cells of this strain deposit cellulose and complete sporulation because of failure to execute an early checkpoint in coat formation (West et al., 2002). Images from the mature spores that formed revealed an unusual mat of novel, filamentous projections at their surfaces (Fig. 5b, c). The filaments replaced the normally continuous electron density of the outer layer and extended outwardly approximately one thickness of the coat. They had apparent diameters on the order of 10 nm, similar to the apparent thickness of the outer layer. The filamentous mat covered most of the surface of the spore (Fig. 6a). Occasional spores had an attenuated or missing middle layer but still exhibited the modified outer layer (data not shown), indicating that the filaments did not depend on the full deeper layers of the coat. The filamentous mat was also formed when NC1 was expressed in SP85-null HW70 cells (Fig. 5d), showing that filaments do not require SP85 and are not the result of checkpoint delay which does not occur in this strain (West et al., 2002). The filaments appear to contain coat proteins because they were immunogold-labelled with mAb 83.5 (Fig. 6c), which recognizes outer-layer proteins (Fig. 6b).



View larger version (79K):
[in this window]
[in a new window]
 
Fig. 5. Electron microscopy of spores expressing normal and mutant NC1 domain fusions. (a) Normal strain Ax3. (b–d) Examples of expression of NC1 in strain Ax3 (b, c) or the SP85-null strain HW70 (d). (e) Expression of mutant NC1(B) in Ax3. (f, g) Expression of the quadruple mutant NC1(B–E) in Ax3.

 


View larger version (184K):
[in this window]
[in a new window]
 
Fig. 6. Electron microscopy of spores expressing the NC1 domain fusion. (a) Section of complete spore showing filamentous mat around its entire perimeter. (b, c) Immunogold labelling of sections of Ax3 (b) or ANC1 (c) cells with mAb 83.5 against the Fuc-epitope normally associated with the outer layer.

 
The filamentous layer was also seen variably on spores that express the C1C2 domain fusion (data not shown). These spores were often mosaic with filaments on one side with the remainder of the spore exhibiting a coat with decreased electron density, suggesting that filaments did not depend on the presence of the N domain in the NC1 fusion. Occasional filaments were also seen in spores expressing NC1C2, but their frequency and extent were greatly reduced relative to NC1-expressing spores.

Site-directed mutagenesis of the NC1 domain fusion
The C1 domain consists of four tandem repeats of the C4C motif (Fig. 3a), which is about 25 aa in length and predicted to fold with two disulfide bonds (Table 2), based on sequence similarity to the N-terminal subdomain of the EGF motif (West, 2002). In an attempt to dissect the contributions of each C4C motif to these functions, the third predicted loop of each motif was subjected to site-directed mutagenesis in which one to two residues on each side of the predicted {beta}-turn in the middle of the loop were modified (Table 2). These mutations were designed to alter the chemical character of the loops without disrupting global folding of the motif (Handford et al., 2000). Each mutant construct, NC1(B), NC1(C), NC1(D) and NC1(E), and a quadruple mutant NC1(B–E) in which all four motifs were mutagenized simultaneously, was expressed in strain Ax3 and clones expressing high levels of the NC1 mutant proteins were chosen for further analysis.


View this table:
[in this window]
[in a new window]
 
Table 2. NC1 mutations

The C4C motif is predicted to fold as shown on the right. Normal (wild-type) sequences of each C4C motif, from Cys3 to Cys4 (in bold), are shown above the mutant sequences created (underlined). The putative {beta}-turn residues are also in bold.

 
Each mutant NC1 protein was expressed at a level comparable to that of normal NC1 based on Western blot analysis of isolated spore-like cells using mAb 5F5 directed against the N-terminal domain of SP85 (Fig. 4c). The NC1s had similar apparent Mr values relative to one another, were incorporated into coats as efficiently as normal NC1 and appeared to be expressed at substantially higher levels than SP85 (Fig. 4c). A similar degree of cleavage of NC1 as described above occurred for each of the mutants. The levels of expression of the coat proteins SP96 and SP75 were relatively normal (Fig. 4b). However, in the SP85-null strain HW70, included in this analysis, a novel band accumulated in the interspore matrix that appeared to replace SP96 in the coat, exactly as observed previously for SP96 in the strain expressing NC2 (West et al., 2002).

Strains expressing mutant NC1s formed fruiting bodies with 30–60 % of the normal number of cellulose-positive spores, compared to the 10 % produced by normal NC1-expressing cells in these experiments (Table 3). This indicated that the mutations partially inactivated the inhibitory effects of NC1 on checkpoint execution, showing that each of the four C4C motifs contributes to its previously described checkpoint blockade activity (West et al., 2002).


View this table:
[in this window]
[in a new window]
 
Table 3. Characteristics of mutant spores

 
Electron microscopy showed that NC1 mutants altered in any one of the C4C motifs still efficiently induced the outer-layer mat of filaments [NC1(B) is shown in Fig. 5e]. In contrast, spores expressing NC1(B–E), simultaneously mutagenized in all four C4C motifs, failed to form the filamentous mat. Instead, coats had irregular outer layers with uneven electron density and occasional stunted filaments (Fig. 5f, g). Therefore the filament-inducing activity of NC1 depends on the amino acid sequence of the C1 domain and is separate from the checkpoint blockade activity of NC1.

Cellulose-binding activities of normal and mutant NC1s
A recombinantly expressed version of the C region, consisting of the C1 and C2 domains, was previously shown to bind cellulose in vitro (Zhang et al., 1998). To determine whether binding maps to the C1 or C2 domain, the binding activities of the NC1 and NC2 constructs were compared. Cells expressing the domain fusions in a dcsA-null background (see Table 1), to eliminate endogenous cellulose, were sporulated in suspension in the presence of 20 mM 8-Br-cAMP and the supernatants (cA-ISM) were used as a source of the NC1 and NC2 proteins. The cA-ISMs were incubated in the presence of Avicel cellulose as described in Methods and separated into bound and unbound fractions by centrifugation. As shown in Fig. 7(a), the majority of NC1 bound to Avicel. In contrast, very little binding of NC2 was detected. Similar results were observed when the domain fusions were expressed in SP85-null cells (Fig. 7a) or normal Ax3 cells (see below), showing that it was not necessary to express the protein in the absence of cellulose and that binding did not depend on any SP85 in these preparations.



View larger version (45K):
[in this window]
[in a new window]
 
Fig. 7. Cellulose-binding activities of normal and mutant domain fusions in vitro. (a) dcsA (cellulose synthase)-null or SP85 (B1)-null strains expressing NC1 or NC2 were induced to sporulate in suspension by 8-Br-cAMP as described in Methods. The secretory fraction (cA-ISM) was incubated with Avicel cellulose particles for 16 h. Bound (B) and non-bound (NB) fractions were separated by centrifugation, loaded onto an SDS-PAGE gel, Western-blotted and probed with pAb NT7 against the N domain of SP85. Strain names are explained in the Table 1 legend. (b) Secretory fractions from the normal strain Ax3 and Ax3 cells expressing NC1, NC1(B) or NC1(B–E) were tested for binding to Avicel as in (a) (lanes 1 and 2). An aliquot of the unbound fraction (shown in lane 1) was reincubated with a second batch of Avicel and binding assayed again (lanes 3 and 4). NC1(B–E) does not bind Avicel as well as NC1 and NC1(B); residual SP85 in all fractions bound Avicel efficiently; and an N-fragment that appeared in some cases did not. The blot was probed with mAb 5F5.

 
To test the Avicel-binding activities of the NC1 mutants, which are expressed in Ax3 cells, cA-ISM fractions were isolated as above. Since it was observed above that not all NC1 in the preparation was pulled down with Avicel (Fig. 7a), a tandem assay was performed in which the unbound fraction from the first round (Fig. 7b, lanes 1) was reincubated in a second round with a fresh batch of Avicel and then centrifuged again (lanes 3 and 4). Under these conditions, the non-bound NC1 from the first binding was found to bind quantitatively to Avicel in the second round. As internal controls, trace amounts of SP85 present in these preparations bound Avicel efficiently in the first round, and a breakdown product of NC1 consisting of the N domain did not bind as seen previously (Zhang et al., 1999). Each of the singly mutated NC1 proteins bound Avicel as efficiently as NC1, as shown for NC1(B) (Fig. 7b). In contrast, the quadruple mutant showed inefficient binding in both rounds. This suggests that multiple C4C motifs contribute to cellulose binding, and establishes a correlation between weak cellulose binding and inability to induce the filamentous mat at the outer layer.


   DISCUSSION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Outer-layer structure in the absence of critical coat proteins
Electron microscopy of mutant spores lacking SP96, SP70 and SP60 shows a dramatic loss of outer-layer material (Fig. 1c). This correlates well with prior evidence that these proteins are associated with the outer layer and with the increased permeability to macromolecular tracers of coats lacking these proteins (Fosnaugh et al., 1995; Srinivasan et al., 2000). Mutant coats exhibit a residual array of electron-dense dots with a spacing of about 35 nm (Fig. 1d, e), a structural motif that has not been previously observed. The accumulation of dots in strain TL56 suggests that outer-layer formation is a two-step process consisting of assembling the dots and in-fill with the coat proteins SP96, SP70 and SP60. Earlier work with single gene disruptions suggested that SP96, SP70 and SP60 each contribute incrementally to outer-layer function (Fosnaugh et al., 1995; Srinivasan et al., 2000) and so it will be interesting to examine the contributions of each of these proteins at the ultrastructural level.

The coats from HW70, lacking SP85, also exhibit loss of electron density from the outer layer (Fig. 2b). The residual material does not show as clear a periodic pattern as TL56, but grazing sections suggest the existence of residual strands (Fig. 2d). The similar spacing of these densities with those of TL56 coats suggests that they may be related structures. This is supported by the appearance of periodic dots when NC2 is expressed in SP85-null spores (see below), suggesting that in SP85- spores the dots may be obscured by additional material. SP85-null coats are missing SP65 (Zhang et al., 1999) and the outer-layer protein SP96 (Figs 2e and 4b), each of which is recovered in the interspore matrix. Failure of SP85-null spores to incorporate SP96 suggests that the same end point, in-fill of outer-layer proteins, is inhibited and implies that a hierarchy of protein–protein interactions involving SP85 and SP65 is required for incorporation of SP96. The less severe defect seen in SP85- compared to TL56 outer layers correlates with the less severe permeability defect observed in SP85- spores, which required pre-treatment with urea and reducing agent to detect (Zhang et al., 1999).

Role of SP85 in outer-layer formation
The outer layer is also altered when fragments and domain fusions of SP85 fragments are expressed. NC2-expressing spores had an outer-layer morphology similar to that of TL56 spores with the appearance of a regular array of similarly spaced punctate dots (Fig. 3). Consistent with this interpretation, cells expressing NC2 fail to incorporate the outer-layer protein SP96 into the coat and instead accumulate a truncated isoform of SP96 in the interspore matrix (West et al., 2002), as now also observed for SP85-null spores (Fig. 4b). Spores expressing the N domain are missing patches of the outer layer altogether. This correlates with other defects including increased coat permeability, decreased coat buoyant density and more rapid germination, as reported previously for this strain (Zhang et al., 1999). The mechanism of action of N may involve binding to a critical distal component as N is not incorporated into the mature coat and is recovered in the interspore matrix (Zhang et al., 1999). Coats of spores expressing C1C2 exhibit a mosaic character, with some regions resembling NC2. Thus the effect of NC2 does not seem to obligatorily depend on the presence of the N domain, or the accumulation of free N domains observed when NC2 was overexpressed. Spores expressing C1C2 also showed increased permeability, decreased buoyant density and more rapid germination (Zhang et al., 1999). The NC1C2 domain fusion was least severe, consistent with the greatest similarity to full-length SP85 and this correlated with the least severe effect of this construct on coat functions (Zhang et al., 1999). These results are consistent with a cross-bridging model for normal SP85 function in which discrete domains bind other components in the coat, thereby linking them together. Overexpression of individual domains would be expected to compete with endogenous SP85 for ligands, resulting in a failure to cross-bridge them into the inter-dot regions of the outer layer. The somewhat more severe effects of expressing the domains compared to deletion of SP85 suggest the existence of mechanisms that compensate for the absence of SP85, which are also effectively competed by the domain constructs.

In contrast to the other mutant constructs, NC1 expression led to the formation of a novel structure. This consists of a thick mat of filamentous material which replaces the normal continuous outer layer (Figs 5 and 6). The filaments may represent a displaced normal coat subunit. The only linear structures known in the coat are cellulose microfibrils. We have been unsuccessful in labelling these filaments using cellulase-gold probes (C. M. West and others, unpublished data), but the sensitivity was low and number of microfibrils per filament may be small. A coat protein epitope is associated with the filaments (Fig. 6c) though it cannot be determined if this represents the core or a coating. Instead of cellulose, the filaments might be formed of protein or another polysaccharide. SP85 is not required for the filaments as the double mutant (expression of NC1 in SP85-null cells) also expresses them, albeit with altered appearance. NC1 might seed the ectopic formation of filamentous structures at the outer layer, as a result of improper localization due to the absence of a critical localizing domain present in full-length SP85. Elsewhere, fibres with radial orientation have been implicated in the formation of cell walls in Chlamydomonas and Cryptococcus (Woessner & Goodenough, 1994; Pierini & Doering, 2001).

Involvement of cellulose-binding activity in SP85 function
Previously, a recombinant version of the C1C2 domain fusion was found to bind cellulose in vitro (Zhang et al., 1998) and a comparison of the NC1 and NC2 domain fusions here shows that this activity maps to the C1 domain (Fig. 7). The C1 domain contains four Cys-rich C4C motifs. Based on the hypothesized similarity to the N-terminal subdomain of the EGF-module (Table 2), the sequence between the third and fourth conserved Cys-residues is predicted to form a loop with a centrally positioned Gly (or similar) residue forming a critical {beta}-turn, possibly analogous to the conformation of the RGD motif of fibronectin (Tani et al., 2002). Residues adjacent to the Gly residue were altered in each C4C motif individually and also simultaneously in a fifth construct in an attempt to disrupt cellulose-binding. Although mutagenesis of any of the C4C motifs interfered with the ability of NC1 to block the early checkpoint in cellulose deposition, only the quadruple mutation interfered with cellulose binding. This was also the only mutant that affected the filament-inducing activity of NC1. Although it is not known if cellulose binding is the only activity affected by these mutations, as C1C2 also binds SP65 via a site separate from cellulose binding (Zhang et al., 1999), this correlation suggests that cellulose binding is important for NC1 action. If the above model that NC1 nucleates or anchors naturally occurring filaments in an ectopic location is correct, then the apparent dependence on cellulose binding could be explained if cellulose is the basis for NC1 attachment to the coat, or if the filaments contain cellulose to which NC1 binds. In plant walls, cellulose fibrils are thought to be anchored by hemicelluloses (Delmer, 1999; Brett, 2000), but additional mechanisms are likely to be involved especially in walls like the spore coat that lack this polysaccharide (West, 2002). In any case, the result implicates cellulose binding in the contribution of SP85 to outer-layer formation.

Relationship of outer-layer formation to earlier events in coat formation
Secretion of coat protein precursors is thought to be the initiating event of the extracellular phase of coat assembly (West, 2002). An early function of SP85 is to prevent cellulose synthesis until coat proteins are secreted, hydrated and possibly reorganized to form a precoat (West et al., 2002). The outer layer itself does not appear until cellulose deposition is initiated (Zhang et al., 2001). Since SP85 contributes in an essential way to outer-layer formation and the dominant negative effect of NC1 depends on structural motifs also required for efficient cellulose binding, it is likely that the role of cellulose in outer-layer formation is transmitted at least in part through SP85. The contribution of SP85 to outer-layer formation is expected to be indirect because it is localized in the inner layer and is not long enough to stretch across the estimated 160 nm thickness of the middle layer of the coat. If the 108 aa of the TR regions fold as left-handed polyproline type II helices (Kelly et al., 2001), they would have a combined length of 35 nm, and the remaining 405 residues, which are likely to fold as globular domains, are unlikely to contribute more than this length again. As summarized above, SP85 appears to contribute to the outer layer via a hierarchy of intermolecular contacts. Based on previous studies (Zhang et al., 1999) and results shown here, a cellulose-SP85-SP65-SP96 linkage might compose the molecular backbone that links elements of the outer layer to the coat (West, 2002). The C1 domain mediates cellulose binding (Fig. 7), and appears to also bind SP65 (C. M. West and others, unpublished data) via a separate site, and the C2 domain may mediate SP96 binding based on the effects of NC2 on SP96 incorporation (Fig. 4b). SP70 and SP60, previously suggested to interact with SP85 (McGuire & Alexander, 1996), might also be anchored via this proposed module. The filaments seen in NC1-expressing spores might represent misplaced copies of this module or, alternatively, reflect an additional function for SP85 in the packaging of cellulose microfibrils, which are disorganized in these mutants based on abnormal sensitivity to acid hydrolysis (West et al., 2002).

The previously described contribution of SP85 to early checkpoint regulation (West et al., 2002) appears to be separate from its outer-layer function. First, the early effect of NC1 on checkpoint function does not depend on cellulose or its cellulose-binding activity, whereas its ability to induce outer-layer filaments seems to depend on its cellulose-binding activity. Second, NC1 mutations that affect checkpoint execution do not affect filament induction (Table 3). Third, the effect of NC1 on checkpoint execution depends on endogenous SP85, whereas induction of filaments does not (Fig. 5d). Finally, some of the SP85 fragment constructs have other effects on the outer layer without altering the timing of checkpoint execution.


   ACKNOWLEDGEMENTS
 
We are grateful to Scherwin Henry of the UF ICBR Hybridoma Core Laboratory for preparing the monoclonal antibodies, and Dr W. F. Loomis for providing strain TL56. These studies were supported in part by a grant (MCB-9730036) from the National Science Foundation.


   REFERENCES
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Aparicio, J. G., Erdos, G. W. & West, C. M. (1990). Spore coat is altered in modB glycosylation mutants of Dictyostelium discoideum. J Cell Biochem 42, 255–266.[Medline]

Bayer, E. A., Shimon, L. J., Shoham, Y. & Lamed, R. (1998). Cellulosomes – structure and ultrastructure. J Struct Biol 124, 221–234.[CrossRef][Medline]

Blanton, R. L., Fuller, D., Iranfar, N., Grimson, M. J. & Loomis, W. F. (2000). The cellulose synthase gene of Dictyostelium. Proc Natl Acad Sci U S A 97, 2391–2396.[Abstract/Free Full Text]

Boulianne, R. P., Liu, Y., Aebi, M., Lu, B. C. & Kues, U. (2000). Fruiting body development in Coprinus cinerus: regulated expression of two galectins secreted by a non-classical pathway. Microbiology 146, 1841–1853.[Abstract/Free Full Text]

Brett, C. T. (2000). Cellulose microfibrils in plants: biosynthesis, deposition, and integration into the cell wall. Int Rev Cyt 199, 161–199.

Delmer, D. P. (1999). Cellulose biosynthesis: exciting times for a difficult field of study. Annu Rev Plant Physiol Plant Mol Biol 50, 245–276.[CrossRef]

Erdos, G. W. & West, C. M. (1989). Formation and organization of the spore coat of Dictyostelium discoideum. Exp Mycol 13, 169–182.

Fosnaugh, K. L., Fuller, D. & Loomis, W. F. (1995). Structural roles of the spore coat proteins in Dictyostelium discoideum. Dev Biol 166, 823–825.[CrossRef]

Frisardi, M., Ghosh, S. K., Field, J., VanDellen, K., Rogers, R., Robbins, P. & Samuelson, J. (2000). The most abundant glycoprotein of amebic cyst walls (Jacob) is a lectin with five Cys-rich, chitin-binding domains. Infect Immun 68, 4217–4224.[Abstract/Free Full Text]

Gonzalez-Yanes, B., Mandell, R. B., Girard, M., Henry, S., Aparicio, O., Gritzali, M., Brown, R. D., Erdos, G. W. & West, C. M. (1989). The spore coat of a fucosylation mutant in Dictyostelium discoideum. Dev Biol 133, 576–587.[Medline]

Handford, P. A., Downing, A. K., Reinhardt, D. P. & Sakai, L. Y. (2000). Fibrillin: from domain structure to supramolecular assembly. Matrix Biol 19, 457–470.[CrossRef][Medline]

Harb, O. S., Gao, L.-Y. & Abu Kwaik, Y. (2000). From protozoa to mammalian cells: a new paradigm in the life cycle of intracellular bacterial pathogens. Environ Microbiol 2, 251–265.[CrossRef][Medline]

Kelly, M. A., Chellgren, B. W., Rucker, A. L., Troutman, J. M., Fried, M. G., Miller, A.-F. & Creamer, T. P. (2001). Host–guest study of left-handed polyproline II helix formation. Biochemistry 40, 14376–14383.[CrossRef][Medline]

Loomis, W. F. (1971). Sensitivity of Dictyostelium discoideum to nucleic acid analogues. Exp Cell Res 64, 484–486.[Medline]

Maeda, M. (1992). Efficient induction of sporulation of Dictyostelium prespore cells by 8-bromocyclic AMP under both submerged- and shaken-culture conditions and involvement of protein kinase(s) in its action. Dev Growth Differ 34, 263–275.

Mateos, F. V., Rickauer, M. & Esquerre-Tugaye, M.-T. (1997). Cloning and characterization of a cDNA encoding an elicitor of Phytophthora parasitica var. nicotianae that shows cellulose-binding and lectin-like activities. Mol Plant Microbe Interact 10, 1045–1053.[Medline]

McDowell, E. M. & Trump, B. F. (1976). Histologic fixatives suitable for diagnostic light and electron microscopy. Arch Pathol Lab Med 100, 405–414.[Medline]

McGuire, V. & Alexander, S. (1996). PsB multiprotein complex of Dictyostelium discoideum: demonstration of cellulose binding activity and order of protein subunit assembly. J Biol Chem 271, 14596–14603.[Abstract/Free Full Text]

Nakao, H., Yamamoto, A., Takeuchi, I. & Tasaka, M. (1994). Dictyostelium prespore-specific gene (DP87) encodes a sorus matrix protein. J Cell Sci 107, 397–403.[Abstract/Free Full Text]

Orlowski, M. & Loomis, W. F. (1979). Plasma membrane proteins of Dictyostelium: the spore coat proteins. Dev Biol 71, 297–307.[Medline]

Pierini, L. M. & Doering, T. L. (2001). Spatial and temporal sequence of capsule construction in Cryptococcus neoformans. Mol Microbiol 41, 105–115.[CrossRef][Medline]

Sassi, S., Sweetinburgh, M., Erogul, J., Zhang, P., Teng-umnuay, P. & West, C. M. (2001). Analysis of Skp1 glycosylation and nuclear enrichment in Dictyostelium. Glycobiology 11, 283–295.[Abstract/Free Full Text]

Simrell, C. R. & Klein, P. A. (1979). Antibody responses of tumor-bearing mice to their own tumors captured and perpetuated as hybridomas. J Immunol 123, 2386–2394.[Medline]

Srinivasan, S., Griffiths, K. R., McGuire, V., Champion, A., Williams, K. L. & Alexander, S. (2000). The cellulose-binding activity of the PsB multiprotein complex is required for proper assembly of the spore coat and spore viability in Dictyostelium discoideum. Microbiology 146, 1829–1839.[Abstract/Free Full Text]

Tani, P. H., Loftus, J. C. & Bowditch, R. D. (2002). In vitro selection of fibronectin gain-of-function mutations. Biochem J 365, 287–294.[CrossRef][Medline]

Wang, W. & Malcolm, B. A. (1999). Two-stage PCR protocol allowing introduction of multiple mutations, deletions and insertions using QuikChange site-directed mutagenesis. Biotechniques 26, 680–682.[Medline]

Wang, Y., Slade, M. B., Gooley, A. A., Atwell, B. J. & Williams, K. L. (2001). Cellulose-binding modules from extracellular matrix proteins of Dictyostelium discoideum stalk and sheath. Eur J Biochem 268, 4334–4345.[Abstract/Free Full Text]

West, C. M. (2002). Comparative analysis of spore coat formation, structure and function in Dictyostelium. Int Rev Cyt 222, 237–293.

West, C. M. & Erdos, G. W. (1988). The expression of glycoproteins in the extracellular matrix of the cellular slime mold Dictyostelium discoideum. Cell Differ 23, 1–16.[CrossRef][Medline]

West, C. M., Mao, J., van der Wel, H., Erdos, G. W. & Zhang, Y. (1996). SP75 is encoded by the DP87 gene and belongs to a family of modular Dictyostelium discoideum outer layer spore coat proteins. Microbiology 142, 2227–2243.[Abstract]

West, C. M., Zhang, P., McGlynn, A. C. & Kaplan, L. (2002). Outside–in signaling of cellulose synthesis by a spore coat protein in Dictyostelium. Euk Cell 1, 281–292.[CrossRef]

Woessner, J. P. & Goodenough, U. W. (1994). Volvocine cell walls and their constituent glycoproteins: an evolutionary perspective. Protoplasma 181, 245–258.

Zhang, P., McGlynn, A. C., Loomis, W. F., Blanton, R. L. & West, C. M. (2001). Spore coat formation and timely sporulation depend on cellulose in Dictyostelium. Differentation 67, 72–79.[CrossRef][Medline]

Zhang, Y., Brown, R. D. & West, C. M. (1998). Two proteins of the Dictyostelium spore coat bind to cellulose in vitro. Biochemistry 37, 10766–10779.[CrossRef][Medline]

Zhang, Y., Zhang, P. & West, C. M. (1999). A linking function for the cellulose-binding protein SP85 in the spore coat of Dictyostelium discoideum. J Cell Sci 112, 4367–4377.[Abstract/Free Full Text]

Received 10 September 2002; revised 6 November 2002; accepted 11 November 2002.