Stephen M. Wolniak
Department of Cell Biology &
Molecular Genetics

Formation of the Cytoskeleton and Motile Apparatus in Spermatids of Marsilea
University of Maryland  
College Park, Maryland 20742    

Introduction | Published Results | Ongoing Experiments
Recent publications from the Wolniak laboratory

Marsilea vestita is a water fern (Figure 1) that grows as a weed at the edges of vernal pools or in rice paddies. By all outward appearances, the plant resembles a large four-leafed clover (Figure 2), but in reality, it is only distantly related to the flowering plants. The meiotic products from the sporophyte, the microspores and megaspores, are found within hardened structures known as sporocarps (Figure 3). After being placed into water, the dry microspores develop into male gametophytes and the megaspores develop into female gametophytes. Fertilization in lower plants involves the fusion of a motile male gamete, the spermatozoid, with a non-motile female gamete that is encased in a flask-shaped array of vegetative cells known as an archegonium. Plant spermatozoids are structurally complex cells (Figure 1) that differentiate within an antheridium. The release of certain chemicals (i.e., pheromones) from female gametes stimulate changes in spermatozoid swimming behavior and direction; these cues effectively guide the spermatozoids toward the egg. This change in swimming behavior is known as chemotaxis, a word that was coined for the response of fern spermatozoids to chemicals present in egg exudates [Pfeffer, 1884].

Almost 90 years ago, Lester Sharp [1912] described the rapid process of spermatozoid formation in the male gametophytes of M. vestita. He found that the placement of dry spores into water would result in sperm formation; the process reached completion in about eleven hours with the release of 32 of these spirally shaped spermatozoids. The development of the male gametophyte occurs entirely within the microspore wall. Sharp tracked the process microscopically and determined that during the first 5.5 hours, there were nine mitotic divisions, which occurred in precise division planes to produce a total of 39 cells (Figure 4). When the division cycles are complete, the gametophyte consists of one prothallial cell, six sterile jacket cells and 32 spermatids. Then, each of the 32 spermatid cells differentiates into a motile spermatozoid. Each spermatozoid had ~140 cilia. Midway through the process, and before the last two division cycles are complete, Sharp observed that the appearance of a particle known as the blepharoplast (i.e., eyelash [Webber, 1899]) preceded sperm elongation and assembly of the cilia. Sharp saw the blepharoplast initially as a dark dot in the cells. In 1976, Peter Hepler provided high resolution images of the blepharoplast (Figure 5), and demonstrated that this spherical structure was the site of basal body synthesis. Basal bodies serve as the templates for ciliary and flagellar axonemes. This process is important because in animal cells, basal bodies are formed in close association with centrioles. Plant cells lack centrioles. The process of blepharoplast formation provides a unique opportunity to study basal body formation in a cell where preexisting centrioles are lacking, and a situation where the process is triggered at a precise time, by an easily repeated process (placing spores into water). Blepharoplasts form 4 hours after the spores are placed into an aqueous medium. During the last half of gametophyte maturation, the development of the spermatids centers on formation of a complex cytoskeletal array known as a multilayered structure (MLS, Figure 6) , the elongation and coiling of the cell body, nuclear elongation and chromatin packing, and finally, ciliogenesis. The structural aspects of this complicated morphogenetic process have been documented by Myles and Hepler (1977).

Top of Page

Published Results

Studies recently initiated in my laboratory have focused on the identity of some of the proteins that are present in the blepharoplast, and we have extended these findings in experiments designed to test whether the synthesis of these components is necessary and sufficient for the formation of basal bodies. Marsilea gametophytes provide a unique opportunity to study de novo basal body formation in populations of synchronously developing populations of cells. The process is rapid and is activated by the simple addition of water. Marsilea vestita is not a model system in the standard sense, where genetic approaches permit the generation of organisms exhibiting permanent shifts in structure, development, or in physiological capacity or response, so we have developed a variety of alternative strategies and probes to study this process. We have devised protocols to isolate mRNA from these gametophytes and we have made a cDNA library from gametophytes at all stages of development [Hart and Wolniak, 1999]. We have devised immunolabeling protocols for in situ detection of antigens in cells, for identification of polypeptides isolated from the gametophytes [Klink and Wolniak, 2001]. In parallel, we have devised protocols for the detection of specific mRNAs in cells, by using in situ hybridization in our plastic-embedded sections of gametophytes [Tsai and Wolniak, 2001]. Many aspects of the technique development phase in this project were time-consuming processes. We performed a random screen of our cDNA library and have now sequenced and analyzed several hundred cDNA inserts. Approximately 50% of these cDNAs represent known gene products that are involved in cell cycle regulation, spindle and cytoskeletal organization, and cell motility. Thus, we have established a species-specific gene bank to study spermiogenesis in this organism. We have developed and employed a series of antisense and RNAi strategies and technologies with Marsilea gametophytes [Klink and Wolniak, 2000, 2001].

For the past several years, we have focused specifically on identifying what components are necessary for blepharoplast formation and the subsequent assembly of basal bodies. Our initial experiments [Hart and Wolniak, 1998] revealed that large quantities of components used in the formation of the motile apparatus are stored as proteins that are present in the dry spore (Figure 7). We did not see any distinct differences in protein isolates from gametophytes treated with a-amanitin (Figure 7, an inhibitor or transcription) or cycloheximide (Figure 7, an inhibitor of translation). Prime examples of stored proteins are the tubulins, which represent the bulk of the fiberous portion of the cytoskeleton and the ciliary axonemes; in both structures, the tubulin subunits are assembled into microtubules. New ß-tubulin is synthesized in these cells only late in spermiogenesis, when the spermatids are engaged in ciliary axoneme formation. We have also found that large quantities of stored mRNA are also present in the dry spore. This mRNA is used for the translation of new proteins (Figure 8). Remarkably, there is little if any transcription in the gametophyte during its 11 hour developmental cycle; development progresses quite far in the presence of a-amanitin, a transcriptional inhibitor (Figure 9 and Figure 10). In contrast, in the absence of translation brought about by treatment with cycloheximide (Figure 9 and Figure 10), there is little discernable development of the gametophyte. Thus, development of the gametophyte relies on the translation of stored mRNAs. One of these proteins made from stored mRNA is centrin (Figure 11), which functions in the blepharoplast (Figure 12) and elsewhere to orchestrate patterns and extents of macromolecular assembly of proteins like the tubulins. We found that centrin is made from stored mRNA (Figure 11) and that the abundance of the protein increases dramatically at the time the blepharoplast appears in the cells [Hart and Wolniak, 1998, 1999] (Figure 12). We isolated a cDNA from our library that encodes a legitimate centrin that we have named MvCen1 (Figure 13). This predicted protein compares well with known centrins from a variety of eukaryotes (Figure 13). We developed the working hypothesis that the timing of centrin synthesis is directly coupled with the activity of centrin in making blepharoplasts form.

Top of Page

By employing RNAi strategies, we have found that centrin translation is indeed necessary for the formation of the blepharoplast [Klink and Wolniak, 2001]. Antisense centrin RNA blocks development at a stage when the blepharoplast should be present, but blepharoplasts are not detectable in these cells. Sense centrin-RNA inhibits development, because of RNAi (RNA interference) effects, as shown with dsRNA, which is inhibitory at significantly lower concentrations (Figure 14). It is clear that we have induced a form of post-transcriptional gene silencing, which blocks progression in spermatid development in a stage-specific fashion, through the targeted destruction of centrin mRNA in the gametophyte (Figure 15 and Figure 16). The concentration-dependence of inhibition using centrin-derived RNA probes strongly supports the notion that our treatments of cells with centrin antisense RNA or centrin dsRNA are gene-specific blocks to centrin translation, because these probes target the destruction of centrin mRNA (Figure 15 and Figure 16). In control experiments, we have been able to determine that ß-tubulin antisense RNA has no effect on development until late in spermiogenesis, when ß-tubulin translation is required for ciliary growth. We have been able to show that an irrelevant RNA, a noncoding sequence from HIV, when added to our cells as an antisense or dsRNA construct, exerts absolutely no effect on spermiogenesis (Figure 15 and Figure 16). Immunoblot analyses of centrin protein production in the presence of any of these RNA probes reveals that the block to centrin translation only occurs when antisense or dsRNA from centin is added to the gametophytes (Figure 14).

We next asked if centrin translation from stored mRNA is dependent on cell division in the gametophytes [Tsai and Wolniak, 2001]. We blocked cell proliferation in the gametophytes in three independent ways, using pharmacological treatments with hydroxyurea or olomoucine, or by RNAi treatments with dsRNA derived from cyclin B, or dsRNA from cyclin A (Figure 17). Divisions are completely arrested by hydroxyurea treatments administered at the time of imbibition (Figure 17). In contrast, one or two divisions with a few incomplete cell plates resulted from treatments with olomoucine (Figure 17), a cyclin-dependent kinase inhibitor. Isolomoucine was without effect on divisions or development of the gametophytes (data not presented). dsRNA probes for Marsilea cyclin A or cyclin B block mitosis in the gametophyte; antisense or dsRNA for cyclin A allow one or two divisions, while similar probes made from cyclin B arrest divisions completely (Figure 17). None of these treatments forestalls or limits centrin translation, as depicted on western blots obtained from protein isolates of cell populations harvested at 2 h intervals during development (Figure 18), but in the absence of mitosis, we have been unable to detect blepharoplasts. Thus, centrin translation appears to be necessary but not sufficient for blepharoplast formation. Cytoplasmic partitioning appears important for organizing components that assemble into the blepharoplast as these cells acquire an organized centrosome and initiate de novo basal body assembly.

Top of Page

Ongoing Experiments

Through the largesse of active members in the cytoskeletal, centrosomal, and axonemal research communities, we have obtained a number of different antibodies that bind to antigens present in our developing spermatids. We have expanded on our immunoblot analyses [Hart and Wolniak, 1998], looking for changes in the abundance of some 70 antigens in protein isolates obtained at different stages during spermiogenesis (Klink and Wolniak, unpublished observations). In contrast to alpha-, beta- and gamma-tubulin, which are abundant as proteins in the dry spores, but in a fashion similar to centrin, the translation of p28, RanBPM and Xgrip109 from stored mRNA precedes or accompanies blepharoplast appearance. Thus, we have additional candidate proteins whose synthesis parallels centrin translation patterns; these candidates will be the subjects for further analysis of blepharoplast composition and formation.

We have expanded our research interests in new directions, asking how a single cell can rapidly undergo a series of division cycles, and produce two distinct kinds of cells [Tsai et al., 2004]. The division cycles in the gametophyte ultimately give rise to nine sterile cells and 32 spermatids. Two clusters of 16 spermatids are formed from the two spermatogenous initials, which differ compositionally and structurally from the adjacent sterile jacket cells that surround them. When we began to look at patterns of stored protein accumulation, and new translation, it was not particularly surprising to find high levels of stored tubulin and new centrin protein in the spermatogenous cells, but largely absent from the sterile cells. We began to look at the movement and distribution of proteins (with immunolabeling assays) and stored mRNAs (with in situ hybridization assays) into the cellular domains that would later become spermatogenous initials. Surprisingly, the distribution of most stored mRNAs does not match that of the proteins (Figure 19 and Figure 20). Most of the mRNAs are present in all of the cells of the gametophyte (Figures 19, 20, 21), while the stored proteins like ß-tubulin (Figure 19) or newly translated proteins like centrin (Figure 20) become exclusively localized in the spermatogenous initials. We were surprised to find that cyclin B mRNA was present in all cells of the gametophyte (Figure 21), though the jacket cells, once formed, cease to divide. We were able to find several mRNAs that exhibit nonrandom distributions in spermatogenous cells; two of these transcripts encode a PRP-19 like protein and an RNA helicase (Figure 22). RNAi experiments using dsRNAs made from either of these transcripts results in anomalous division patterns in the gametophyte (Figure 22). We found that patterns of polyadenylation match the distribution of transcription patterns, mostly in spermatogenous cells (Figure 23), and we also showed that the cytoplasmic polyA polymerase (PAP) protein is present and abundant in spermatogenous cells, but PAP is barely detectable in sterile jacket cells (Figure 24). These results suggest that translational capacity in the spermatogenous cells is regulated at several levels, and some of this control apparently involves mRNA processing.

In ongoing work, we have used these mRNAs in assays to study transcript movement prior to cell partitioning within the fixed volume of the gametophyte. These experiments provide insights on cell fate determination in plants, a process that is tightly coupled with cell position and cell size in tissues and organs where cell movements do not occur.

Top of Page

Recent Publications from the Wolniak laboratory

Hart, P. and S.M. Wolniak. 1998. Spermiogenesis in Marsilea vestita: a temporal correlation between centrin expression and blepharoplast differentiation. Cell Motil. Cytoskelet. 41: 39-48. [Cover Illustration 42(2)]

Hardin, S. and S.M. Wolniak. 1998. Molecular cloning and characterization of maize ZmMEK1, a protein kinase with a catalytic domain homologous to mitogen- and stress-activated protein kinase kinases. Planta 206: 577-584.

Hart, P. and S.M. Wolniak. 1999. Molecular cloning of a centrin homolog from Marsilea vestita and evidence for its translational control during spermiogenesis. Biochem. Cell Biol. 77: 101-108.

Wolniak, S.M., V.P. Klink, P.E. Hart and C.-W. Tsai. 2000. Control of the development and motility in the spermatozoids of lower plants. Grav. Sp. Biol. Bull. 13: 85-93.

Klink, V.P. and S.M. Wolniak. 2001. The efficacy of RNAi in the study of the plant cytoskeleton. J. Pl. Growth Reg. 19: 371-384.

Klink, V.P. and S.M. Wolniak. 2001. Centrin is necessary for the formation of the motile apparatus in spermatids of Marsilea. Mol. Biol. Cell 12: 761-776.

Hardin, S. and S.M. Wolniak. 2001. Expression of the mitogen-activated proteinkinasekinase ZmMEK1 in the primary root of maize. Planta 213: 916-926.

Tsai, C.W. and Wolniak. S.M. 2001. Cell cycle arrest allows centrin translation but not basal body formation during spermiogenesis in Marsilea. J. Cell Sci. 114: 4265-4272 [Cover Illustration 114(23)].

Klink, V.P. and .M. Wolniak. 2003. Changes in the abundance and distribution of conserved centrosomal, cytoskeletal and ciliary proteins during spermiogenesis in Marsilea vestita. Cell Motil. Cytoskelet. 56: 57-73.

Tsai, C.W., C.M. Van der Weele and S.M. Wolniak. 2004. Differential segregation and modification of mRNA during spermiogenesis in Marsilea vestita. Develop. Biol. 269: 319-330.

Top of Page


|||||||||| Return to S. Wolniak's Home Page |||||||||||||||||

|||||||||| Return to S. Wolniak's Teaching Interests Page |||||||||||||||||

|||||||||| Return to Departmental Home Page |||||||||||||||||

|||||||||| Translation in Ukrainian|||||||||||||||||


Marsilea vestita
University of Maryland

College of Chemical & Life Sciences * University of Maryland * College Park, MD 20742