BIOLOGICAL OVERVIEW

How is spatial asymmetry generated in the egg and the surrounding follicle cells. The answer question is found in a complex series of signaling events that first sends information from the egg to surrounding follicle cells, and only subsequently generates spatially differentiated populations of follicle cells. Pipe, produced by follicle cells, is one of the last proteins to be involved in these events, while the proteins of the Gurken-Egfr pathway are among the first. The gene Egfr is required in the somatic follicle cells, and when mutant, causes the generation of ventralized phenotypes in eggs and embryos. Gurken, a ligand for Egfr, is produced by the developing egg and secreted locally, altering the developmental fate of dorsal follicle cells, thus initiating the process that gives rise to the D/V asymmetry of follicle cells, which give rise to a ventral cell fate. Epistasis experiments have shown that the Grk-Egfr pathway negatively regulates the production of the ventralizing signal in the follicle cells, although the exact nature of this interaction is unknown. The follicle cells form a single-cell epithelial layer that surrounds the nurse cells and the developing oocyte. Their role during oogenesis is multifunctional as they are responsible for the secretion of both structural components of the egg, like the chorion, vitelline membrane, and yolk proteins (for review, see Spradling 1993), and regulatory molecules like Nudel (Hong, 1995) that participate in the establishment of embryonic polarity. Thus, the follicle cells form a population of cells that are highly secretory in nature and are, in addition, the recipients and transmitters of extracellular signals.

The establishment of DV polarity in Drosophila embryos is dependent on ventrally restricted activation of the uniformly distributed Toll receptor, a complex process that involves multiple proteins including Pipe. Toll's putative ligand-binding domain is predicted to extend into the perivitelline space between the embryonic membrane and the vitelline layer of the eggshell. The ligand for Toll is thought to be encoded by the spätzle gene, whose product is secreted from the embryo as an inactive precursor that is processed into the active ligand by a proteolytic cascade including the products of the genes easter (ea), snake (snk), and gastrulation defective (gd). This proteolytic cascade is believed to act ventrally in the perivitelline space of the egg. However, the mechanism that determines the site of Spätzle processing within the egg has remained enigmatic (Sen, 1998 and references).

Despite the identification of many of the genes involved in determination of D/V polarity, an understanding of the direct agent that communicates the established DV polarity in the follicle cell layer to bring about the localized activation of a protease cascade has been elusive. One would want to search among the genes expressed in follicle cells to discover the agent that brings about localized activation of the protease cascade. Three of the dorsal group genes, windbeutel (wind), nudel, and pipe, have been shown by genetic mosaic experiments to be required in the somatic follicle cells rather than in the germ line (Stein, 1991). These genes are therefore candidates for encoding proteins that may directly produce the ventral signal in the follicle cells. The gene nudel has been cloned and shown to encode a modular protein with an extracellular matrix domain and a serine protease domain (Hong, 1995). It has been suggested that Nudel is secreted by the follicle cells and may possibly be incorporated in the vitelline membrane, thus specifying the site of generation of the active Spätzle ligand, after fertilization of the oocyte. This idea of a localized protease would fit well with the model of proteolytic activation of the ligand for the Toll receptor by sequential processing of a series of serine protease pro-enzymes.

Nudel is not, however, the spatially restricted signal responsible for the stimulation or generation of the spatially delimited protease cascade. When clones of nudel mutants are examined for the effect on D/V polarity, no spatially localized requirement for nudel activity is found. These results imply that wild-type ndl activity in subpopulations of follicle cells, regardless of their position, must be sufficient to mediate its dorsoventral patterning function. In contrast, windbeutel and pipe are required only within a restricted ventral region of the follicular epithelium (Nilson, 1998).

Likewise, Windbeutel is not the spatially restricted signal. wind encodes a protein that is a putative resident of the endoplasmic reticulum (ER). The ER plays a role in the translocation, folding, and degradation of membrane bound and secreted proteins; the product of the gene wind could be involved in the modification and/or folding of a factor secreted by the follicle cells over the ventral side of the oocyte. Nevertheless, careful examination of the wind expression pattern reveals that while only the follicle cells that are located over the oocyte express the WIND transcript, there does not seem to be any restriction of wind expression relative to the dorsoventral axis within the columnar follicle cell epithelium. Rather, all the follicle cells over the oocyte uniformly express wind. Interestingly, the transcript appears to be concentrated in the apical cytoplasm of the follicle cells, which in the follicular epithelium is the side facing the oocyte. The functional significance of this localization is not known but Wind could be involved in the secretion of the sought for spatially restricted agent (Konsolaki, 1998).

That leaves Pipe. Its transcript is in fact spatially restricted to ventral follicle cells, leading to the belief that the long sought spatially restricted signal has been found. Pipe is neither a protease nor a protein involved in secretion. The molecular cloning of pipe reveals it to be a heparan sulfate 2-O-sulfotransferase (HSST). Based on its structural similarity to Chinese hamster HSST, it appears likely that a spatially restricted Pipe protein acts to mediate the attachment of sulfate moieties to sugars that are borne by proteins required for DV pattern formation. The targets of HSSTs are proteoglycans (proteins modified by sugars). Proteoglycans comprise a class of proteins that contain covalently attached carbohydrate side chains or (glycosamineglycans) GAGs. GAG side chains are characterized by the presence of particular alternating pairs of monosaccharides and by the types of linkages that exist between the monosaccharides. Marked heterogeneity is observed in purified GAGs and in the specificity of their biological function as a result of sulfate addition at specific positions in the monosaccharide constituents. Based on Pipe's structural similarity to mammalian HSST, it appears likely that Pipe acts to mediate the sulfation at the 2-O position of hexuronic acid moeities of heparan sulfate/heparin or dermatan sulfate GAG(s) attached to a proteoglycan required for DV pattern formation (Sen, 1998).

This finding suggests a previously unsuspected involvement of polysaccharides and their modification in the regulation of DV polarity determination in the Drosophila embryo. One of the sulfotransferase (ST) isoforms encoded by the pipe locus, pipe-ST2, exhibits a striking, ventrally restricted pattern of expression in the follicular epithelium of stage 9/10 follicles. Expression of transgene-encoded pipe-ST2 in the ovaries of females carrying mutant alleles of pipe leads to the restoration of lateral and ventral pattern elements in progeny embryos. Further, ectopic expression of pipe-ST2 in the follicle cell layer is capable of determining the polarity of the resulting embryos. These findings indicate that pipe plays a pivotal role in the process that defines the DV axis of the embryo and that its spatially regulated activity may provide the link between the establishment of DV polarity in the follicle and the transmission of DV patterning information to the developing egg and future embryo (Sen, 1998).

To assess the influence of egg chamber polarity on the expression pattern of the pipe gene, its expression was examined in the ovaries of females mutant for genes that affect the DV polarity of the follicle and/or the embryo. In the ventralized follicles from females mutant for gurken, virtually uniform distribution of the PIPE-ST2 transcript is observed throughout the follicle cell layer. In egg chambers from fs(1)K101 mutant females, only a patch of follicle cells at the posterior end of the egg chamber express the gene. This is consistent with observations that in fs(1)K10 mutants, mislocalization of the GRK mRNA at the anterior end of the oocyte leads to a perpendicular shift in the DV axis, such that the follicle cells at the posterior of the egg chamber assume a ventral fate. In this series of experiments nudel is expressed uniformly in the follicle cells of the wild-type ovary, and this distribution does not change in any of the genotypes examined. This suggests that the alterations in pipe-ST2 expression seen in the grk and fs(1)K10 mutant backgrounds specifically reflect the effects of changes in follicle cell polarity (Sen, 1998).

To investigate whether the ventrally restricted expression of the pipe gene results from regulation of transcriptional initiation, an examination was carried out of the ability of 5' sequences from the pipe gene to drive E. coli lacZ expression in a ventral pattern. Since fragment A includes the 5' end of the putative pipe cDNA, it is considered likely that this 5' end also contains promoter and enhancer/silencer elements required for the correct expression of pipe. Transformants carrying lacZ downstream of fragment A were generated and their ovaries dissected and stained for beta-galactosidase activity. Stage 10 egg chambers exhibit ventral beta-galactosidase staining that is virtually identical in distribution to that seen for the PIPEmRNA itself. When the same transgenic insert is introduced into the gurken mutant maternal background, uniform beta-galactosidase activity is detected. These observations confirm that the ventrally restricted expression of pipe arises through transcriptional regulation that is mediated via Grk/Egfr signaling (Sen, 1998).

The results of the experiments described above indicate that the spatial pattern of pipe expression is dependent on the DV polarity of follicle cells as defined by the Egfr and gurken genes. To determine whether the alterations in pipe expression generated by mutations in these genes are responsible for the corresponding changes in the embryonic DV axis, experiments were carried out to express the pipe-ST2 cDNA in specific patterns in the follicular epithelium in females that otherwise lacked pipe activity. Uniform expression of pipe-ST2 in the stage 9/10 egg chamber, leads to strong ventralization of progeny embryos derived from both pipe/pipe and pipe/+ females. These embryos produced cuticles with rings of ventral denticle material. Observed under oil, differentiated embryos exhibit muscular movement, indicating the formation of mesoderm. Consistent with the ventralization of the cuticle, gastrulation orientations are variable in these embryos. Strikingly, when pipe-ST2 is expressed in a pipe/pipe mutant background in a dorsal region of the follicle, the embryos exhibit gastrulation movements that are inverted with respect to the intrinsic orientation of the eggshell. The most ventral structures differentiated by most embryos are Filzkörper, although some embryos also develop ventral denticle material. When pipe is expressed in the posterior of embryos, the resulting embryos exhibit gastrulation movements that are consistent with the posterior expression of pipe. These embryos develop furrow invagination, normally formed in the ventral portion of the embryo, at the posterior end only. Consistent with these observations, these embryos form rings of denticles in the posterior half only, indicating that restricted expression of pipe along the anterior-posterior axis of the egg chamber leads to spatially restricted phenotypic rescue of the embryos (Sen, 1998).

A persistent enigma in the understanding of Drosophila DV pattern formation has been the mechanism by which the DV pattern of the egg chamber is linked to the formation of the embryonic DV axis. The identification and analysis of pipe may provide a long-sought answer to this question. Future investigations will seek to identify the specific sulfotransferase activities catalyzed by Pipe and the identity of the proteoglycan carrying the GAGs acted upon by Pipe. These investigations should contribute to the elucidation of the role of carbohydrate modification in the pathway that defines embryonic polarity (Sen, 1998).

GENE STRUCTURE

Bases in 5' UTR - 145

Exons - 3

Bases in 3' UTR - 305

PROTEIN STRUCTURE

Amino Acids - 399

Structural Domains

The conceptually translated sequence contains three regions of open reading frame, all of which exhibit strong amino acid sequence similarity to portions of Chinese hamster heparan sulfate 2-O-sulfotransferase (Kobayashi, 1996 and 1997). Two in-frame methionine codons spaced ten codons apart provide potential sites for translational initiation. The second of these is contained within a perfect consensus site for Drosophila translational initiation. The first and second pipe ORFs flank consensus splice acceptor and donor sequences and presumably an unspliced intron. Splicing of the putative intron would result in a 399 amino acid ORF (ST1) that exhibits 49.2% sequence similarity (27.8% identity) to HSST over its entire sequence. Probes were used to identify a 1.6 kb cDNA clone from a stage 10 follicle cell library in which the first 95 amino acids of ST1 are joined in frame to the first methionine residue encoded in ORF3. This spliced isoform is referred to as ST2. When the ST2 open reading frame is aligned with CHO HSST, it exhibits 28.1% identity and 52.8% similarity to that protein over its entire length. Because the two isoforms share the first 95 amino acid residues, both proteins exhibit the typical type II transmembrane structure of a Golgi resident protein, with a 23 amino acid hydrophilic cytoplasmic domain followed by a 15 amino acid hydrophobic transmembrane domain, and a large overall hydrophilic domain that presumably provides catalytic function (Sen, 1998).

EVOLUTIONARY HOMOLOGS

Heparan sulfate 2-sulfotransferase, which catalyzes the transfer of sulfate from adenosine 3'-phosphate 5'-phosphosulfate to position 2 of L-iduronic acid residue in heparan sulfate, was purified 51,700-fold to apparent homogeneity with a 6% yield from cultured Chinese hamster ovary cells. The isolation procedure included a combination of affinity chromatography on heparin-Sepharose CL-6B and 3',5'-ADP-agarose, which was repeated twice for each, and finally gel chromatography on Superose 12. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis of the purified enzyme showed two protein bands with molecular masses of 47 and 44 kDa. Both proteins appear to be glycoproteins, because their molecular masses decrease after N-glycanase digestion. When completely desulfated and N-resulfated heparin and mouse Engelbreth-Holm-Swarm tumor heparan sulfate are used as acceptors, the purified enzyme transfers sulfate to position 2 of L-iduronic acid residue but does not transfer sulfate to the amino group of glucosamine residue or to position 6 of N-sulfoglucosamine residue. Heparan sulfates from pig aorta and bovine liver, however, are poor acceptors. The enzyme shows no activities toward chondroitin, chondroitin sulfate, dermatan sulfate, and keratan sulfate. The optimal pH for the enzyme activity is around 5.5. The enzyme activity is minimally affected by dithiothreitol and is stimulated strongly by protamine. The Km value for adenosine 3'-phosphate 5'-phosphosulfate is 0.20 microM (Kobayashi, 1996).

Heparan-sulfate 2-sulfotransferase (HS2ST) internal amino acid sequences were obtained from the peptides after digestion of the purified protein with a combination of endoproteinases. Mixed oligonucleotides based on the peptide sequences were used as primers to obtain a probe fragment by reverse transcriptase-polymerase chain reaction using CHO cell poly(A)+ RNA as template. The clone obtained from a CHO cDNA library by screening with the probe is 2.2 kilobases in size and contains an open reading frame of 1068 bases encoding a new protein composed of 356 amino acid residues. The protein predicts a type II transmembrane topology similar to other Golgi membrane proteins. Messages of 5.0 and 3.0 kilobases are observed in Northern analysis. Evidence that the cDNA clone corresponds to the purified HS2ST protein is as follows: (1) The predicted amino acid sequence contains all five peptides obtained after endoproteinase digestion of the purified protein; (2) the characteristics of the predicted protein fit those of the purified protein in terms of molecular mass, membrane localization, and N-glycosylation, and (3) when the cDNA containing the entire coding sequence of the enzyme in a eukaryotic expression vector is transfected into COS-7 cells, the HS2ST activity increases 2.6-fold over controls, and the FLAG-HS2ST fusion protein purified by affinity chromatography shows the HS2ST activity alone (Kobayashi, 1997).

date revised: 25 April 2024

Home page: The Interactive Fly © 1995, 1996 Thomas B. Brody, Ph.D.

The Interactive Fly resides on the
Society for Developmental Biology's Web server.