Glycodelin A : An apoptogenic lipocalin:Molecular mechanism of immunosuppression by Glycodelin A

Abstract

Glycodelin (previously known as placental protein 14, progestagen-associated endometrial protein, pregnancy-associated secreted endometrial ?2-globulin, etc.) is a 162 amino acid, glycosylated protein secreted mainly by the primate reproductive tissues. Glycodelin has been classified as a member of the lipocalin superfamily (small, hydrophobic molecule transporters) as it has the lipocalin signature motifs. The primary structure of the protein is 40-55% identical to ?-lactoglobulin, the second type member of the lipocalin family, the first being plasma retinol binding protein. Interestingly, no ligand for this molecule has been identified to date. Glycodelin can be subclassified as an immunocalin (lipocalins with immunological function) based on the immunosuppressive properties of one of its isoforms, which is present in amniotic fluid (GdA). Apart from immunosuppression, other functions have been ascribed to this protein, such as contraception, angiogenesis, and morphogenesis. It has been reported that the specific glycosylation pattern on GdA is responsible for its contraceptive activity. Similarly, a lysine-rich stretch of the protein has been identified as the angiogenic region.

GdA was shown to be immunosuppressive by its ability to inhibit two-way mixed lymphocyte reaction, phytohemagglutinin and anti-CD3 monoclonal antibody-induced T cell proliferation, as well as NK cell proliferation in vitro. GdA is synthesized by the secretory and decidualized endometrium when stimulated by progesterone. The presence of a glucocorticoid response element in the promoter region of the protein supports this observation. During pregnancy, the GdA level in the amniotic fluid peaks around the 12-14th weeks of gestation. The endometrial level of GdA during implantation is very high, suggesting a physiological role for the protein, i.e., protecting the fetus from maternal immune attack. Indeed, low levels of GdA are frequently associated with pregnancy loss, and the protein is used as a marker for general endometrial health.

The main objective of this investigation was to delineate the molecular mechanism underlying the immunosuppressive activity of GdA. To study this function in molecular detail, glycodelin was required in pure form and in sufficient quantity. Two approaches were used to obtain the protein: (a) heterologous expression, where the full-length glycodelin cDNA was cloned from endometrial tissue, and the protein was expressed in prokaryotic as well as eukaryotic expression systems, and (b) purification from amniotic fluid by immunoaffinity chromatography using a monoclonal antibody against glycodelin expressed in Pichia pastoris.

Chapter 1 describes the cloning strategy of glycodelin in E. coli as well as in the yeast Pichia pastoris and in the mammalian cell line HEK-293. Expression and purification procedures employed for the recombinant proteins are also discussed in this chapter. Chapter 2 describes the strategy adopted for the purification of the native GdA. Detailed procedures for obtaining and characterizing monoclonal antibodies against glycodelin, standardization of a glycodelin-specific radioimmunoassay, and ultimately purification of the protein from amniotic fluid have been discussed in this chapter.

Glycodelin A purified from amniotic fluid was shown to inhibit proliferation of PBMCs induced either by anti-CD3 monoclonal antibody or phytohemagglutinin, as reported earlier. It was observed that the protein exerts its immunoinhibitory function by inducing apoptosis in activated T cells. This effect was found to be independent of the antigen-presenting cells present in the PBMCs as confirmed by using human T cell lines Jurkat (JR4, A3) and MOLT-4. The mechanism leading to GdA-induced apoptosis was studied in Jurkat cells. It was observed that CD95/Fas and initiator caspase 8 were not involved in the GdA-induced death signaling, though the overall process was caspase-dependent. It was also observed that GdA induces a decrease in the mitochondrial membrane potential. This would lead to the spillage of many pro-apoptotic factors in the cytosol, initiating the apoptotic pathway. The methods and the results obtained regarding the study of the glycodelin-induced apoptosis in T cells and macrophages are described in Chapter 3.

The function of the other isoform of glycodelin found in seminal plasma (GdS) has not been documented, and studies in our laboratory have shown GdS to be apoptotically inactive. The only difference between GdA and GdS lies at the level of glycosylation. GdS has been reported to be fucose-rich and lacking sialic acid modification, whereas GdA has low fucose residues but extensive sialic acid modifications. The role of glycosylation in the apoptotic activity of glycodelin was therefore studied. The glycosyl groups present on GdA are not able to induce apoptosis in the absence of the protein backbone, suggesting that neither the protein alone nor the glycans alone, but only the glycoprotein, can induce apoptosis in Jurkat cells. Further, when GdA was treated with different glycosidases, loss of apoptotic activity was observed only with neuraminidase treatment. Combined with the fact that there were discernible differences in the tertiary structure of GdA and GdS, we propose that there is a glycosylation-induced conformational difference between the two isoforms which may lead to functional differences. The results obtained are discussed in Chapter 4.

To summarize, this study describes an immunoaffinity purification method and a radioimmunoassay for glycodelin, establishes GdA as an apoptogenic protein, and highlights the possibility of sialic acid-induced conformational differences in the glycodelin isoforms which may lead to the difference in their apoptotic function.