Characterization of a putative 82 kDa nuclear ligand for the N-acetylglucosamine-binding protein CBP70

Abstract

Since nuclear lectins were first characterized several years ago, six lectins have been isolated. Furthermore, the existence of nuclear glycoproteins containing N-linked complex-oligosaccharide chains or O-linked GlcNAc residues was evidenced. These latter are abundant in the nucleus and are well-studied so far. The presence of both glycoprotein and lectin in the cell nucleus led us to postulate that these two proteins could interact and play a role in some nuclear activities such as the modulation of transcription and/or nuclear cytoplasmic exchanges or by the disruption of protein-protein interactions. In such context, the recent data concerning the GlcNAc-binding activity of CBP70 argued this postulate. However, to study the possible role of a glycoprotein-lectin complex, it was critical to isolate the two partners. Because CBP70 was also a cytoplasmic protein, the lectin was isolated in both cytoplasmic and nuclear compartments in order to investigate the putative ligand in the two cellular compartments. The results obtained with cross-linking experiments on isolated and membranedepleted nuclei incubated with the CBP70 bearing an iodinatable, cleavable, photoreactive cross-linking agent (sulfosuccinimidyl 2-(p-azidosalicylamido) ethyl-1,3'-dithiopropionate) and immunoprecipitation experiments with polyclonal antibodies raised against CBP70, revealed that both nuclear and cytoplasmic CBP70 have the same 82 kDa nuclear ligand which is absent in the cytoplasmic fraction. In addition, this ligand is glycosylated, containing GlcNAc residues, and, therefore, the complex between CBP70 and the 82 kDa polypeptide could be due to a glycoprotein-lectin interaction. These results raised the possibility that nuclear glycoprotein-lectin interaction could be involved in nuclear activities.

CBP70, a glycosylated nuclear lectin

“…Indeed, a cytoplasmic form of CBP70 was recently described in several cell types, including HL60 cells [Hadj-Sahraoui et al, 1996;Felin et al, 1997]. Furthermore, an 82 kDa nuclear ligand for CBP70, which contains GlcNAc residues, was characterized, that showed for the first time, a nuclear glycoprotein-lectin interaction [Felin et al, 1997]. This result strengthens the postulate emitted some years ago concerning a possible involvement of such interactions in the nuclear physiology [Hubert et al, 1989].…”

“…Initially, CBP70 was isolated from HL60 cell nuclei [Sève et al, 1993]; recently, a cytoplasmic form of this lectin was described, on the basis that the two forms have the same molecular mass, the same carbohydrate binding sites that recognize both Glc and GlcNAc, and are recognized by the same antibody raised against nuclear CBP70 [HadjSahraoui et al, 1996;Felin et al, 1997].…”

“…In this context, very clean nuclear and cytoplasmic fractions from HL60 cells were obtained according to a previously reported method [Sève et al, 1993;Felin et al, 1997]. Then, the combination of two successive rounds of affinity chromatography (GlcNAc and Glc) allowed us to obtain both purified forms of CBP70.…”

“…Recently, we have found that the nuclear and the cytoplasmic CBP70 recognized different ligands [Felin et al, 1997]. Therefore, appropriate glycosylation, could confer to both forms of CBP70 a correct tridimensional structure that enable each of them to recognize, or be recognized, by their specific ligand (e.g., proteins, glycoproteins, or lectins) among all their different potentials ligands available.…”

“…In addition, CBP70 is not a nuclear lectin stricto sensu. Indeed, a cytoplasmic form of CBP70 was recently described in several cell types, including HL60 cells [Hadj-Sahraoui et al, 1996;Felin et al, 1997]. Furthermore, an 82 kDa nuclear ligand for CBP70, which contains GlcNAc residues, was characterized, that showed for the first time, a nuclear glycoprotein-lectin interaction [Felin et al, 1997].…”