Genetic mapping and functional studies of a natural inhibitor of the insulin receptor tyrosine kinase: the mouse ortholog of human alpha2-HS glycoprotein.

Fetuin/alpha2-HS glycoprotein (alpha2-HSG) homologs have been identified in several species including rat, sheep, pig, rabbit, guinea pig, cattle, mouse and human. Multiple physiological roles for these homologs have been suggested, including ability to bind to hydroxyapatite crystals and to specifically inhibit the tyrosine kinase (TK) activity of the insulin receptor (IR). In this study we report the identification, cloning, and characterization of the mouse Ahsg gene and its function as an IR-TK inhibitor. Genomic clones derived from a mouse Svj 129 genomic library were sequenced in order to characterize the intron-exon organization of the mouse Ahsg gene, including an 875 bp subclone containing 154 bp upstream from the transcription start site, the first exon, and part of the first intron. A second genomic subclone harboring a 3.45 kb Bgl II fragment contained exons 2, 3 and 4 in addition to two adjacent elements within the first intron-a repetitive element of the B1 family (92 bp) and a 271 bp tract of (T,C)n*(A,G)n. We have mapped mouse Ahsg at 16 cM adjacent to the Diacylglycerol kinase 3 (Dagk3) gene on chromosome 16 by genotyping interspecific backcross panels between C57BL/6J and Mus spretus. The position is syntenic with human chromosome 3q27, where the human AHSG gene resides. Using recombinant mouse alpha2-HSG expressed from a recombinant baculovirus, we demonstrate that mouse alpha2-HSG inhibits insulin-stimulated IR autophosphorylation and IR-TKA in vitro. In addition, mouse alpha2-HSG (25 microg/ml) completely abolishes insulin-induced DNA synthesis in H-35 rat hepatoma cells. Based on the sequence data and functional analysis, we conclude that the mouse Ahsg gene is the true ortholog of the human AHSG gene.

Alpha2-HSG, a specific inhibitor of insulin receptor autophosphorylation, interacts with the insulin receptor.

Human fetuin, [alpha2-Heremans Schmid Glycoprotein (alpha2-HSG)], is a natural inhibitor of insulin receptor tyrosine kinase activity (IR-TKA). Previously, we have demonstrated that alpha2-HSG inhibits the mitogenic pathway without affecting the metabolic arm of insulin signal transduction. In this study, we demonstrate the time-course and specificity of inhibition, its interaction with IR and probable physiological role. In intact rat1 fibroblasts overexpressing the human insulin receptor (HIRc B), incubation of recombinant human alpha2-HSGbac (1.8 microM) inhibited insulin-induced IR autophosphorylation by over 80%. This inhibitory effect of alpha2-HSGbac on insulin-induced IR autophosphorylation was blunted by half in 60 min. Interestingly, alpha2-HSGbac at similar concentrations (0.9 or 1.8 microM), had no effect on EGF- or IGF-I-induced cognate receptor autophosphorylation. Anti-alpha2-HSG immunoprecipitates of alpha2-HSGbac-treated HIRc B cell lysates demonstrated the presence of IR. Our data suggest that alpha2-HSGbac preferentially interacts with the activated IR. To further characterize the site(s) of interaction, the effect of alpha2-HSGbac on trypsin-treated IR autophosphorylation was studied. Trypsin-treatment of intact HIRc B cells results in proteolysis of the IR alpha-chain and constitutive activation of IR-TKA. We demonstrate that alpha2-HSGbac (0.1 microM) completely inhibited trypsin-activated IR autophosphorylation and TKA in vitro indicating that this effect was not mediated by its interaction with the proximal 576 amino acid residues of the IR alpha-subunit. The physiological relevance of these observations was explored by characterizing the effects of alpha2-HSG injection in rats. Alpha2-HSGbac (2 microM), acutely injected through the portal vein of normal rats, inhibited insulin-stimulated IR autophosphorylation and IRS-1 phosphorylation in liver and hindlimb muscle. Taken together our results suggest that alpha2-HSG, by interacting with IR, specifically inhibits insulin-stimulated IR autophosphorylation and may play a physiological role in the regulation of insulin signaling.

Genetic analysis reveals ongoing HIV type 1 evolution in infected human placental trophoblast.

To provide a better understanding of the role of placenta in vertical human immunodeficiency virus (HIV) transmission, we have studied the infection of placental trophoblast in a group of 15 mother-neonate pairs. By nested PCR amplification of the C2V3 env gene region, HIV-1 has been found to infect the placenta in five cases (33%). Phylogenetic analysis of the cloned sequences showed that all recovered maternal variants were of the B subtype. Further investigation into the ancestral relationships at the nucleotide level revealed that the trophoblast sequences evolved into a quasispecies population clearly distant from that observed in the mother. As expected, the populations transmitted to the trophoblast were also found to be more homogeneous than those in the mothers when characterized on the basis of pairwise nucleotide sequence distances. With regard to the predicted biological properties, the primary amino acid structure of the V3 loop domain was consistent, with a macrophage-tropic, non-syncytium-inducing phenotype in all patients. We also attempted to determine if any of a number of selected maternal or viral factors was associated with trophoblast infection. However, none of the followed parameters, including maternal age, disease stage, antiretroviral therapy, CCR5delta32 deletion status of the infant, and viral genotype, could be associated with viral transmission. Moreover, in one pair with proven trophoblast infection, HIV was also detected in the cord blood. Taken together, our data suggest that the productive trophoblast infection by HIV-1 in vivo is a relatively frequent event that may bear direct implications for a further transplacental propagation of the virus.

Integration of proviral DNA into the PDGF beta-receptor gene in HTLV-I-infected T-cells results in a novel tyrosine kinase product with transforming activity.

We have previously shown that noninfected human T-cell lines express the canonical 5.7 kb mRNA coding for the type beta platelet-derived growth factor-receptor (PDGF beta-receptor), whereas HTLV-I-infected T-cell lines express a novel PDGF beta-receptor mRNA of 3.8 kb. In this report, we have extended those studies to molecularly characterize the 3.8 kb PDGF beta-receptor mRNA and show that it has resulted from integration of an apparently undeleted HTLV-I provirus into the PDGF beta-receptor gene in an orientation enabling expression of a truncated PDGF beta-receptor mRNA using the 3' HTLV-I long terminal repeat as a promoter. Further, NIH3T3 cells transfected with a plasmid containing the truncated PDGF beta-receptor ORF plasmid generate colonies in soft agar with more cells per colony than untransfected cells, or cells transfected with the Tax 1 or PDGF-B (c-sis) plasmids. These results indicate that the truncated PDGF beta-receptor protein acquires transforming capability and that HTLV-I-induced truncation of PDGF beta-receptor may correlate with HTLV-I-associated neoplasia of human T-cells.

Recombinant human alpha 2-HS glycoprotein inhibits insulin-stimulated mitogenic pathway without affecting metabolic signalling in Chinese hamster ovary cells overexpressing the human insulin receptor.

Insulin acts on its target tissues by specific interaction with the cell surface insulin receptor (IR). The IR possesses an intrinsic tyrosine kinase (TK) activity which is stimulated by insulin binding. This TK activity is required for many aspects of insulin signalling. We had earlier reported that human plasma alpha 2-HS glycoprotein (alpha 2-HSG) inhibits insulin-stimulated mitogenesis at the level of IR-TK (Mol Endo 7: 1445-1455, 1993). In the present study, using recombinant alpha 2-HSG, which possesses 50-100 times the specific activity of plasma alpha 2-HSG, we have further investigated the molecular basis of this effect. We examined the insulin-stimulated Ras signalling pathway in Chinese Hamster Ovary cells overexpressing the human IR. alpha 2-HSG inhibits insulin-induced tyrosine phosphorylation of IRS-1 and the subsequent association of GRB2, as well as Sos, with IRS-1. This inhibition results in reduced guanine nucleotide exchange in p21ras. alpha 2-HSG also inhibits the stimulation of Raf phosphorylation, in response to insulin, leading to inhibition of MEK activity. In a parallel pathway, alpha 2-HSG also inhibits insulin-induced tyrosine phosphorylation of Shc. However, alpha 2-HSG does not affect any of the metabolic actions of insulin rested in these cells. These results suggest that, while insulin's mitogenic effects can be abolished by inhibition of insulin-induced IR-TK, propagation of signals for metabolic activities might utilize alternate of rescue mechanisms.

Baculoviral expression of a natural inhibitor of the human insulin receptor tyrosine kinase.

We had earlier reported that the human serum alpha 2-HS glycoprotein (1) is a physiological and specific inhibitor of the human insulin receptor tyrosine kinase (IR-TK). We have now expressed this human protein in the baculoviral expression system using the Sf-9 and High Five insect cells. The protein was optimally expressed at 72 h post infection. alpha 2-HSGbac completely inhibited the insulin-stimulated autophosphorylation and TK activity of partially purified IR preparations. It also abolished insulin-induced DNA synthesis in the H-35 rat hepatoma cell line. The effective concentration of the baculoviral derived alpha 2-HSG necessary for inhibiting IR-TK activity was significantly lower than that of the protein purified from human plasma.

Basal and Tat-transactivated expression from the human immunodeficiency virus type 1 long terminal repeat in human placental trophoblast rules out promoter-enhancer activation as the partial block to viral replication.

We have analysed the capacity of the trophoblast-derived malignant cell lines BeWo, JAR and JEG-3, and primary cultures of highly purified trophoblast cells to support the basal and Tat-mediated trans-activation-enhanced transcriptional activity of two distinct human immunodeficiency virus type 1 (HIV-1) isolates. Kinetic studies based on expression of long terminal repeat (LTR)-chloramphenicol acetyltransferase (CAT) constructs revealed that LTRs of both the prototype strain 3B and the highly cytopathic Zairean variant NDK were activated significantly in all target cells. Overall, the strongest activation was observed in primary trophoblasts. A novel modification of quantitative PCR was used to normalize LTR expression for transfection efficiency, enabling the calculation of specific expression rates in terms of muU CAT enzyme per fmol of transfected DNA. Using the latter criterion we determined that LTRs of both viruses were activated in decreasing order from trophoblasts to JAR, JEG-3 and BeWo cells; furthermore, the expression of HIV-1 3B LTR always significantly surpassed that of HIV-1 NDK. The effects of trans-activation on either of the LTRs, when assayed in cotransfection assays with various amounts of HIV-1 NDK-Tat expression vector, increased in a dose-dependent fashion and were comparable in a particular neoplastic cell line. Furthermore, the cell-specific LTR activity patterns did not correspond to the abundance of transcription factors binding specifically to the viral NF kappa B and SP1 motifs. Unlike SP1-binding proteins which were relatively abundant, substantially smaller amounts of proteins with NF kappa B specificity were found in all cells. Despite this apparent deficit in NF kappa B activity, trophoblasts supported a high basal activity of both LTRs. These data indicate that an insufficiency of basal or Tat-trans-activated LTR activity cannot account for the low level of HIV-1 replication in this important cell type.

Serum alpha 2-HS-glycoprotein is an inhibitor of the human insulin receptor at the tyrosine kinase level.

The insulin-dependent tyrosine kinase activity (TKA) of the insulin receptor (IR) plays an essential role in insulin signaling. Thus, dysregulation of IR-TKA might be an important element in the states of insulin resistance. A phosphorylated rat hepatic glycoprotein (pp63) acting as an inhibitor of IR-TK has been described. In search of the human homolog of pp63, we isolated a cDNA clone from a human liver lambda gt11 cDNA library. DNA sequence analysis reveals identity with the mRNA product of a human gene AHSG encoding a serum protein, alpha 2-Heremans Scmid-glycoprotein (alpha 2HSG), with heretofore unknown physiological function. Northern blot analysis demonstrates a 1.8-kilobase mRNA in human liver and HepG2 hepatoma cells. alpha 2HSG, purified from human serum, specifically inhibits insulin-stimulated IR autophosphorylation in vitro and in vivo as well as exogenous substrate tyrosine phosphorylation. alpha 2HSG also inhibits both insulin-induced tyrosine phosphorylation of IRS-1 and the association of IRS-1 with the p85 subunit of phosphatidylinositol-3 kinase in H-35 hepatoma cells. alpha 2HSG inhibits insulin-dependent mitogenesis, but does not affect insulin-stimulated induction of the metabolic enzyme tyrosine aminotransferase. alpha 2HSG does not compete with insulin for binding to IR. Finally, the action of alpha 2HSG is specific toward the IR-TK; its effect does not extend to insulin-like growth factor-I-stimulated TKA. Our results allow us to assign a biochemical function for human alpha 2HSG, namely regulation of insulin action at the IR-TK level.

Coexpression of the genes for platelet-derived growth factor B-chain receptor and macrophage colony-stimulating factor 1 receptor during monocytic differentiation.

Receptors for platelet-derived growth factor (PDGF) have not been identified previously to our knowledge in human myeloid cells that also produce PDGF. Here we report that phorbol ester-treated myeloid cells differentiated along the monocytic lineage express both a full-length 5.5-kilobase (kb) mRNA and a predominant, truncated 4.6-kb mRNA coding for the PDGF B-chain receptor (PDGF-BR). PDGF-BR was identified in phorbol ester-differentiated myeloid cells by indirect immunofluorescence with an antibody specific to PDGF-BR. This anti-PDGF-BR was also used in immunoprecipitation studies to demonstrate that lysates of phorbol ester-differentiated myeloid cells contain PDGF-BR molecules of 37 kDa to 130 kDa. The results also show that the tandemly linked genes for PDGF-BR and the macrophage colony-stimulating factor 1 receptor are coexpressed in the phorbol ester-differentiated myeloid cells. Expression of these two receptor genes has not been shown previously in any cell type to our knowledge.

Platelet-derived growth factor and its receptor in blood cell differentiation and neoplasia.

Platelet-derived growth factor (PDGF) is a family of dimeric protein molecules synthesized by differentiated, non-dividing and proliferating blood cells. Experimental findings indicate that PDGF is involved in development and/or maintenance of physiological functions of certain normal blood cells. Also, PDGF synthesis correlates with certain blood cell proliferative diseases caused either spontaneously or associated with viral infection. There is increasing evidence that the diverse effects of PDGF in both normal and abnormal physiological functions of blood cells may be regulated at the level of its receptor. New experimental findings are discussed relating to PDGF receptors in normal leukemic, and virally-infected human cells of myeloid and lymphocytic lineages. At specific developmental stages this regulation may take the form of PDGF and its receptor being expressed or co-expressed; the unmodified or modified form of receptor that specifically interacts with PDGF; the cellular site at which the PDGF-receptor interacts with its ligand; and co-expression of the PDGF-receptor with other receptors associated with specific cell lineage or functions. Elucidation of events involved in synthesis, processing, and interactions of PDGF isoforms and their respective receptors will enable us to develop pharmacological means that may either interfere with, or enhance these desired blood cell functions. This review focuses on PDGF and its receptor in human blood cell differentiation and neoplasia.

Coexpression of the genes for platelet-derived growth factor and its receptor in human T-cell lines infected with HTLV-I.

Several HTLV-I-infected T-cell lines express the two genes encoding platelet-derived growth factor (PDGF). Therefore, we examined the question of a possible self-stimulatory mechanism of proliferation involving PDGF in these cells. Using a nucleic acid probe and an antibody specific for the PDGF-B receptor (PDGFR-B), we examined four established human T-cell lines infected with HTLV-I, as well as several noninfected T-cell lines for expression of the PDGF-B receptor. Previous reports indicate that lymphocytes do not display receptors for PDGF; our results show that two noninfected T-cell lines (HUT-78, CCRF-HSB-2) expressed the canonical 5.5-kb PDGFR mRNA, and two HTLV-I-infected T-cell lines (C10/MJ, MT-2) expressed a novel PDGFR mRNA of 4.8 kb. Concomitantly, the cell lines expressing PDGFR mRNA also synthesize PDGFR proteins immunoprecipitated by the antibody to PDGFR-B. Differences were observed in the molecular weight of PDGFR molecules immunoprecipitated from uninfected and HTLV-infected T-cells. Immunofluorescence studies demonstrated that the PDGFR-B proteins are localized primarily on the cell external membrane. The results suggest that the HTLV-I-infected T-cells acquire an autostimulatory mechanism of cell proliferation that involves PDGF.

Transforming growth factor type beta can act as a potent competence factor for AKR-2B cells.

Transforming growth factor type beta (TGF beta) is a pleiotropic regulator of cell growth with specific high-affinity cell-surface receptors on a large number of cells; its mechanism of action, however, is poorly defined. In this report, we utilized the mouse fibroblast line AKR-2B to explore the question of the temporal requirements during the cell cycle in regard to both the growth inhibitory and the growth stimulatory action of TGF beta. The results indicate that AKR-2B cells are most sensitive to the inhibitory action of TGF beta during early to mid-G1. In addition, TGF beta need be present only briefly (as little as 1 min) in order to exert its inhibitory effect on EGF-induced DNA synthesis. Likewise, the stimulatory effect of TGF beta in the absence of EGF requires only an equally brief exposure to TGF beta. Use of homogeneous 125I-labeled TGF beta in a cell-binding assay demonstrates that TGF beta bound to cell-surface receptors can readily exchange into the culture medium T1/2 = 120 min), helping to rule out the possibility that persistent receptor-bound TGF beta is the source of a continuous stimulus. The data indicate that TGF beta exposure induces a stable state in the cell (T1/2 = 20 h) similar to but distinct from the state of "competence" induced by platelet-derived growth factor (PDGF).

Transforming growth factor alpha and beta expression in human colon cancer lines: implications for an autocrine model.

Three human colon cancer lines (SW480, SW620, WIDR) secrete different levels of transforming growth factor beta (TGF beta)-like and transforming growth factor alpha (TGF alpha)/epidermal growth factor (EGF)-like molecules into serum-free conditioned media as measured by competing activity in TGF beta and EGF radioreceptor assays. SW480 cells, the highest producers of TGF beta-like activity, lack detectable TGF beta receptors while SW620 cells, the highest producers of TGF alpha/EGF-like activity, lack EGF receptors. This study investigated the production of these growth factors at the mRNA level and examined the mechanism of loss of detectable receptors. Using complementary DNA probes for TGF beta and TGF alpha, it was demonstrated that mRNA levels correlated with the amounts of TGF beta and TGF alpha produced; TGF beta gene expression was highest in SW480 cells and TGF alpha gene expression was highest in SW620 cells. Acid washing of the SW480 cells prior to performing the TGF beta binding assay resulted in the unmasking of substantial levels of TGF beta receptors. Neither acid washing nor preincubation with suramin uncovered EGF receptors in SW620 cells. Also, and in contrast to the other two lines, EGF receptor expression could not be detected in SW620 cells by Northern gel analysis of receptor messenger RNA or by immunological analysis of receptor protein. Thus two distinct mechanisms (occupation of TGF beta receptor in SW480 cells, or absence of EGF receptor in SW620 cells) explain the lack of detectable TGF beta and EGF receptors in the binding assays. The autocrine hypothesis remains viable for TGF beta in SW480 cells but not for TGF alpha in SW620 cells; this would not discount a paracrine role in this latter case.