Identification of a protein encoded by the trans activator gene tatIII of human T-cell lymphotropic retrovirus type III.

The human T-cell lymphotropic virus type III (HTLV-III/LAV) is a retrovirus associated with acquired immune deficiency syndrome. The region on the viral genome that is necessary for trans-activation of the HTLV-III/LAV long terminal repeat called tatIII has previously been determined to lie between nucleotides 5365 and 5607. Here we report that a bacterial fusion protein containing amino acid sequences specified by the first coding exon of the tatIII gene is recognized by some patient antisera. We also demonstrate that lymphoid and epithelial cells that express the trans activator function express a 14-kilodalton (kDa) protein recognized by a patient antiserum that reacts with the bacterial tatIII fusion protein. Cells transiently transfected with a deletion mutant of the trans activator protein produce a 12-kDa protein rather than the 14-kDa protein. These observations indicate that the tatIII region contains a functional gene and is capable of expressing a protein that migrates with an apparent molecular size of 14 kDa in some lymphoid and epithelial cells transfected with plasmids containing the tatIII region. We propose that the product of the trans activator gene be designated p14tat-III.     

ATP-driven calcium transport in membrane vesicles of Streptococcus sanguis.

Calcium transport was investigated in membrane vesicles prepared from the oral bacterium Streptococcus sanguis. Procedures were devised for the preparation of membrane vesicles capable of accumulating 45Ca2+. Uptake was ATP dependent and did not require a proton motive force. Calcium transport in these vesicles was compared with 45Ca2+ accumulation in membrane vesicles from Streptococcus faecalis and Escherichia coli. The data support the existence of an ATP-driven calcium pump in S. sanguis similar to that in S. faecalis. This pump, which catalyzes uptake into membrane vesicles, would be responsible for extrusion of calcium from intact cells.     

Reconstitution of ATP-dependent calcium transport from streptococci

Membrane vesicles of three streptococcal strains (Streptococcus faecalis, Streptococcus lactis, and Streptococcus sanguis) were extracted with octyl-beta-D-glucoside in the presence of Escherichia coli lipid and glycerol. For reconstitution, the detergent extract was mixed with bath-sonicated E. coli lipid, in the presence of octyl-beta-D-glucoside, and proteoliposomes were formed by a 25-fold dilution. ATP-dependent calcium accumulation by proteoliposomes was comparable to that found in parent vesicles. Recovery of this calcium transport activity was dependent on the inclusion of an osmolyte protein stabilant (glycerol, etc.) during solubilization. The properties of ATP-driven calcium transport were studied in the reconstituted system. In proteoliposomes, ATP-linked calcium accumulation was not affected by the protonophore, carbonyl cyanide p-trifluoromethoxyphenylhydrazone, or by the ionophores, valinomycin and nigericin, in the presence of potassium, or by N,N'-dicyclohexylcarbodiimide, an inhibitor of the F0F1-ATPase. On the other hand, calcium transport was completely blocked by micromolar levels of orthovanadate; half-maximal inhibitions were observed at 0.4, 4, and 4 microM vanadate, for S. faecalis, S. lactis, and S. sanguis, respectively. This marked sensitivity to orthovanadate suggests operation of an E1E2-type ion-motive pump. These data demonstrate that, in a reconstituted system, calcium transport is not linked to an ATP-dependent proton circulation via the F0F1-ATPase, but rather is driven by a calcium-translocating ATPase. Thus, calcium extrusion from the cytosol of enteric, lactic acid, or oral streptococci is mediated by an ATP-linked process analogous to the ion-motive ATPases of eukaryotic membranes.     

Increasing the cAMP content of IM-9 cells alters the phosphorylation state and protein kinase activity of the insulin receptor

The effect of 8-bromo-cAMP and forskolin on the phosphorylation state and protein kinase activity of the insulin receptor was evaluated in cultured IM-9 lymphoblasts. 8-Bromo-cAMP (1 mM) or forskolin (10 microM) enhanced the phosphorylation of the insulin receptor purified from 32P-labeled cells by affinity chromatography on wheat germ agglutinin-agarose and immunoprecipitation with monoclonal antibody. In the absence of insulin, phosphorylation of the beta subunit of the receptor was increased approximately 2-fold by raising intracellular cAMP. Phosphoamino acid analysis of the beta subunit following treatment of cells with forskolin revealed an increase in phosphoserine and phosphothreonine residues. In contrast, the insulin-stimulated phosphorylation of the receptor occurred on serine, threonine, and tyrosine residues and was diminished by prior exposure of cells to forskolin. Pulse-chase experiments indicated that forskolin did not enhance the turnover of phosphate on the receptor of cells previously exposed to insulin. Furthermore, extracts from forskolin-treated cells did not differ from control extracts in their capacity to dephosphorylate 32P-labeled receptor isolated from cells treated with insulin. The insulin-dependent tyrosine protein kinase activity of the receptor isolated from forskolin-treated cells was approximately 50% as active as the receptor isolated from either control or insulin-treated cells. This was assessed using both histone and a peptide synthesized in accordance with the deduced amino acid sequence of a potential autophosphorylation site of the human receptor (Thr-Arg-Asp-Ile-Tyr-Glu-Thr-Asp-Tyr-Tyr-Arg-Lys) as substrates for the protein kinase reaction. These results suggest that agents that raise intracellular cAMP increase phosphorylation of the insulin receptor on serine and threonine residues, reduce insulin-mediated receptor phosphorylation on tyrosine, serine, and threonine residues, and inhibit the insulin-dependent tyrosine protein kinase activity of the receptor. Thus cAMP may attenuate insulin action by altering the state of phosphorylation of the insulin receptor.     

Antibodies to the N-terminal portion of cartilage-inducing factor A and transforming growth factor beta. Immunohistochemical localization and association with differentiating cells

Two apparently homologous proteins, designated CIF-A and CIF-B, were previously isolated from bovine bone on the basis of their cartilage-inducing activity in culture. CIF-A has been shown to probably be identical to transforming growth factor beta (TGF-beta). To address the question of tissue localization, antibodies to CIF-A were produced using a synthetic polypeptide identical to N-terminal residues 1-30. The antibodies were immunoreactive with bovine CIF-A and human TGF-beta, did not recognize CIF-B, and did not recognize other molecular weight species in crude bovine bone extracts. The antibodies were used to immunohistochemically localize CIF-A/TGF-beta in fetal bovine bone and other tissues. There was abundant staining of osteocytes throughout cancellous and cortical bone as well as chondrocytes within the articular cartilage, although growth plate-associated chondrocytes were not labeled. In addition, immunoreactive cells were detected in bone marrow (megakaryocytes and some mononuclear cells), fetal liver (hematopoietic stems cells), and the thymus (Hassall's corpuscle and some medullary thymocytes). In the kidney, the antibodies labeled a population of epithelial cells lining the calyces. Tissues which did not have detectable amounts of CIF-A/TGF-beta included the thyroid, adrenal, salivary gland, and aorta. Results presented here suggest that the factor may function in vivo as a general development and repair factor and may play a significant role in the differentiation of many cell types including chondrocytes, osteocytes, T-lymphocytes, and red blood cells.     

Cytodiagnosis of in situ and early carcinoma of the upper gastrointestinal tract

Three cases of adenocarcinoma of the stomach, two in situ and one superficially invasive, and one of superficially invasive squamous-cell carcinoma of the esophagus are presented to illustrate the problems encountered in the diagnosis of early lesions of the upper gastrointestinal (GI) tract and the contribution that cytodiagnosis can make. The symptomology and roentgenographic findings in these cases were largely nonspecific. While endoscopic biopsies were repeatedly negative in three of the four cases, endoscopic brushing cytology consistently indicated the presence of a malignancy. Surgery was finally performed on the basis of the cytologic findings, confirming the presence of early malignancy. The cytologic findings, with histologic correlations, are presented in an effort to define some specific criteria for the diagnosis of early malignancy of the upper GI tract.     

Growth kinetics as a function of ploidy in diploid, tetraploid, and octaploid smooth muscle cells derived from the normal rat aorta

The smooth muscle cell population in major arteries of humans and experimental animals is heterogeneous with regard to cellular DNA content. A proportion of cells has polyploid DNA content and this proportion increases with normal aging and with hypertension. We have isolated pure populations of rat aortic smooth muscle cells containing 2C, 4C, and 8C DNA content by cloning of cultures of cells previously subjected to flow cytometric cell sorting. Karyologic analysis of these clonal populations revealed them to be pure diploid, tetraploid, and octaploid populations, respectively, containing 2N (= 42), 4N, and 8N chromosomes. Cell attachment area and nuclear size appeared to increase with the level of ploidy. Studies of the proliferative characteristics of the cells revealed that the growth rate and ultimate cell densities achieved decreased as the ploidy level increased. The intrinsic cellular radiosensitivity of these clones did not vary with ploidy. Increased smooth muscle cell ploidy is, therefore, associated with a decreased rate of proliferation. The emergence of smooth muscle cells with polyploid DNA content under normal and pathologic conditions is probably due to mitotic polyploidization without net cell proliferation and may be related to the need for expression of differentiated functions.     

Human B cell development. II. Subpopulations in the human fetus

In man, during fetal development the B cell populations show distinct phenotypes at different tissue sites. The pre-B and B lymphocytes of the fetal liver and bone marrow express IgM and B cell markers, B1 (CD20) and BA-1 (CD24). These "early" cells are negative with a number of other reagents, anti-IgD, RFB4 (CD22), RFB6 (CD21), and RFA-2, which on the other hand recognize peripheral B cells. These peripheral B lymphocytes in the developing fetus are heterogeneous. The diffusely distributed B cells in the earliest lymph node samples, 16 to 17 wk of gestational age, and from 16 to 21 wk in the spleen, are strongly IgM+ (IgD+,RFB4+,RFB6+, and RFA-2+) but lack T cell-associated markers such as T1 (CD5, p 67,000 dalton equivalent of murine Ly-1) and Tü-33. In fetal lymph nodes, primary nodules develop around the follicular dendritic (FD) cells from 17 wk onward, and contain a virtually pure population of B cells; B1+,BA1+,RFB4+,RFB6+,RFA-2+, which simultaneously express IgM,IgD together with T1 (CD5), a T cell-associated antigen. A sizeable subpopulation of these IgM+,T1+ cells are also positive for Tü-33, another T cell-associated marker. In the spleen, the B cells of the IgM+,IgD+,T1+ type appear in smaller numbers and only relatively late around wk 22. These cells are diffusely distributed at first, and start accumulating around the small FD cell clusters as soon as these emerge about the 23rd gestational wk. At that time, the IgM+,T1+B cells can also be washed out from the peritoneal and pleural cavities. The T1+,IgM+B cells may represent the normal equivalent cells of B chronic lymphoid leukemia and centrocytic lymphoma, and appear to be the counterpart of Ly-1+,IgM+B cells in the mouse.     

Purification and distribution of a novel macrophage-specific calmodulin-binding glycoprotein

The murine macrophage-like cell line J774.16 and peritoneal exudate cells elicited with thioglycollate or starch contain a major calmodulin (CaM)-binding protein (CaMBP) which is absent in trifluoperazine-resistant variants of J774, resident peritoneal macrophages and peritoneal macrophages elicited with concanavalin A, lipopolysaccharide, proteose peptone, or Bacillus Calmette-Guerin. Resident murine peritoneal cells maintained in tissue culture for 3 days begin to accumulate this protein, as do human peripheral blood monocytes after 7 days of culture. A specific competitive displacement radioimmunoassay was developed with the use of a rabbit antiserum raised to the partially purified CaM-binding protein and [125I]CaM covalently cross-linked to the principal CaM-binding protein in the preparation. The radioimmunoassay confirmed the unique cellular distribution of this protein, suggesting that it may be a marker for certain stages of macrophage differentiation. Monoclonal antibodies were prepared, and one of these was used to further purify the protein by immunoaffinity chromatography. A protein of Mr 50,000 to 60,000 was isolated. The protein could be selectively adsorbed to wheat germ agglutinin agarose and subsequently eluted with N-acetyl glucosamine. This property, plus the sensitivity of the protein to endoglycosidase F, led to the conclusion that it is a glycoprotein. The cellular distribution, subcellular localization, and evidence for glycosylation suggest that this protein may be a macrophage-specific receptor with a high affinity for Ca2+-CaM.