The CD28 and CTLA-4 receptors associate with the serine/threonine phosphatase PP2A.

CD28 and CTLA-4 are related members of a family of T lymphocyte cell surface receptors that function to regulate T cell activation. We have found that the cytoplasmic domains of both CTLA-4 and CD28 can associate with members of the PP2A family of serine/threonine phosphatases. The association of PP2A with CD28 was negatively regulated by tyrosine phosphorylation of the CD28 cytoplasmic domain. Inhibition of PP2A activity in Jurkat leukemia T cells by treatment with okadaic acid or by expression of a dominant-negative mutant enhanced T cell activation induced by CD28 engagement. Interactions between cell surface receptors such as CTLA-4 and CD28 and serine/threonine phosphatases may represent a novel mechanism for modulating the intracellular signal transduction pathways associated with cell activation.

Regulation of cytotoxic T lymphocyte-associated molecule-4 by Src kinases.

Cytotoxic T lymphocyte-associated molecule-4 (CTLA-4) is a cell surface receptor expressed on activated T cells that can inhibit T cell responses induced by activation of the TCR and CD28. Studies with phosphorylated peptides based on the CTLA-4 intracellular domain have suggested that tyrosine phosphorylation of CTLA-4 may regulate its interactions with cytoplasmic proteins that could determine its intracellular trafficking and/or signal transduction. However, the kinase(s) that phosphorylate CTLA-4 remain uncharacterized. In this report, we show that CTLA-4 can associate with the Src kinases Fyn and Lck and that transfection of Fyn or Lck, but not the unrelated kinase ZAP70, can induce tyrosine phosphorylation of CTLA-4 on residues Y201 and Y218. A similar pattern of tyrosine phosphorylation was found in pervanadate-treated Jurkat T cells stably expressing CTLA-4. Phosphorylation of CTLA-4 Y201 in Jurkat cells correlated with cell surface accumulation of CTLA-4. CTLA-4 phosphorylation induced the association of CTLA-4 with the tyrosine phosphatase SHP-2, but not with phosphatidylinositol 3-kinase. In contrast, Lck-induced phosphorylation of CD28 resulted in the recruitment of phosphatidylinositol 3-kinase, but not SHP-2. These findings suggest that phosphorylation of CD28 and CTLA-4 by Lck activates distinct intracellular signaling pathways. The association of CTLA-4 with Src kinases and with SHP-2 results in the formation of a CTLA-4 complex with the potential to regulate T cell activation.

Proliferation-associated differences in the spatial and temporal expression of gap junction genes in rat liver.

After a 70% partial hepatectomy (PH), the steady-state levels of Connexin (Cx)32, Cx26, and Cx43 messenger RNA (mRNA) transcripts each displayed unique patterns of temporal expression. Within 1 hour after surgical resection, increased expression of all three Cx mRNAs was observed. Subsequently, the level of Cx32 mRNA transcripts transiently decreased to a nadir at 12 hours. Comparisons of the spatial changes with previously reported hepatocyte proliferation kinetics induced by PH demonstrated that hepatocytes before S-phase "remodel" their GJs. Within 1 to 5 hours post-PH, midzonal hepatocytes exhibited diffuse membrane staining different from the normal punctate distribution. Subsequently, midzonal hepatocytes expressed colocalized punctate Cx32 and Cx26 immunostaining. Because the changes occurred in midzonal hepatocytes before 24 hours post-PH, near the peak of hepatocyte DNA synthesis, these findings indicate that Cx26 is enhanced in hepatocytes before the onset of S-phase. In contrast to the restricted expression of Cx43 in Glisson's capsule in adult liver, Cx43 protein and mRNA were enhanced specifically in proliferating bile duct and perisinusoidal cells post-PH. PH performed during continuous administration of 2-acetylaminofluorene (AAF) prevented changes in Cx32 and Cx26 staining observed in the absence of AAF. Proliferating oval cells were found to express diffuse Cx43 immunoreactivity. On day 11 post-PH and AAF, basophilic hepatocytes displayed both punctate Cx32 and Cx26 staining, whereas bile ducts and perisinusoidal cells expressed Cx43. These findings indicate that alterations in Cx32 and Cx26 expression occur rapidly in hepatocytes stimulated to proliferate and that several nonparenchymal liver cell types upregulate Cx43 expression when induced to proliferate. Differentiation of oval cells into basophilic hepatocytes resulted in their expression of Cx32 and Cx26.

Differences in the expression of connexin genes in rat hepatomas in vivo and in vitro.

Gap-junctional intercellular communication (GJIC) in normal rat liver cells involves at least three different connexins (Cxs)--Cx32, Cx26, Cx43--depending on the cell type, position in the lobule, or both. Whereas rat hepatocyte primary cultures expressed Cx32 and Cx26 as observed in vivo, cell lines derived from normal rat liver (WB-F344, Clone 9, RLEC, and BRL) expressed Cx43 and to a lesser extent Cx26. Hepatoma cells propagated in vitro were either deficient in GJIC and Cx expression (7777, 8994, H4IIE-C3) or communicated via gap junctions composed of Cx43 protein (N1S1-67, 9618A). Analysis of neoplasms that resulted from injection of hepatoma cells into rat femoral muscle showed differences in Cx expression when compared with cells grown in vitro. Whereas hepatoma cells 7777 and H4IIE-C3 failed to express Cx mRNAs in culture, these cells transplanted in vivo expressed levels of Cx32 mRNA comparable to those in normal liver. However, detectable Cx32 immunostaining was observed in less than 5% of the neoplastic cells in vivo. These results indicate that Cx32 protein was posttranscriptionally downregulated in 7777 and H4IIE-C3 tumor cells. Unexpectedly, 9618A cells expressed Cx43 mRNA and protein in cell culture but expressed Cx32 mRNA in vivo. In contrast, N1S1 transplants continued to express Cx43 mRNA and protein in vivo. Unlike the punctate Cx43 staining observed in suspension cultures of N1S1 cells, diffuse intracellular Cx43 staining was observed in N1S1-derived neoplasms in vivo, although the electrophoretic pattern of Cx43 isolated from N1S1 tumors grown in vivo (43 kDa) was different from that observed in suspension cell cultures (43 and 45 kDa). Thus, the findings reported here demonstrate that Cx expression in hepatoma cells depends on the environment, whether in vivo or in vitro, in which the cells are propagated.

Colocalized alterations in connexin32 and cytochrome P450IIB1/2 by phenobarbital and related liver tumor promoters.

Direct intercellular signal transduction is achieved by the passage of small molecules through gap junctions (GJ). Previous studies in our laboratory showed that the liver tumor promoter phenobarbital (PB) reversibly decreases the abundance of the GJ protein connexin32 (Cx32) in both preneoplastic-altered hepatic foci and centrolobular hepatocytes (M. J. Neveu et al., Cancer Commun., 2: 21-31, 1990). Because the inhibitory effects of PB on GJ intercellular communication are prevented by the nonspecific cytochrome P-450 inhibitor SKF-525A (J. E. Klauning, et al., Toxicol. Appl. Pharmacol., 102: 533-563, 1990), we investigated whether alterations in Cx32 are coincident with changes in the major PB-inducible cytochrome P-450, termed b/e or IIB1/2. Immunostaining of liver cryosections from rats fed dietary PB demonstrated that centrolobular hepatocytes that exhibit reduced Cx32 express enhanced cytochrome P450IIB1/2 protein. In contrast, no change in the periportal distribution of connexin26 immunoreactivity was found in PB-treated rats. In addition, rats were treated with the structurally related barbiturates pentobarbital, amobarbital, barbital, and barbituric acid. We found that the extent of the hepatic lobule occupied by coincident centrolobular alterations in Cx32 and P-450 staining correlates with the ability of the compounds to promote liver oncogenesis. To determine the molecular mechanisms responsible for the modification in Cx32 staining, we examined the mRNA and protein levels of Cx32 and P450IIB1/2 in total-tissue homogenates from PB-treated rats. Northern blotting demonstrated thatdietary PB dramatically induced P-450IIB1 mRNA, but the same RNA samples failed to show alterations in Cx32 steady-state transcripts. Consistent with these findings, the level of Cx32 protein in total liver homogenates did not change in rats chronically fed PB. Examination of Cx32 solubility in 20 mM NaOH demonstrated that PB treatment results in the generation of a NaOH-soluble form of Cx32 (i.e., 47 kDa). In addition, trypsinized paraffin-embedded liver sections from PB-treated rats exhibited diffuse cytoplasmic Cx32 staining that was restricted to centrolobular cells. Our results show that PB and related barbiturate tumor promoters reversibly down-regulate punctate Cx32 staining in centrolobular hepatocytes posttranslationally, possibly through modification(s) in the transport, assembly, and/or turnover of GJs.

Multiple mechanisms are responsible for altered expression of gap junction genes during oncogenesis in rat liver.

Although several abnormalities in gap junction (GJ) structure and/or function have been described in neoplasms, the molecular mechanisms responsible for many of the alterations remain unknown. The identification of a family of GJ proteins, termed connexins, prompted this study of connexin32 (Cx32), connexin26 (Cx26) and connexin43 (Cx43) expression during rat hepatocarcinogenesis. Using antibody, cDNA and cRNA probes, we investigated connexin mRNA and protein expression in preneoplastic and neoplastic rat livers. In normal liver, Cx32 is expressed in hepatocytes throughout the hepatic acinus, Cx26 is restricted to periportal hepatocytes, and Cx43 is expressed by mesothelial cells forming Glisson's capsule. Most preneoplastic altered hepatic foci generated by diethylnitrosamine (DEN) initiation and either phenobarbital (PB) or 2,3,7,8-dichlorodibenzo-p-dioxin (TCDD) promotion exhibited decreased Cx32 or increased Cx26 staining. Foci from either protocol failed to display Cx43 immunoreactivity. In the majority of PB-promoted foci, Cx32 immunoreactivity decreased independently of changes in mRNA abundance. Continuous thymidine labeling, following cessation of PB promotion, showed that downregulation of Cx32 staining is reversible in foci that are promoter-dependent for growth, but irreversible in lesions that are promoter-independent for growth. Hepatic neoplasms from rats initiated with DEN and promoted with PB or TCDD also displayed modified connexin expression. While all 24 neoplasms studied were deficient in normal punctate Cx32 and Cx26 staining, altered cellular localization of these proteins was apparent in some tumors. Immunoblotting of crude tissue extracts revealed that neoplasms with disordered Cx32 staining showed immunoreactive bands with altered electrophoretic mobility. These observations show that hepatomas may downregulate Cx32 expression through changes in the primary structure of Cx32 or by post-translational modifications. Northern blotting of total tumor mRNAs failed to demonstrate consistent changes in the abundance of Cx32, Cx26 or Cx43 transcripts. Some tumors expressed steady-state transcripts without observable immunoreactivity, indicating that some hepatomas downregulate connexin immunoreactivity independently of mRNA abundance. Increased levels of Cx43 mRNA and protein were found in several neoplasms, but immunostaining was always localized to nonparenchymal cells. Areas of bile duct proliferation and cholangiomas displayed Cx43 staining, whereas, cholangiocarcinomas were deficient in immunoreactivity. These findings show that alterations in the expression of connexins, by either downregulation or differential induction, represent common modifications during hepatocarcinogenesis. Although our results imply that connexins represent useful markers for the boundary between tumor promotion and progression, preneoplastic and neoplastic rat hepatocytes fail to use a common mechanism to modify connexin expression.

Specificity of gap junction communication among human mammary cells and connexin transfectants in culture.

In a previous paper (Lee et al., 1992), it was shown that normal human mammary epithelial cells (NMEC) express two connexin genes, Cx26 and Cx43, whereas neither gene is transcribed in a series of mammary tumor cell lines (TMEC). In this paper it is shown that normal human mammary fibroblasts (NMF) communicate and express Cx43 mRNA and protein. Transfection of either Cx26 or Cx43 genes into a tumor line, 21MT-2, induced the expression of the corresponding mRNAs and proteins as well as communication via gap junctions (GJs), although immunofluorescence demonstrated that the majority of Cx26 and Cx43 proteins present in transfected TMEC was largely cytoplasmic. Immunoblotting demonstrated that NMEC, NMF, and transfected TMEC each displayed a unique pattern of posttranslationally modified forms of Cx43 protein. The role of different connexins in regulating gap junction intercellular communication (GJIC) was examined using a novel two-dye method to assess homologous and heterologous communication quantitatively. The recipient cell population was prestained with a permanent non-toxic lipophilic dye that binds to membranes irreversibly (PKH26, Zynaxis); and the donor population is treated with a GJ-permeable dye Calcein, a derivative of fluorescein diacetate (Molecular Probes). After mixing the two cell populations under conditions promoting GJ formation, cells were analyzed by flow cytometry to determine the percentage of cells containing both dyes. It is shown here that Cx26 and Cx43 transfectants display strong homologous communication, as do NMEC and NMF. Furthermore, NMEC mixed with NMF communicate efficiently, Cx26 transfectants communicate with NMEC but not with NMF, and Cx43 transfectants communicate with NMF. Communication between Cx26 TMEC transfectants and NMEC was asymetrical with preferential movement of calcein from TMEC to NMEC. Despite the presence of Cx43 as well as Cx26 encoded proteins in the GJs of NMEC, few Cx43 transfectants communicated with NMEC. No heterologous GJIC was observed between Cx26- and Cx43-transfected TMEC suggesting that heterotypic GJs do not form or that Cx26/Cx43 channels do not permit dye transfer.

Reversible alteration in the expression of the gap junctional protein connexin 32 during tumor promotion in rat liver and its role during cell proliferation.

Although numerous biochemical markers can identify putative preneoplastic altered hepatic foci (AHF) in rat liver, no consistent pattern of expression during hepatocarcinogenesis has emerged. Using quantitative stereologic analyses we demonstrated that decreased expression of the major hepatocyte gap junction protein, connexin 32 (Cx32), in rat AHF is a consistent observation in several protocols of multistage hepatocarcinogenesis. This change was observed after initiation by either ethylnitrosourea (ENU) or diethylnitrosamine (DEN), followed by promotion with phenobarbital (PB), dioxin, chlorendic acid, C.I. Solvent Yellow, or tamoxifen. AHF generated by Wy-14,643, ciprofibrate, and a choline/methionine-deficient dietary regimen also showed decreased Cx32 expression. The decrease of Cx32 in AHF was rapidly reversible after withdrawal of PB, and this change preceded a reduction in placental isozyme of glutathione-S-transferase (GST) expression in the same AHF. Within 20 days of withdrawal, fewer than 4% of GST-positive AHF were Cx32 deficient, while the volume of total AHF decreased 30%. Chronic PB treatment also resulted in a reversible decrease in Cx32 specifically in mid- and centro-lobular hepatocytes. Continuous thymidine labeling demonstrated that Cx32 could be uncoupled from the cell cycle, suggesting that some liver promoters may act directly to alter the expression of Cx32. These observations suggest that a decrease in Cx32 content was a relatively common epigenetic change in AHF induced during hepatocarcinogenesis by a number of initiating and promoting agents but that this change was not sufficient for carcinogenesis. This change, however, may be necessary for the mechanism(s) of tumor promotion, since Cx32-positive AHF did not proliferate as readily as Cx32-deficient AHF.

Expression of the c-raf protooncogene, gamma-glutamyltranspeptidase, and gap junction protein in rat liver neoplasms.

Female Harlan Sprague-Dawley and F-344 rats were subjected to a 70% hepatectomy and 18 h later given one dose of 30 mg diethylnitrosamine/kg body weight. Beginning 1 week later, the animals were fed a diet containing 0.05% phenobarbital. Groups of rats were sacrificed 6 and 15 months later, and the livers were either frozen for cryostat sectioning or used to isolate RNA. Primary liver tumors present in these animals were used for RNA isolation, and a portion was taken for histopathological analysis. Eleven of 13 primary lesions, consisting of either neoplastic nodules or hepatocellular carcinomas, showed elevated levels of mRNA for the c-raf protooncogene. Increased c-raf mRNA in these tumors appeared to be unrelated to their cellular proliferative status inasmuch as the levels of c-raf mRNA did not correlate with levels of H4 histone mRNA. Decreased expression of the major rat liver gap junction protein mRNA was observed in all of the primary tumors. Immunocytochemical analysis using an anti-gap junction antibody revealed a decrease in gap junction immunoreactivity in some but not all preneoplastic focal lesions. All preneoplastic foci having positive gamma-glutamyltranspeptidase enzyme staining also exhibited a marked increase in gamma-glutamyltranspeptidase mRNA as determined by in situ hybridization. The possible relation of alterations of the mRNA levels of c-raf and the gap junction protein to the further development of preneoplastic foci is discussed.