Hepatitis B virus X protein activates the p38 mitogen-activated protein kinase pathway in dedifferentiated hepatocytes

Hepatitis B virus X protein (pX) is implicated in hepatocarcinogenesis by an unknown mechanism. Employing a cellular model linked to pX-mediated transformation, we investigated the role of the previously reported Stat3 activation by pX in hepatocyte transformation. Our model is composed of a differentiated hepatocyte (AML12) 3pX-1 cell line that undergoes pX-dependent transformation and a dedifferentiated hepatocyte (AML12) 4pX-1 cell line that does not exhibit transformation by pX. We report that pX-dependent Stat3 activation occurs only in non-pX-transforming 4pX-1 cells and conclude that Stat3 activation is not linked to pX-mediated transformation. Maximum Stat3 transactivation requires Ser727 phosphorylation, mediated by mitogenic pathway activation. Employing dominant negative mutants and inhibitors of mitogenic pathways, we demonstrate that maximum, pX-dependent Stat3 transactivation is inhibited by the p38 mitogen-activated protein kinase (MAPK)-specific inhibitor SB 203580. Using transient-transreporter and in vitro kinase assays, we demonstrate for the first time that pX activates the p38 MAPK pathway only in 4pX-1 cells. pX-mediated Stat3 and p38 MAPK activation is Ca(2+) and c-Src dependent, in agreement with the established cellular action of pX. Importantly, pX-dependent activation of p38 MAPK inactivates Cdc25C by phosphorylation of Ser216, thus initiating activation of the G(2)/M checkpoint, resulting in 4pX-1 cell growth retardation. Interestingly, pX expression in the less differentiated hepatocyte 4pX-1 cells activates signaling pathways known to be active in regenerating hepatocytes. These results suggest that pX expression in the infected liver effects distinct mitogenic pathway activation in less differentiated versus differentiated hepatocytes.     

The mitogenic function of hepatitis B virus X protein resides within amino acids 51 to 140 and is modulated by N- and C-terminal regulatory regions

The hepatitis B virus (HBV) X protein (pX) is implicated in hepatocarcinogenesis by an unknown mechanism. pX variants encoded by HBV genomes found integrated in genomic DNA from liver tumors of patients with hepatocellular carcinoma (HCC) generally lack amino acids 134 to 154. Since deregulation of mitogenic pathways is linked to oncogenic transformation, herein we define the pX region required for mitogenic pathway activation. A series of pX deletions was used to construct tetracycline-regulated pX-expressing cell lines. The activation of the mitogenic pathways by these pX deletions expressed in the constructed cell lines was measured by transient transreporter assays, effects on endogenous cyclin A expression, and apoptosis. Conditional expression of pX51-140 in AML12 clone 4 cell line activates the mitogenic pathways, induces endogenous cyclin A expression, and sensitizes cells to apoptosis, similar to wild-type (WT) pX. By contrast, pX1-115 is inactive, supporting the idea that amino acids 116 to 140 are required for mitogenic pathway activation. Moreover, this pX deletion analysis demonstrates that WT pX function is modulated by two regions spanning amino acids 1 to 78 and 141 to 154. The N-terminal X1-78, expressed via a retroviral vector in WT pX-expressing 4pX-1 cells, coimmunoprecipitates with WT pX, indicating this pX region participates in protein-protein interactions leading to pX oligomerization. Interestingly, pX1-78 interferes with WT pX in mediating mitogenic pathway activation, endogenous gene expression, and apoptosis. The C-terminal pX region spanning amino acids 141 to 154 decreases pX stability, determined by pulse-chase studies of WT pX and pX1-140, suggesting that increased stability of naturally occurring pX variants lacking amino acids 134 to 154 may play a role in HCC development.     

Bacterial expression and characterization of the CREB bZip module: circular dichroism and 2D 1H-NMR studies.

In this paper we describe the expression and purification from bacteria of the recombinant basic leucine zipper (bZip) domain of the cAMP response element binding protein, CREB327. The bZip peptide, CREB259-327, purified to near homogeneity, maintains the sequence-specific CRE site recognition demonstrated by in vitro competition assays. Alkylation of the three cysteine residues of CREB259-327 was employed to prevent aggregation of the peptide due to cysteine oxidation. The Kd of the purified native and modified CREB259-327 for the CRE site was determined by gel retardation assays to be on the order of 10(-7) M. We employed CD spectroscopy to study the folding properties of the native and modified CREB259-327. The CD analyses of the native/modified CREB259-327 peptide demonstrated a 20% increase in the alpha-helical content upon binding to the cAMP response-element. Only a 5% increase in the alpha-helical content of CREB259-327 is observed upon binding to the AP-1 site. This observation contrasts with CREB from the GCN4 protein (Weiss, M.A., et al., 1990, Nature 347, 575-578). In addition, the two-dimensional (2D) 1H-NMR studies of the bZip CREB peptide further support the distinct features of the CREB protein, in comparison to GCN4. Analysis by CD and 2D NMR of the dimerization domain of CREB suggests that the distinct DNA binding characteristics of CREB reside in the basic portion of the bZip module.     

Mutational analysis of human T-cell leukemia virus type 2 Tax.

A mutational analysis of human T-cell leukemia virus type 2 (HTLV-2) Tax (Tax-2) was performed to identify regions within Tax-2 important for activation of promoters through the CREB/ATF or NF-kappaB/Rel signaling pathway. Tax-2 mutations within the putative zinc-binding region as well as mutations at the carboxy terminus disrupted CREB/ATF transactivation. A single mutation within the central proline-rich region of Tax-2 disrupted the transactivation of the NF-kappaB/Rel pathway. Surprisingly, this mutation, which is thought to be in a separate activation domain, was suppressed by mutations within or around the putative zinc-binding region, suggesting an interaction between these two regions. These analyses indicate that the functional regions or domains important for transactivation through the CREB/ATF or NF-kappaB/Rel signaling pathway are similar, but not identical, in Tax-1 and Tax-2. Identification of these distinct Tax-2 mutants should facilitate comparative biological studies of HTLV-1 and HTLV-2 and ultimately lead to the determination of the functional importance of Tax trans-acting capacities in T-lymphocyte transformation by HTLV.