Stable knockdown of CREB, HIF-1 and HIF-2 by replication-competent retroviruses abrogates the responses to hypoxia in hepatocellular carcinoma.

The fast proliferation of tumor cells develops faster than the vasculature, resulting, in most malignant tumors, in generation of hypoxic regions. Hypoxia renders solid tumors resistant to radiation and chemotherapeutics while providing opportunities for tumor-selective therapies targeting tumor hypoxia. Here we exploit two properties of tumors: propagation of tumor cells and ongoing generation of hypoxic regions to construct a system that preferentially leads to the death of tumor cells and thus hinders tumor growth. We constructed murine leukemia virus replication-competent (RCR) viruses that infect only propagating cells. These viruses express small hairpin RNAs (shRNAs) targeting cyclic AMP-response-element binding protein (CREB), hypoxia-inducible factors 1 (HIF)-1 or HIF-2 individually or all three together (X3). These viruses efficiently infected in vitro human hepatocellular carcinoma (HepG2 and FLC4) cells and established persistence of the virus and knocked down the expression of the regulators of the hypoxia-responding genes. Knockdown of either HIF-1 or CREB or both in hypoxia reduced the expression of hypoxia-response elements- and CRE-mediated gene expression, diminished cell proliferation and increased caspase-3 activity. We did not detect any significant effect of the efficiently knocked down HIF-2 on any of the functions tested in vitro. Moreover, severe combined immunodeficiency mice implanted subcutaneously with HepG2 stably infected with recombinant RCRs showed reduction of tumor growth and vascular endothelial growth factor expression, and no hypoxia-guided neovascularization. Combined treatment (RCRs+doxorubicin) improved efficacy in the context of in vitro hypoxia and in vivo (with either vACE-CREB or vACE-X3). This synergistic effect may lead to an improved efficacy and safety profile of the treatment that may result in fewer side effects.

Expression of an anti apoptotic recombinant short peptide in mammalian cells.

Understanding the mechanisms of the apoptotic and anti apoptotic processes may lead to a better way to control these cascades. Here we demonstrated for the first time the feasibility to express a short functional peptide in mammalian cells that abrogates the apoptosis cascade through interference with the proteolytic activity of the initiator caspase 9 and the executing caspase 3 enzymes. The expression of a short peptide that includes the pseudo-substrate motif of the apoptosis inhibitor protein P35 (Asp-Gln-Met-Asp) leads to the abrogation of cell death induced through either the mitochondrial or the death receptors pathways. Short open reading frames have been detected in several mammalian mRNAs, primarily upstream of the main long reading frame (uORFs), however, direct evidence for de-novo peptides translation has not been provided. Utilizing biochemical and imaging techniques we demonstrate here that the functional recombinant peptide was localized to the cytpoplasmic fraction of the cell. In conclusion, this work demonstrates that ribosomes recognize short ORFs to translate stable short recombinant peptides in mammalian cells. Expression of these intracellular peptides results in the knock down of apoptotic processes to generate apoptosis resistant stable cells.

Two cysteine residues in the DNA-binding domain of CREB control binding to CRE and CREB-mediated gene expression.

The cAMP-responsive element-binding protein (CREB) has been implicated in the regulation of numerous physiological functions including those of several hypoxia-responding genes. All CREB transcription-regulated genes harbor the eight base-pair cAMP-responsive element (CRE) or the seven base-pair AP-1 sequence. Utilizing mutational analysis and biochemical assays, we found that reduction of two cysteine residues located in the DNA-binding basic domain of CREB, enhances the binding efficiency of CREB to DNA and regulates CRE-mediated gene expression. Substitution of these residues to serine renders insensitivity to reduction, hypoxia and to the sulfhydryl-specific modifying agent, N-ethylmaleimide. These substitutions enhance the binding of CREB to its cognate DNA sites under oxidative conditions, and of the CREB-dependent gene expression during normoxia. These findings are supported by results of molecular modeling of the CREB-CRE interactions. We also found that HTLV-1 Tax enhancement of CREB binding to the cellular and the viral DNA sites and activation of the CRE-dependent gene expression are independent of CREB activation exerted by redox conditions. The genetic biochemical and molecular modeling presented in this work indicate that the two cysteine residues in the bZIP domain of CREB regulate the binding efficiency of CREB to its cognate DNA sites and as a consequence the activation of CREB-mediated gene expression.

Imaging transgene expression in live animals.

Monitoring the expression of therapeutic genes in targeted tissues in disease models is important to assessing the effectiveness of systems of gene therapy delivery. We applied a new light-detection cooled charged-coupled device (CCCD) camera for continuous in vivo assessment of commonly used gene therapy delivery systems (such as ex vivo manipulated cells, viral vectors, and naked DNA), without the need to kill animals. We examined a variety of criteria related to real-time monitoring of luciferase (luc) gene expression in tissues including bone, muscle, salivary glands, dermis, liver, peritoneum, testis, teeth, prostate, and bladder in living mice and rats. These criteria included determination of the efficiency of infection/transfection of various viral and nonviral delivery systems, promoter specificity, and visualization of luciferase activity, and of the ability of luciferin to reach various organs. The exposure time for detection of luc activity by the CCCD camera is relatively short (approximately 2 minutes) compared with the intensified CCD camera photon-counting method (approximately 15 minutes). Here we transduce a variety of vectors (such as viruses, transfected cells, and naked DNA) by various delivery methods, including electroporation, systemic injection of viruses, and tail-vein, high-velocity-high-volume administration of DNA plasmids. The location, intensity, and duration of luc expression in different organs were determined. The distribution of luciferin is most probably not a barrier for the detection of in vivo luciferase activity. We showed that the CCCD photon detection system is a simple, reproducible, and applicable method that enables the continuous monitoring of a gene delivery system in living animals.

Transcriptional regulation of the human sterol 27-hydroxylase gene (CYP27) and promoter mapping.

Recent evidence suggests that sterol 27-hydroxylase may play a role in cholesterol homeostasis and affect atherogenesis. The major objective of the study was to map and characterize the sterol 27-hydroxylase (CYP27) promoter region. Here we show that CYP27 gene has a TATA-less promoter and transcription initiates at a cluster of sites. The basic promoter is located between -166 and -187 bp from the translation initiation site. Possible positive transcription regulation sites are located at position -187 to -320 and -857 to -1087 bp. A negative transcription regulator site is located in position -320 to -413 bp. An enhancer sequence is located upstream to position -1087. CYP27 is upregulated by dexamethasone and downregulated by cyclosporin A and cholic acid. The dexamethasone responsive element is located between 1087 and 678 bp upstream to the putative ATG. Cyclosporin A affects bile acid metabolism by repressing CYP27 at the transcriptional level. The cyclosporin A- responsive element is mapped to between 1087 and 4000 bp upstream of the ATG. Cholic acid represses sterol 27-hydroxylase mRNA level by affecting the stability of its mRNA. The results obtained here imply that CYP27 has a potentially important role in cholesterol homeostasis in human cells, and is regulated by several substances that were previously shown to affect bile acid metabolism.

Emergence of multiple human cytomegalovirus ganciclovir-resistant mutants with deletions and substitutions within the UL97 gene in a patient with severe combined immunodeficiency.

Infection with multiple ganciclovir-resistant human cytomegalovirus mutants, containing different substitutions and deletions in the UL97 gene, was found in a patient with severe combined immunodeficiency (SCID) within 3 weeks of ganciclovir therapy. A novel 11-codon deletion at positions 590 to 600 was identified. These unique findings may be related to the nature of the immunodeficiency in the SCID patient.

A sensitive and versatile bioluminescence bioassay for HIV type 1 based on adenoviral vectors.

The construction and characterization of a versatile bioassay for the quantification of HIV-1 viral infection and HIV-1 Tat protein activity based on recombinant adenoviral vectors carrying an HIV LTR-driven luciferase reporter gene is described. The assay system consists of a set of two adeno-reporter vectors, one of which is responsive to HIV-1 Tat protein activity, and the second of which is not, by virtue of a deletion of the TAR site within the HIV LTR. This configuration of the reporter genes allows one to distinguish Tat-specific activation from Tat-non-specific HIV LTR-mediated gene expression. The adenoviral HIV LTR-mediated luciferase gene expression is highly responsive to Tat and increases linearly with increasing levels of HIV-1 infection, reaching levels of between 3- and 1000-fold induction. The adeno-reporter viruses can be utilized to detect Tat activity and HIV-1 infection in a wide range of cell types, including 293, CEM, HUT-78, Jurkat, and HeLa-derived cell lines. The resulting bioassay is convenient, sensitive, and readily adaptable to automated procedures. These characteristics of the adeno-reporter assay make it a valuable reagent for studies of HIV infection and for analysis of HIV-inhibitory agents.

Gene regulation mediated by interaction between HTLV-1 promoter elements and transcription factors Tax and CREB.

In this work we examine the role of three genetic control components in the regulation of HTLV-1 transcription: cyclic AMP-responsive element (CRE)-binding protein (CREB), the HTLV-1 trans-activator Tax, and the three Tax-responsive elements (TREs). We demonstrate that the in vivo efficiency of the HTLV-1 promoter basal expression in cell culture depends on the spacing between the three TRE elements, located at the HTLV-1 LTR (long terminal repeat), whereas the level of transcription activation mediated by Tax is affected by the number of TREs. In the presence of only one TRE, the enhancement of expression by Tax is affected by the distance between the single TRE and the transcription start site. Following CREB binding to the LTR, additional DNase I hypersensitive sites are generated in the region between the two distal TREs (I and II), while in the presence of Tax, such sites are generated also in the region between TREs II and III. Neither cooperative binding of CREB to the TREs nor preferential binding of CREB to a particular TRE was observed. Tax binding to the CREB/TRE complex does not change the DNase I protection pattern. Taken together, these results suggest that the basal CREB-mediated transcription is determined by the number and the position of the viral TREs relative to each other. Tax protein stabilizes the protein/DNA complex and suppresses the spacing limitations, probably by bridging between the CREB/TRE complexes and the basal initiation transcription complex.

Characterization of the human cytomegalovirus UL97 gene product as a virion-associated protein kinase.

The cellular localization and virion association of the human cytomegalovirus (HCMV) UL97 protein were studied. UL97 protein demonstrated early nuclear localization followed by late perinuclear accumulation. It was found to be a structural virion constituent detected in all three enveloped forms of extracellular viral particles and shown to be phosphorylated by the virion-associated protein kinase. UL97 protein immunoprecipitated from virions and from infected cells demonstrated protein kinase activity manifested by autophosphorylation. This activity was reduced in the presence of a ganciclovir-resistance mutation at residue 460, implicated in nucleotide binding. A mutant virus, from which the proposed UL97 kinase catalytic domain had been deleted, could not be propagated in the absence of a helper wild-type virus. The characterization of UL97 protein as a virion-associated protein kinase which appears essential for viral replication, provides further insight into HCMV replication and could identify a potential novel target for antiviral therapy.

Early emergence of ganciclovir-resistant human cytomegalovirus strains in children with primary combined immunodeficiency.

Children with primary combined immunodeficiency (CID) and human cytomegalovirus (HCMV) infection often deteriorate despite antiviral therapy. In this study, the emergence of ganciclovir-resistant strains was examined in 6 children with CID and HCMV infection, using sequence analysis of the HCMV UL97 gene and virus susceptibility assays. Mutations in the proposed ATP binding site associated with ganciclovir resistance were found in 4 of the 6 children. In 1 patient with B severe CID, an unusual multiplicity of mutations was found in the UL97 substrate binding domain between aa 590-606. All mutations were detected within 10 days to 3 weeks from initiation of therapy. The emergence of resistant strains in children with CID appears earlier than in other groups of HCMV-infected patients. These findings may have relevance to the cellular pathways involved in viral DNA repair and mutagenesis, and they indicate the need for early and frequent genotypic monitoring and prompt therapeutic modification in this patient population.

Mycoplasmas regulate HIV-LTR-dependent gene expression.

Mycoplasmas have been incriminated in setting the stage for HIV infection and full-blown AIDS. We tested the possible involvement of mycoplasmas in activation of HIV. Two cell lines, 293 fibroblasts and Jurkat CD4+ T-cells, transfected with plasmids harboring a transcription fusion construct between HIV-long terminal repeat (HIV-LTR) and either luc or cat genes, were infected with several mycoplasmas (M. fermentans; M. penetrans, M. pirum and Ureaplasma urealyticum) and the reporter gene expression was monitored. The data presented here suggest that mycoplasmas, and specifically their membranes, play a role in the activation of HIV-LTR mediated transcription.

Translation efficiency of the human T-cell leukemia virus (HTLV-2) gag gene modulates the frequency of ribosomal frameshifting.

The regulation of ribosomal frameshifting during translation of the polycistronic mRNA of human T-cell leukemia virus (HTLV) was studied in a cell-free system. Translation inhibitors such as cycloheximide and puromycin antibiotics were much more effective in blocking the synthesis of the frameshift polypeptide Gag-Pro than the synthesis of the Gag product. The preferential inhibition of the frameshift polypeptide Gag-Pro by the two antibiotics was not a reflection of the different sizes of the two gene products, but rather a consequence of the effect of the inhibitors on ribosomal translation efficiencies. To further analyze the effect of translation efficiencies on ribosomal frameshifting, we compared the translation of 5'-capped RNA to noncapped RNA. The translation of 5'-capped RNA was higher, as expected. Consequently, ribosomal frameshifting producing the Gag-Pro polypeptide was enhanced when compared to the translation of noncapped RNA. Taken together these results indicate that efficiencies of translation, in conjunction with the cis regulatory genetic elements at the frameshift sites, determine the ratio of the polypeptides Gag, Gag-Pro, and Gag-Pro-Pol produced in the HTLV-infected cell. Thus, physiological changes which affect the cellular translation machinery may alter the optimal ratio of these three polyprotein products needed for virus maturation.

Transactivation of human T-cell leukemia virus type 1 by herpes simplex virus type 1.

HTLV-1 transcription depends upon activation by the HTLV-1 tax gene product. In addition, various substances and cellular transcription factors are also known to activate the HTLV-1 long terminal repeat (LTR)-mediated transcription in the absence of Tax. In this work we demonstrate that infection of either Jurkat or 293 cell lines with herpes simplex I (HSV-1), a widespread infectious virus of humans, activates HTLV-1 LTR-mediated gene expression. Further investigation revealed that each of the immediate-early (IE) gene products--ICPO, ICP4, and ICP27--of HSV-1 transactivates the HTLV-1 LTR-mediated gene expression in the absence of Tax. The HSV-1 activation is additive to Tax activation in its presence in the cell. Three 21 base repeats upstream of the TATA box are known as the TAX responsive elements (TRE). Recombinant HTLV-1 minimal promoter composed of the HTLV-1 TATA box fused to a synthetic 21 base TRE is responsive to Tax but not to HSV-1 activation. It thus can be concluded that HSV-1 IE gene products and Tax transactivates HTLV-1 LTR mediated gene expression through different transcription complexes. The results presented in this work may point to one possible way for the transition of HTLV-1 from a quiescent to an actively replicating stage.

The sequences of and distance between two cis-acting signals determine the efficiency of ribosomal frameshifting in human immunodeficiency virus type 1 and human T-cell leukemia virus type II in vivo.

We have analyzed in cell culture the sequence elements that control the level of ribosomal frameshifting in the human T-cell leukemia virus type II (HTLV-2) gag-pro junction. The slippery sequence of HTLV-2 is sufficient to dictate a basal level of frameshifting. This level is enhanced by its upstream sequence context and by the downstream stem-loop structure which is located at an optimal distance of 7 bases. Frameshifting in human immunodeficiency virus gag-pol is similar to that of HTLV-2 gag-pro. However, experiments using hybrid cassettes of HTLV-2 and human immunodeficiency virus type 1 frameshift elements show that while the slippery sequence of HTLV-2 is less efficient, the stem-loop structure is a more efficient enhancer.

Two cis-acting signals control ribosomal frameshift between human T-cell leukemia virus type II gag and pro genes.

The open reading frame of the human T-cell leukemia virus type II pro gene is arranged at a -1 position relative to the gag gene. Synthesis of the Gag-Pro fusion polyprotein is facilitated by ribosomal frameshift into the reading frame of the pro gene. Cloning of a synthetic 41-bp oligonucleotide corresponding to the gag-pro junction within a heterologous gene (nef of human immunodeficiency virus type I) and mutation analysis revealed that two cis-acting signals, an adenosine residue stretch and a dyad symmetry sequence, flanking the UAA termination codon, are required for efficient ribosomal frameshifting between gag and pro. The stability of the stem-loop structure is crucial for frameshifting.

An RNA secondary structure juxtaposes two remote genetic signals for human T-cell leukemia virus type I RNA 3'-end processing.

Sequence analysis of the human T-cell leukemia virus type I (HTLV-I) long terminal repeat (LTR) does not reveal a polyadenylation consensus sequence, AAUAAA, close to the polyadenylation site at the 3' end of the viral RNA. Using site-directed mutagenesis, we demonstrated that two cis-acting signals are required for efficient RNA processing in HTLV-I LTR: (i) a remote AAUAAA hexamer at a distance of 276 nucleotides upstream of the polyadenylation site, and (ii) the 20-nucleotide GU-rich sequence immediately downstream from the poly(A) site. It has been postulated that the folding of RNA into a secondary structure juxtaposes the AAUAAA sequence, in a noncontiguous manner, to within 14 nucleotides of the polyadenylation site. To test this hypothesis, we introduced deletions and point mutations within the U3 and R regions of the LTR. RNA 3'-end processing occurred efficiently at the authentic HTLV-I poly(A) site after deletion of the sequences predicted to form the secondary structure. Thus, the genetic analysis supports the hypothesis that folding of the HTLV-I RNA in the U3 and R regions juxtaposes the AAUAAA sequence and the poly(A) site to the correct functional distance. This unique arrangement of RNA-processing signals is also found in the related retroviruses HTLV-II and bovine leukemia virus.

A bioluminescence assay for gene expression by continuously growing mammalian cells: application for detection of human immunodeficiency virus type 1 (HIV-1).

An in-situ assay for monitoring regulated gene expression in continuously growing mammalian cells is described. This technique can be used for the detection of the transactivator (Tat) protein in human immunodeficiency virus(HIV)-infected cells. Human kidney cells 293, harboring the luc gene, and fused to the HIV-1 long terminal repeat, were isolated and served as tester cells. Tat is supplied by transfection with a tat-carrying plasmid, or alternatively by addition of Tat-containing cell extracts, made from virus-infected or plasmid-transfected cells. Light emitted from the tester cells is recorded on film continuously, or by a photo sensor. Transactivation by HIV Tat results in a pronounced increase in light emission from the tester cells (up to 3000-fold). This assay, which detects HIV-specific gene products, may be used as a diagnostic tool for the detection of active HIV present in peripheral blood.

cis Acting RNA sequences control the gag-pol translation readthrough in murine leukemia virus.

The pol gene of the Moloney murine leukemia virus (M-MuLV) is expressed as a Gag-Pol fusion protein through an in-frame suppression of the UAG termination codon located between the two genes. The role of nucleotide context in suppression was investigated, in a rabbit reticulocyte lysate translation system, using site-directed mutagenesis. The results indicate that the translational readthrough is mediated by at least 50 bases long RNA sequence located 3' to the gag UAG termination codon. Within this sequence a short purine-rich sequence adjacent to the amber codon, highly conserved among different retroviruses, appears essential for M-MuLV suppression. Two alternative putative stem and loop like RNA structures can be drawn at the gag-pol junction, one abutting the gag UAG codon, and the second downstream to it. None of these structures appears to be important to the suppression process.

Expression, purification, and crystallization of the HIV-1 reverse transcriptase (RT).

The HIV-1 pol gene proteins (protease, reverse transcriptase, and endonuclease) were expressed in Escherichia coli N4830-1 by the use of the inducible expression vector pWS60 into which the pol gene was inserted. The p66/p51 heterodimer of reverse transcriptase (RT) was isolated in a highly pure and active form. Crystals of the p66/p51 heterodimer were obtained by the vapor diffusion hanging drop technique. The present crystal quality is still not adequate for high resolution X-ray investigation.

Translation of gag, pro, and pol gene products of human T-cell leukemia virus type 2.

Sequence analysis of human T-cell leukemia proviral DNA revealed three open reading frames arranged at a -1 position relative to one another. On the basis of homology to other retroviruses, these open reading frames were assigned to the gag, pro, and pol genes. To characterize the primary protein products of these genes and their modes of synthesis, a DNA clone of human T-cell leukemia virus type 2 was transcribed and translated in vitro. Analysis of the viral proteins revealed three polyproteins with molecular masses of 58, 75, and 112 kilodaltons at relative frequencies of 100:13:0.9, respectively. These proteins were mapped on the viral genome by both internal deletions and 3'-end truncations at gag, pro, and pol, respectively. The results indicate that translation of the pol gene requires two independent frameshift events, and the readthrough frequencies at the two frameshift sites appeared to be similar.