Induction of cellular genes is mediated by the Bel1 transactivator in foamy virus-infected human cells

To gain insight into human foamy virus (HFV; also called spumaretrovirus)-induced alterations of cellular genes, the expression profiles of defined genes in HFV-infected primary human cells were analyzed by cDNA array assays. Several distinct cellular genes activated by HFV infection were identified; the identities of the cellular genes were confirmed by RNA blot analyses. Compared with mock-infected controls, the concentrations of cellular Kip2, Egr-1, COUP-TF1, insulin-like growth factor II (IGF-II), and EphB3 mRNAs were significantly increased in HFV-infected cells and showed a gene-specific and time-dependent induction. Immunoblot analyses with antibodies against some of the cellular gene products revealed increased levels of the corresponding proteins. To investigate mechanisms of HFV-induced alterations in cellular gene expression, the capacity of known HFV genes to increase expression of defined cellular genes was analyzed by transient expression experiments. Plasmids that encode the HFV Bel1 transcriptional transactivator were necessary and sufficient to strongly increase expression of p57Kip2, IGF-II, and EphB3 genes in 293T cells. Potential mechanisms and consequences of activation of cellular genes during HFV infection and Bel1 transactivation of the Kip2 gene are discussed.

Sp1 is involved in a protein kinase C-independent activation of human T cell leukemia virus type I long terminal repeat by 12-O-tetradecanoylphorbol-13-acetate

The long terminal repeat (LTR) of human T cell leukemia virus type-I (HTLV-I) contains binding sites for several cellular transcription factors that can activate its expression independently of the viral transactivator Tax protein. In a previous study, we have shown that 12-O-tetradecanoylphorbol-13-acetate (TPA) induces a Tax-independent activation of the viral LTR expression. We have also noted that although most other TPA biological effects are attributed to its protein kinase C (PKC)-activating potential, this particular effect of TPA is PKC independent and antagonized by PKC activity. In addition, we have demonstrated that deletion of the ets-responsive region 1 (ERR-1) from the LTR abolishes its response to TPA. In the present study, we demonstrate that TPA exerts this effect by enhancing the binding of the Sp1 transcription factor to an Sp1-binding site located within ERR-1. This Sp1-binding stimulation was not diminished by a potent PKC-specific inhibitor like bisindolylmaleimide-I, indicating that it did not depend on PKC activity. However, no increase in Sp1 protein level could be detected in the TPA-treated cells, suggesting that TPA exerted its effect by a posttranslational modification of Sp1 protein rather than by stimulating its synthesis. Although the proximal Tax responsive 21-bp element also contains an Sp1-binding site, the present study shows that the modified Sp1 protein mediates the TPA effect on LTR only through the Sp1 site of the ERR-1.

Rapid syncytium formation between human T-cell leukaemia virus type-I (HTLV-I)-infected T-cells and human nervous system cells: a possible implication for tropical spastic paraparesis/HTLV-I associated myelopathy

Tropical spastic paraparesis/HTLV-I associated myelopathy (TSP/HAM), is characterized by infiltration of human T cell leukaemia virus type-I (HTLV-I)-infected T-cells, anti-HTLV-I cytotoxic T cells and macrophages into the patients' cerebrospinal fluid and by intrathecally formed anti-HTLV-I antibodies. This implies that the disease involves a breakdown of the blood-brain barrier. Since astrocytes play a central role in establishing this barrier, the authors investigated the hypothesis that the HTLV-I infected T cells disrupt this barrier by damaging the astrocytes. The present study revealed the HTLV-I-producing T cells conferred a severe cytopathic effect upon monolayers of astrocytoma cell line in co-cultures. Following co-cultivation, HTLV-I DNA and proteins appeared in the monolayer cells, but after reaching a peak their level gradually declined. This appearance of the viral components was proved to result from a fusion of the astrocytic cells with the virus-producing T cells, whereas their subsequent decline reflected the destruction of the resulting syncytia. This fusion could be specifically blocked by anti HTLV-I Env antibodies, indicating that it was mediated by the viral Env proteins expressed on the surface of the virus-producing cells. Similar fusion was observed between the HTLV-I-producing cells and certain other human nervous system cell lines. If such fusion of HTLV-I-infected T cells occurs also with astrocytes and other nervous system cells in TSP/HAM patients, it may account, at least partially, for the blood-brain barrier breakdown and some of the neural lesions in this syndrome.

Evidence that protein kinase A activity is required for the basal and tax-stimulated transcriptional activity of human T-cell leukemia virus type-I long terminal repeat

The present study was undertaken to investigate the role of protein kinase A (PKA) in the control of human T-cell leukemia virus type-I (HTLV-I) long terminal repeat (LTR) expression, since this issue is still controversial. For this purpose we employed two human T-cell lines; the Jurkat cells in which long exposure to diBu-cAMP severely down-regulated the catalytic subunit of PKA (PKA-C), and H-9 cells in which such exposure markedly increased PKA-C level. Transient transfection assays revealed that addition of diBu-cAMP 1 h before or after transfection profoundly increased HTLV-I LTR directed CAT expression and synergistically enhanced its stimulation by the viral transactivator tax gene product in both cell lines. However longer exposure to diBu-cAMP before transfection reduced LTR-CAT expression to below its basal level and completely abolished its stimulation by tax in Jurkat cells, and this diBu-cAMP inhibitory effect could be abrogated by co-transfection of a PKA-C expressing vector. By contrast, in H-9 cells, this long exposure to diBu-cAMP continued enhancing LTR-CAT expression and its tax-mediated transactivation, and this stimulatory effect of diBu-cAMP could be diminished by the PKA-specific inhibitor N-12-(p-bromocinnamylamine)ethyll-5-isoquinolinsulfonamid e (H-89). Notably, in the absence of diBu-cAMP treatment H-89 reduced LTR-CAT expression to below its basal level and prevented its stimulation by tax in both cell lines. Together these findings indicate not only that cAMP-activated PKA stimulates HTLV-I LTR expression and its transactivation by tax, but even in the absence of PKA activating signals the basal HTLV-I LTR expression as well as its stimulation by tax are both dependent on a basal PKA activity.

Deletion analysis of both the long terminal repeat and the internal promoters of the human foamy virus

Deletion analyses of the long terminal repeat (LTR) and internal promoters (IP) of human foamy virus (HFV) showed that a negative acting element resides in the U5 region of the 5' LTR reducing reporter gene expression tenfold. The basal activity of the IP was higher than that obtained with LTR promoter constructs and strongly elevated in permissive BHK-21 cells whereas semi-permissive COS-7 cells showed low basal activity. Since the basal activity of the IP is critical for initiating HFV gene expression by providing Bel 1 transactivator early after infection, the basal activity of the IP may be the crucial factor that contributes to whether cells are permissive for HFV infection or not. Deletion mutagenesis allowed to define the minimal IP region. A region strongly transactivated by Bel 1 extends from -136 to +58 relative to the cap site of the IP. The major Bel 1 response element of the IP required for transactivation is located upstream of the cap site between -136 and -88 relative to the internal cap site. A DNA fragment reported to be protected by recombinant Bel 1 was deleted with marginal reduction of Bel 1 transactivation. HFV gene expression directed by the IP and LTR promoters is thus multiply regulated by positive and negative acting response elements in cis and their binding partners in trans.

The long terminal repeats of human immunodeficiency virus type-1 and human T-cell leukemia virus type-I are activated by 12-O-tetradecanoylphorbol-13-acetate through different pathways

The LTRs of HIV-1 and HTLV-I have been shown by several laboratories to be activated by 12-O-tetradecanoylphorbol-13-acetate (TPA). This agent is a potent activator of protein kinase C (PKC). However, long exposure to TPA downregulates PKC in many cell types. We demonstrated that TPA treatment of Jurkat cells for more than 24 hr resulted in a sever depletion of this enzyme. Therefore, to explore the role of PKC in the effect of TPA on these LTRs, we transfected Jurkat cells with HIV-1 LTR-CAT or HTLV-I LTR-CAT construct after 72 hr of TPA pretreatment. While this TPA pretreatment considerably reduced the HIV-1 LTR basal expression, it strongly stimulated the expression of HTLV-I LTR. Furthermore, when TPA was added after transfection, a strong stimulation of HIV-1 LTR was observed, which could be abrogated by PKC inhibitors like H7 and chelerythryn. However, under these conditions TPA stimulated HTLV-I LTR to a lesser extent than did the long-term TPA pretreatment. Moreover, this stimulation was enhanced by the PKC inhibitors. Thus our data indicate that while the effect of TPA on HIV-1 LTR is strictly dependent on PKC activity, its effect on HTLV-I LTR is exerted via a different pathway that not only does not require PKC activation but rather seems to be antagonized by the activated PKC. Using a deletion mutant of HTLV-I LTR we mapped the PKC-independent effect of TPA to the c-ets responsive region 1 (ERR-1) located in U3 of this LTR.

A survey of heparin monitoring in Australasia

Full dose heparin therapy is monitored by a variety of laboratory methods, of which the activated partial thromboplastin time (APTT) is the most popular. A large number of APTT reagents are currently available, with different sensitivities to heparin evident in many. Within the literature it is apparent that there is a lack of consensus, and indeed some confusion, regarding the therapeutic ranges for the APTT for standard heparin therapy in the treatment of venous thromboembolic disease. Accordingly we conducted an Australasian survey to evaluate current laboratory and clinical practices in monitoring heparin therapy, to determine the extent of variation in the approach and to stimulate the process of standardisation of acceptable procedures and methodology. Results of the survey demonstrate that currently there is no uniform practice used to establish therapeutic ranges for monitoring standard heparin therapy. Furthermore, results suggest that current practice may lead to subtherapeutic anticoagulation in many laboratories.

Molecular biological characterization of the human foamy virus reverse transcriptase and ribonuclease H domains

Foamy viruses form a separate group of retroviruses encoding a pol protein with at least four domains based on comparative sequence alignments. The polymerase and ribonuclease H domains of the human foamy virus (HFV) pol gene were expressed in Escherichia coli either individually or in combination. The histidine-tagged HFV fusion proteins were subsequently purified to near homogeneity by affinity Ni2+ chelate column chromatography. The polymerase and RNase H activities were characterized by performing conventional DNA polymerase and ribonuclease H assays and in situ gel assays. Six purified recombinant HFV proteins were enzymatically active either individually as DNA polymerase and ribonuclease H or as combined domains. The HFV enzymatic activities were characterized with respect to cation preferences and pH optima. Western blots with antibodies against the RNase H domain, in situ reverse transcriptase (RT), and RNase H gel assays showed that in HFV-infected cells pol proteins of 120 and 80 kDa were detectable. A novel activity band of 60 kDa was found in situ RT gel assays. Recombinant RNase H protein additionally purified by fast performance liquid chromatography was capable of removing the primer for minus-strand DNA synthesis when labeled tRNA(Lys1,2) model substrates were used. Specific cleavages occurred at the phosphodiester bonds one to three nucleotides 5' of the RNA-DNA junction. The results revealed biochemical properties of the HFV pol gene products that define functional domains of the HFV pol gene that are distinct but comparable to other retroviruses.

Expression of c-myc and c-ras oncogenes in the neoplastic and non-neoplastic cells of Hodgkin's disease

The oncogenes c-myc and c-ras are known to elicit a cooperative tumorigenicity. In this study we investigated their role in the pathogenesis of Hodgkin's disease. The expression of these oncogenes was determined in Hodgkin's disease patients by avidin-biotin peroxidase complex immunohistochemical staining and was compared to their expression in patients with non-Hodgkin's lymphomas and inflammatory reactive lymph nodes. Of 29 examined patients with different histological types of Hodgkin's disease, 21 (72.4%) showed an elevated expression of c-myc and 28 (96.5%) of c-ras. Although this expression was marked especially in the neoplastic Reed-Sternberg cells, it was also noted in the numerous reactive cells present in the involved lymph nodes. By contrast, a much lower frequency of increased expression of these oncogenes was recorded in 19 patients with different grades of non-Hodgkin's lymphoma and in 29 patients with inflammatory reactive lymph nodes. The elevated expression of c-myc and c-ras in the neoplastic Reed-Sternberg cells may reflect an oncogenic event that directly activates these genes. However, their increased expression in the surrounding non-neoplastic cells probably results from signal transduction induced by certain growth-promoting factors possibly released by the Reed-Sternberg cells and that act paracrinally to stimulate the proliferation of the neighboring cells. Furthermore, the continuous c-ras elevation may impair the normal cell cycle control and thereby promote mutagenesis and overt malignancy.

Mutational analysis of the reverse transcriptase and ribonuclease H domains of the human foamy virus

Human foamy or spuma virus (HFV) codes for a distinct set of pol gen products. To determine the minimal requirements for the HFV enzymatic activities, defined residues of the reverse transcriptase (RT) and ribo-nuclease H (RNase H) domain of the HFV pol gene were mutated by site-specific PCR mutagenesis. The mutant gene products were bacterially expressed, purified by Ni2+ chelate affinity chromatography and characterised by Western blotting. The enzymatic activities of the individual recombinant HFV pol mutant proteins were characterised by the situ RT, RNase H and RNase H assays. Two substitution mutants reached RT activity levels higher than that of the intact recombinant HFV RT-RH-His. When the catalytically essential D508 was substituted by A508, 5% of RNase H activity was retained while DNA polymerase activity increased 2-fold. A deletion of 11 amino acid residues in the hinge region completely abolished DNA polymerase while RNase H activity decreased 2-fold. A deletion mutant in the C-terminal RH domain showed no RNase H but retained RNase H activity indicating that the activities are genetically separable. The combined data reveal that the HFV DNA polymerase and RNase H activities are interdependent.

Analysis of the phylogenetic placement of different spumaretroviral genes reveals complex pattern of foamy virus evolution

Foamy or spumaviruses are complex retroviruses. Phylogenetic trees have been constructed previously for either the polymerase or integrase domains showed a clustering of the foamy viruses relatively distant from other retroviruses. The most related retrovirus was found to be murine leukemia virus, irrespective of the method used or foamy viral gene analyzed. We analyze bel genes of different foamy viruses and compared the corresponding phylogenetic trees with those obtained from the pol genes that were constructed with refined computer programs. In addition, the nucleocapsid protein sequence of foamy viruses was used for a comparative phylogenetic analysis. Known biological properties of the individual FV protein domains are discussed to ascertain the apparent phylogenetic relatedness.

DNA binding properties of the zinc-bound and zinc-free HIV nucleocapsid protein: supercoiled DNA unwinding and DNA-protein cleavable complex formation

The HIV nucleocapsid (NC) protein contains, as those of other retroviruses, two Cys-His arrays which function as zinc finger binding domains. The nucleic acid binding properties of retroviral NC have been previously demonstrated. In this study, we characterized the DNA binding ability of the zinc-bound and zinc-free forms of HIV NC. We found that in addition to binding single-stranded DNA, both forms bind and unwind supercoiled plasmid DNA. The binding ability of the zinc-bound form was higher than the zinc-free form. In addition we showed the formation of NC protein-DNA cleavable complex which is the result of a presumably covalent bond formed between the protein and the phosphate moiety of the DNA backbone. The NC unwinding activity and the protein-DNA cleavable complex formation resembles the first step of the relaxing mechanism of DNA topoisomerase. Our results shed light on the possibility of a novel physiological function for the HIV NC protein in the viral life cycle.

Stimulation of virus production and induction of self-syncytium formation in human T-cell leukemia virus type I- and type II-infected T cells by 12-O-tetradecanoylphorbol-13-acetate

Treatment of human T-cell leukemia virus type I (HTLV-I)- and HTLV-II-infected T-cell lines with 12-O-tetradecanoylphorbol-13-acetate (TPA) stimulated virus release. However, this stimulation was mainly detected at 42 to 48 h of treatment, whereas later virus release declined rapidly. During the first 48 h, TPA had no effect on cell growth, but later, the number of viable cells was profoundly lower in the TPA-treated than in the untreated cultures. This shift in virus release and cell number resulted from self-fusion of a large proportion of the virus-producing cells, which seemed to consequently enter into a dying process. This fusion, which resulted in syncytium formation, was strongly inhibited by anti-HTLV-I env monoclonal antibodies. Furthermore, no self-fusion was detected in three different uninfected T-cell lines similarly treated with TPA. On the other hand, stimulation of virus production by 3-methylcholanthrene (3-MC) treatment failed to induce self-fusion in the infected cells. Moreover, no syncytium was detected when these 3-MC-treated infected cells were cocultured with any of the TPA-treated uninfected cells. The effects of TPA on virus production and syncytium formation were both abolished by three different protein kinase C inhibitors. Taken together, these data suggest that the self-fusion observed in these experiments required both enhanced virus production and protein kinase C-phosphorylated viral or/and virally induced cellular component(s).

The human foamy virus internal promoter directs the expression of the functional Bel 1 transactivator and Bet protein early after infection

The human foamy virus or spumaretrovirus (HFV) is a complex retrovirus that has the capacity to code not only for the three retroviral genes gag, pol, and env but, in addition, for at least three bel genes. The HFV provirus contains two different and functionally active promoters: the classical retroviral promoter in the 5' long terminal repeat and a recently identified second promoter in the env gene upstream of the bel genes. Both promoter/enhancers are strongly dependent on the HFV transcriptional transactivator protein Bel 1. Here we report that the internal promoter directs the synthesis of viral transcripts that code for functionally active Bel 1 and for Bet proteins that appeared early after HFV infection. The viral mRNAs of the internal promoter have a 112-nucleotide-long leader exon and were spliced predominantly at the first splice donor site in the 5' untranslated region. The data were obtained by transient expression assays, transactivation experiments, and RNA analyses of transcripts derived from HFV-infected cells. The results provide strong evidence for the crucial role the internal promoter plays during HFV infection in generating bel-specific transcripts.

Increase in the basal transcriptional activity of the human foamy virus internal promoter by the homologous long terminal repeat promoter in cis

The human foamy or spumaretrovirus HFV is a complex and exogenous retrovirus that encodes several bel genes besides the three classical retroviral genes gag, pol, and env. HFV was recently reported to contain two functionally active promoters that are both strongly trans-activated by the HFV trans-activator protein Bel 1. The occurrence of a second internal cap site underscores the complexity of the HFV genome. We have analysed whether there is interference between the HFV long terminal repeat promoter and the internal promoter located in the 3' end of env upstream of the bel genes. Recombinant clones were constructed that carry two different indicator genes, one under the control of the U3 promoter, the other under the control of the internal promoter. The portion of the basal transcriptional activity of the internal promoter that is not trans-activated by Bel 1 was increased two- to threefold in the presence of the long terminal repeat promoter. The rate of trans-activation by Bel 1 of both HFV promoters was not altered in these constructs.

Synergistic effect of interferon-gamma and phorbol myristate acetate on superoxide production by human monocytes

This study demonstrates a synergistic effect of IFN gamma and PMA on superoxide generation by human monocytes. A strict correlation was observed between the induction of superoxide production and PK-C activation by PMA alone. No such correlation was evident for IFN gamma. However, exposure of the cells to IFN gamma for 10 to 15 hr prior to PMA treatment enhanced both superoxide production and PK-C activation. Using protein kinase inhibitors, we noticed that while PMA exerted its effect by activating PK-C, IFN gamma operated via activation of calcium/calmodulin-dependent or some other calcium-dependent protein kinases. These kinases appeared to be involved in the effect of IFN gamma on superoxide production, as well as in its potentiation of PMA activity.

Inhibition of retrovirus-induced disease in mice by camptothecin

We have previously shown that noncytotoxic doses of camptothecin (CPT), a topoisomerase I-specific antagonist, inhibit retrovirus replication in acutely and chronically infected cells. To evaluate the efficacy of CPT as an antiretroviral drug in vivo, we injected newborn BALB/c mice with Moloney murine leukemia virus and adult NFS mice with Friend spleen focus-forming virus. The Moloney murine leukemia virus-injected mice developed lymphoma, and the Friend spleen focus-forming virus-injected mice developed erythroleukemia. CPT, administrated together with the virus or 1 or 2 days after virus injection, prevented the onset of the disease in both cases. We showed that repeated CPT treatments increased the effectiveness of the drug when administrated 3 days after virus injection. This ability of CPT to inhibit retrovirus-induced disease in vivo without causing any apparent toxic side effects suggests its application as a legitimate remedy for the treatment of retroviral diseases.

Stimulation of HTLV-I expression by subtoxic dose of 3-methylcholanthrene

In vivo infection by HTLV-I is characterized by the lack of virus expression in the host's infected cells and in most cases it is clinically inapparent. The present study shows that exposure of HTLV-I-infected human T-cells to a maximal non-toxic dose of 3-MC remarkably enhances virus expression. This stimulatory effect was demonstrated by showing elevated levels of viral RNA and proteins within the cells and increased virus release to the culture medium. The enhanced virus expression was found to increase the ability of the treated cells to infect normal human peripheral lymphocytes in co-culture. Such data point to the possibility that initiation of a pathogenic process in HTLV-I-infected individuals might result from activating the expression of the latent provirus by environmental carcinogens.

Role of natural killer cells in controlling local tumor formation and metastatic manifestation of different 3LL Lewis lung carcinoma cell clones

Using two cell clones derived from 3LL Lewis carcinoma; an NK-resistant (designated A6) and an NK-susceptible (designated F2) clone, we investigated the effect of the NK system on local tumor growth and metastasis. Both clones generated local tumors at the inoculation site somewhat faster in NK-deficient (bg/bg) than in normal mice. In addition, these tumors grew slightly faster in the bg/bg mice. However, the effect of the NK system on the metastatic spread of these clones was much more prominent. The F2 cells were profoundly less metastatic than the A6 cells when inoculated into normal mice, whereas in the NK-deficient mice, both clones were highly metastatic. Thus, the NK system appears to be effective primarily in controlling the metastatic spread of the 3LL cells and has a lesser effect on their local tumor formation.

Effect of temperature on the expression of major histocompatibility complex class-I antigens

In the present study we investigated the effect of temperature on MHC class-I gene expression in BALB/C 3T3 cells incubated for 5 days at 34 degrees C, 37 degrees C and 39 degrees C. FACS analysis revealed no significant difference in the cell surface expression of any of the 3 major class-I antigens at 34 degrees C and 37 degrees C. Strikingly, however, when the level of the respective mRNA was determined, only that of the H-2K was comparable at both temperatures, whereas the levels of the H-2D and H-2L mRNA were profoundly higher at 37 degrees C. These data appear to reflect a differential temperature-related transcriptional control of the different class-I genes or a different temperature effect on the stability of their mRNA. The absence of a parallel increase in surface expression of the corresponding H-2D and H-2L antigens may result from some translational or post-translational limiting factors. At 39 degrees C, however, these limiting factors seem to be overcome since the surface expression of all the 3 antigens was remarkably increased although the level of their encoding mRNA was rather lower than in 37 degrees C. This stimulatory effect might be ascribed to heat shock proteins which are known to arise in cells at heat or other stress conditions. They participate in assembly and disassembly of various protein complexes and in transport of certain proteins across intracellular membranes. Such proteins may have arisen in our cells at 39 degrees C and facilitated the intracellular assembly of the class-I molecules and their transport to the cell surface. The possible implication of such heat shock proteins in the anti-tumor effect of hyperthermia is discussed.

Human T-cell leukemia virus type II Rex binding and activity require an intact splice donor site and a specific RNA secondary structure

The human T-cell leukemia virus type II (HTLV-II) regulatory protein Rex augments cytoplasmic levels of unspliced gag-pol mRNA by acting through a Rex-responsive element (RxRE) in the long terminal repeat. Purified Rex protein binds to long terminal repeat mRNA. Here, using an immunobinding assay to measure the binding of Rex protein to mutated RxRE RNAs, we show that efficient Rex binding requires a stem-bulge-loop RNA secondary structure (nucleotides [nt] 465 to 500) and specific sequences both within the stem-bulge (nt 470 to 476) and within a conserved upstream splice donor site (nt 449 to 455). Rex function in a transient transfection expression system correlates with Rex protein-RxRE RNA binding. The ability of HTLV-II Rex to interact directly with the HTLV-II splice donor site suggests that HTLV-II Rex may increase expression of unspliced gag-pol mRNA, in part, by inhibiting splicing.