An advanced application of protein microarrays: cell-based assays for functional genomics.

Microarrays have become common tools for approaching different experimental questions: DNA, protein and peptide arrays offer the power of multiplexing the assay and by means of miniaturization technology, the possibility to reduce cost and amount of samples and reagents. Recently, a novel technology for functional assays has been proposed. Sabatini and co-workers have shown a cell-based microarrays method (1) that relies on the deposition and immobilization of an array of cDNA plasmids on a slide where cells are subsequently plated; the cDNA is then internalized by "reverse transfection" and cells overexpress or downregulate in each single spot the genes of interest. This approach allows the screening of different phenotypes in living cells of many genes in parallel on a single slide. To overcome some relevant limitations of this approach, we have implemented the technology by means of viral immobilization (2) on a novel surface of cluster-assembled nanostructured TiO2 (3) previously functionalized with an array of a docking protein. In this work, we present the detailed development of the "reverse infection cell-microarray based technology" in U2OS cells on a novel coated slide that represents an advanced application of protein arrays.

Quantitative analysis of the interactions between HIV-1 integrase and retroviral reverse transcriptases.

The RT (reverse transcriptase) of HIV-1 interacts with HIV-1 IN (integrase) and inhibits its enzymatic activities. However, the molecular mechanisms underling these interactions are not well understood. In order to study these mechanisms, we have analysed the interactions of HIV-1 IN with HIV-1 RT and with two other related RTs: those of HIV-2 and MLV (murine-leukaemia virus). All three RTs inhibited HIV-1 IN, albeit to a different extent, suggesting a common site of binding that could be slightly modified for each one of the studied RTs. Using surface plasmon resonance technology, which monitors direct protein-protein interactions, we performed kinetic analyses of the binding of HIV-1 IN to these three RTs and observed interesting binding patterns. The interaction of HIV-1 RT with HIV-1 IN was unique and followed a two-state reaction model. According to this model, the initial IN-RT complex formation was followed by a conformational change in the complex that led to an elevation of the total affinity between these two proteins. In contrast, HIV-2 and MLV RTs interacted with IN in a simple bi-molecular manner, without any apparent secondary conformational changes. Interestingly, HIV-1 and HIV-2 RTs were the most efficient inhibitors of HIV-1 IN activity, whereas HIV-1 and MLV RTs showed the highest affinity towards HIV-1 IN. These modes of direct protein interactions, along with the apparent rate constants calculated and the correlations of the interaction kinetics with the capacity of the RTs to inhibit IN activities, are all discussed.

The location in cartilage of infectious retrovirus in cats infected with feline leukemia virus.

BACKGROUND: Previous studies have suggested that articular cartilage allografts were not likely to transmit infectious retrovirus since viral DNA could not be isolated from chondrocytes of infected individuals. However, the ability of the extracellular matrix of articular cartilage to harbor and transmit a retrovirus has not been examined. We hypothesized that articular cartilage fragments, but not isolated chondrocytes, from cats systemically infected with feline leukemia virus (FeLV) are capable of transmitting infectious retrovirus. METHODS: Fresh cartilage segments and chondrocytes isolated from cats systemically infected with feline leukemia virus were used in this study. Feline embryonic fibroblast cells were cocultured with segments of cartilage, isolated chondrocytes, or fragments of cortical bone from each infected cat. The FeLV p27 antigen was measured in the coculture media by enzyme-linked immunosorbent assay. In addition, FeLV proviral nucleic acids were quantified by real-time quantitative polymerase chain reaction with use of DNA extracted from feline embryonic fibroblast cell cocultures as well as isolated chondrocytes. Immunohistochemistry was used to assess for FeLV p27 antigen in both intact cartilage fragments and isolated chondrocytes. RESULTS: Feline embryonic fibroblast cells cocultured with cartilage fragments from each of the five FeLV-infected cats all demonstrated high levels of proviral DNA, indicating transmission of infective virus. In addition, media from all cocultures of feline embryonic fibroblast cells and chondral fragments became positive for p27 antigen, indicating active viral replication. In contrast, cocultures of feline embryonic fibroblast cells and isolated chondrocytes from all FeLV-infected cats were negative for proviral DNA and p27 antigen. Likewise, no proviral nucleic acids could be detected in isolated chondrocytes from any infected cats. Cocultures of feline embryonic fibroblast cells with cortical bone fragments were positive for proviral DNA and p27 antigen. Immunohistochemical staining of cartilage fragments from FeLV-infected cats demonstrated the presence of p27 antigen throughout the extracellular matrix, but the p27 antigen was not detected in isolated chondrocytes. CONCLUSIONS: Articular cartilage fragments can readily transmit infectious retrovirus, but isolated chondrocytes were likely not the source of the infectious virus because they did not harbor proviral DNA or p27 antigen.

A human T-cell lymphotropic virus type 1 enhancer of Myc transforming potential stabilizes Myc-TIP60 transcriptional interactions.

The human T-cell lymphotropic virus type 1 (HTLV-1) infects and transforms CD4+ lymphocytes and causes adult T-cell leukemia/lymphoma (ATLL), an aggressive lymphoproliferative disease that is often fatal. Here, we demonstrate that the HTLV-1 pX splice-variant p30II markedly enhances the transforming potential of Myc and transcriptionally activates the human cyclin D2 promoter, dependent upon its conserved Myc-responsive E-box enhancer elements, which are associated with increased S-phase entry and multinucleation. Enhancement of c-Myc transforming activity by HTLV-1 p30II is dependent upon the transcriptional coactivators, transforming transcriptional activator protein/p434 and TIP60, and it requires TIP60 histone acetyltransferase (HAT) activity and correlates with the stabilization of HTLV-1 p30II/Myc-TIP60 chromatin-remodeling complexes. The p30II oncoprotein colocalizes and coimmunoprecipitates with Myc-TIP60 complexes in cultured HTLV-1-infected ATLL patient lymphocytes. Amino acid residues 99 to 154 within HTLV-1 p30II interact with the TIP60 HAT, and p30II transcriptionally activates numerous cellular genes in a TIP60-dependent or TIP60-independent manner, as determined by microarray gene expression analyses. Importantly, these results suggest that p30II functions as a novel retroviral modulator of Myc-TIP60-transforming interactions that may contribute to adult T-cell leukemogenesis.

Demonstration of feline foamy virus in experimentally infected cats by immunohistochemistry.

Foamy viruses (FV) are complex retroviruses which are commonly isolated from cats, cattle and non-human primates. The infection is persistent and infected animals have a sustained antibody response. The role of FV in diseases remains unclear, in cats, a possible association with uncharacterized renal symptoms remains to be confirmed. To demonstrate feline FV (FFV) in tissues of experimentally infected cats three polyclonal monospecific antisera from rabbits against three different viral proteins, the structural Gag and the non-structural Bel 1 and Bet proteins were tested for their applicability in immunohistochemistry with paraffin sections. Only the Bet antiserum allowed detection of FFV-specific proteins, the antibodies against Gag and Bel 1 did not work even after pre-treatment of the slides with proteinase K or cooking in a pressure cooking pot. The Bet-reactive antibodies were detected using a commercial streptavidin kit and revealed Bet in the cytoplasm of cells from different lymphoid tissues like lymphnodes, tonsils, thymus and spleen. The method described opens new ways to explore the in vivo replication and tissue specificity of FFV and its possible role in disease.

High prevalence of an IgG response against murine leukemia virus (MLV) in patients with psoriasis.

Increasing evidence suggests that human endogenous retroviruses (HERV) could participate in the pathogenesis of autoimmune diseases such as multiple sclerosis and lupus erythematosus. To assess a possible association of murine leukemia virus (MLV)-like group of HERVs with psoriasis we searched for antibodies against MLV proteins in the sera of patients. We showed that anti-MLV antibodies (total) were detected in both psoriatic and control sera. However, they were detected with a higher frequency in psoriasis when compared with controls (91 vs. 53%, respectively, P=0.001). In addition, the IgG response was dramatically increased in psoriasis (86 vs. 8%, respectively, P<0.0001). This immunoreactivity was observed against the products of both the gag and env genes, and the most antigenic proteins were the gp65-70. Moreover, we observed that anti-p30 MLV antibodies reacted with an epidermal protein with a molecular weight of 50 kDa in protein extracts from both normal and psoriatic skin cultures. These observations suggest that HERVs of the MLV-like group could contribute to the immunopathogenesis of psoriasis.

Adaptation of a sandwich enzyme-linked immunosorbent assay to determine the concentration of bovine leukemia virus p24 and optimal conditions for p24 expression in short-term cultures of peripheral blood mononuclear cells.

Bovine leukemia virus (BLV) is a common retroviral infection of cattle. Infection is accompanied by integration of BLV into the host cell genome and is persistent for the life of the individual as is the presence of anti-BLV antibodies. Lymphosarcoma occurs in a small fraction of infected adult individuals but otherwise there is little or no associated disease. Viremia is undetectable, however, BLV is expressed readily once infected cells are cultured in vitro. A sandwich enzyme-linked immunosorbent assay (sELISA) was optimized, using murine monoclonal antibodies, to quantify the major internal structural protein (p24) produced in short-term cultures of peripheral blood mononuclear cells (PBMCs). Optimal production of BLV p24 was achieved utilizing RPMI supplemented with 10% fetal bovine serum (FBS), pH 7, and 5 x 10(6) cells per ml. Cultures were terminated at 24 h. The sELISA was linear between 30 and 900 ng/ml and the limit of detection was 1.2 ng/ml. At three concentrations of p24, intra- and inter-assay coefficients of variation (CV) varied between 9.2 and 13.3 and 5.1 and 12.9%, respectively.

The promyelocytic leukemia protein does not mediate foamy virus latency in vitro.

Spumaviruses, commonly called foamy viruses, are complex retroviruses that establish life-long persistent infections in the absence of accompanying pathology. Depending upon cell type, infection of cells in tissue culture cells can result in either lytic replication, persistence, or latency. The cellular factors that mediate foamy virus (FV) latency are poorly understood. In this study we show that the only known inhibitor of FV replication, the promyelocytic leukemia protein (PML), which binds the FV transactivator (Tas), does not play an important role in FV latency in vitro. We found no significant differences in PML levels in cells that supported lytic replication compared to those that were latently infected. Furthermore, endogenous PML levels did not change following exposure to phorbol myristate acetate (PMA), which induces FV replication. We demonstrated that FV replication proceeded in the presence of substantial levels of PML, both in fully permissive cells and during reactivation of latent FV. Endogenous PML did not efficiently colocalize with Tas, even after upregulation by alpha interferon (IFN-alpha) treatment. IFN-alpha did, however, partially suppress the reactivation of latent FV by PMA. Finally, depletion of endogenous PML by small interfering RNA did not promote activation of FV in cells that responded to PMA treatment. Taken together, these data indicate that endogenous PML does not play an important role in mediating FV latency.

Construction and functional characterization of feline foamy virus-based retroviral vectors.

Replication-competent feline foamy or spuma virus (FFV) vectors were constructed and functionally tested. The unmodified FFV vector genome expressed by the strong human cytomegalovirus immediate early promoter encodes FFV particles that were replication-competent in cell cultures. Virus derived from the cloned FFV DNA replicated and persisted in experimentally infected cats similar to the FFV isolate FUV. A FFV vector partially deleted in the noncoding area of the U3 region was used to transduce the gene for the green fluorescent protein (Gfp) into cell cultures. Gfp was expressed either by an internal ribosomal entry site (IRES) or as C-terminal fusion protein linked to Bet that was recently shown to be essential for FFV replication. Whereas the genetic stability of the IRES-Gfp construct was comparably low, the Bet-Gfp fusion protein was detectable upon serial cell-free vector passages. However, genetic rearrangements also occurred leading to the concomitant loss of marker gene expression.

Lentiviral vectors containing the human immunodeficiency virus type-1 central polypurine tract can efficiently transduce nondividing hepatocytes and antigen-presenting cells in vivo.

High-titer self-inactivating human immunodeficiency virus type-1 (HIV-1)-based vectors expressing the green fluorescent protein reporter gene that contained the central polypurine and termination tract and the woodchuck hepatitis virus posttranscriptional regulatory element were constructed. Transduction efficiency and biodistribution were determined, following systemic administration of these improved lentiviral vectors. In adult severe combined immunodeficiency (SCID) mice, efficient stable gene transfer was achieved in the liver (8.0% +/- 6.0%) and spleen (24% +/- 3%). Most transduced hepatocytes and nonhepatocytes were nondividing, thereby obviating the need to induce liver cell proliferation. In vivo gene transfer with this improved lentiviral vector was relatively safe since liver enzyme concentration in the plasma was only moderately and transiently elevated. In addition, nondividing major histocompatibility complex class II-positive splenic antigen-presenting cells (APCs) were efficiently transduced in SCID and normal mice. Furthermore, B cells were efficiently transduced, whereas T cells were refractory to lentiviral transduction in vivo. However, in neonatal recipients, lentiviral transduction was more widespread and included not only hepatocytes and splenic APCs but also cardiomyocytes. The present study suggests potential uses of improved lentiviral vectors for gene therapy of genetic blood disorders resulting from serum protein deficiencies, such as hemophilia, and hepatic disease. However, the use of liver-specific promoters may be warranted to circumvent inadvertent transgene expression in APCs. In addition, these improved lentiviral vectors could potentially be useful for genetic vaccination and treatment of perinatal cardiac disorders.

Mitochondrial alterations induced by the p13II protein of human T-cell leukemia virus type 1. Critical role of arginine residues.

Human T-cell leukemia virus type 1 encodes a number of "accessory" proteins of unclear function; one of these proteins, p13(II), is targeted to mitochondria and disrupts mitochondrial morphology. The present study was undertaken to unravel the function of p13(II) through (i) determination of its submitochondrial localization and sequences required to alter mitochondrial morphology and (ii) an assessment of the biophysical and biological properties of synthetic peptides spanning residues 9-41 (p13(9-41)), which include the amphipathic mitochondrial-targeting sequence of the protein. p13(9-41) folded into an alpha helix in micellar environments. Fractionation and immunogold labeling indicated that full-length p13(II) accumulates in the inner mitochondrial membrane. p13(9-41) induced energy-dependent swelling of isolated mitochondria by increasing inner membrane permeability to small cations (Na(+), K(+)) and released Ca(2+) from Ca(2+)-preloaded mitochondria. These effects as well as the ability of full-length p13(II) to alter mitochondrial morphology in cells required the presence of four arginines, forming the charged face of the targeting signal. The mitochondrial effects of p13(9-41) were insensitive to cyclosporin A, suggesting that full-length p13(II) might alter mitochondrial permeability through a permeability transition pore-independent mechanism, thus distinguishing it from the mitochondrial proteins Vpr and X of human immunodeficiency virus type 1 and hepatitis B virus, respectively.

Biochemical analysis of retroviral structural proteins to identify and quantify retrovirus expressed by an NS0 murine myeloma cell line.

A subclone of the NS0 murine myeloma cell line, frequently used to produce recombinant monoclonal antibodies, was found by a transmission electron microscopy method to express a surprisingly high titer of 10(11) retroviral particles per ml of culture supernatant. Infectivity assays showed a very low infectious titer with the restricted host range expected for a murine amphotropic retrovirus. A Western blot assay for the viral capsid protein was developed to confirm the high titer values and provide a means for monitoring batch consistency and virus removal during the purification process. Mass spectrometry of several of the viral Gag proteins demonstrated that the cell line appeared to produce at least two closely related retroviruses. N-terminal sequencing of three of the Gag proteins demonstrated that these retroviruses were members of the murine leukemia retroviral family. Western blot detection with an antibody for the capsid protein gave a linear standard curve over the range of 0.1-3 ng per lane. This allows the detection of viral titers as low as 6x10(7) virions per ml without the need to concentrate the sample. The Western blot method has higher throughput and less variability than transmission electron microscopy methods and has potential for monitoring viral titer and clearance during development of manufacturing processes.

Visualisation of phenotypically mixed HIV-1 and HTLV-I virus particles by electron microscopy.

Virions produced after HIV-1 infection of HTLV-I transformed cells have an expanded tropism that has been attributed to the presence of HTLV-I glycoproteins in the envelope. This report now directly identifies these phenotypically mixed virions by immunogold labelling electron microscopy. Furthermore we estimate there are 2% of these in cell-free supernatant, which represents up to 1 x 10(7) particles/ml from an in vitro infection. HTLV-1 envelope labelling was localised to a single region, suggesting a defined event in packaging of foreign envelope proteins into HIV-1 virus particles.

3D NMR experiments for measuring 15N relaxation data of large proteins: application to the 44 kDa ectodomain of SIV gp41.

A suite of 3D NMR experiments for measuring 15N-¿1H¿ NOE, 15N T1, and 15N T1rho values in large proteins, uniformly labeled with 15N and 13C, is presented. These experiments are designed for proteins that exhibit extensive spectral overlap in the 2D 1H-15N HSQC spectrum. The pulse sequences are readily applicable to perdeuterated samples, which increases the spectral resolution and signal-to-noise ratio, thereby permitting the characterization of protein dynamics to be extended to larger protein systems. Application of the pulse sequences is demonstrated on a perdeuterated 13C/15N-labeled sample of the 44 kDa ectodomain of SIV gp41.

Characterization of a putative retroviral env-related human protein.

In this study, we developed a rabbit polyclonal antibody to a fusion protein containing the envelope (env) transmembrane (TM) region from the human endogenous retroviral family, HERV-E. We used this reagent to document the expression of TM-related protein by Western blot in endothelial, colon and prostate carcinoma and seminoma cell lines and in peripheral blood mononuclear cells from a healthy donor. We detected a 58-kD protein (as compared to murine TM of 15 kD) that had specificity for the env-related antibodies of the polyclonal antiserum.

Expression of human endogenous retrovirus K elements in germ cell and trophoblastic tumors.

Antibodies against proteins encoded by human endogenous retrovirus (HERV)-K family genes are consistently found in the sera of patients with classical seminoma. Furthermore, HERV-K Gag-encoded protein could be detected in corresponding tumor biopsies. Addressing questions as to the extent of HERV gene expression in biologically related lesions, we studied various testicular and ovarian germ cell tumors (GCTs), GCT precursor lesions, and gestational trophoblastic disease (GTD) for the presence of HERV-K gene transcripts in tissue sections. By in situ hybridization using four non-overlapping, isotopically labeled RNA probes specific for HERV-K gag and env sequences on archival tissue samples, consistent HERV-K expression of gag and env genes was found to be common to all GCTs and their testicular precursor lesions with the exception of teratomas, mature and immature, and spermatocytic seminomas. HERV-K expression was also found in malignant GTD (choriocarcinoma) but not in benign GTD (noninvasive molar pregnancy). There was no evidence for HERV-K expression in differentiated embryonal and adult tissues as well as a total of 53 tumors other than GCT or GTD. The findings point to a common molecular pathoetiology of GCT and malignant GTD, have implications for the classification of GCTs, and support the concept of carcinoma in situ as a precursor lesion common to all forms of testicular GCT.

Effects of vif mutations on cell-free infectivity and replication of simian immunodeficiency virus.

To investigate the function of the Vif protein of the simian immunodeficiency virus (SIV), mutations were introduced into the SIVmac239 vif gene without affecting the reading frames of other overlapping genes. The phenotypes of these mutant viruses were examined with respect to viral replication and the expression and processing of viral proteins. Transfection of vif-mutant proviral DNA into established T cell lines resulted in a significant delay in the onset of virus replication compared to that seen with the wild-type provirus. The efficiency of replication of the vif-mutant virus was dependent on cell type. MT-4 cells were permissive for replication of the vif mutant, while replication in CEMx174 cells was severely restricted. Little or no virus replication was observed following cell-free infection of the CEMx174 cell line and macaque peripheral blood mononuclear cells (PBMC). These results indicate that the requirement for vif during the replication of SIVmac239 is dependent on cell type, as has been observed for HIV-1. Following cell-free infection, mutant viruses containing combined deletions in vif and the other regulatory genes (vpx, vpr, and nef) displayed replication kinetics similar to that of viruses containing the deletion of vif alone. Viral protein expression and processing in MT-4 cells of vif-deleted viruses were indistinguishable from those of the wild-type virus. The effects of two different point mutations in vif were examined. One point mutant in vif reverted to the genetic sequence of the wild-type virus within 2 weeks.2 +

Reactivity of primate sera to foamy virus Gag and Bet proteins.

In order to establish criteria for the serodiagnosis of foamy virus infections we investigated the extent to which sera from infected individuals of human and primate origin react with structural and non-structural virus proteins in immunoblot assays. Using lysates from infected cells as the source of virus antigen, antibodies were preferentially detected against the Gag proteins and the non-structural Bet protein. Both the Gag precursor molecules of 70 and 74K apparent M(r) and the cytoplasmic 60K M(r) Bet protein were found to be phosphorylated, the latter being synthesized in large amounts in infected cells. Rabbit antiserum raised against recombinant human foamy virus (HFV) Gag major capsid protein cross-reacted with foamy viruses of chimpanzee, gorilla, orang-utan, rhesus monkey and African green monkey origin. This was reflected by a broad cross-reactivity of the respective monkey sera to the Gag proteins of the various foamy virus isolates. Cross-reactivity of antisera against the Bet protein was restricted to viruses from man and the great apes. Recombinant Gag and Bet proteins expressed in prokaryotes or in insect cells were readily recognized by foamy virus-positive primate sera. Screening serum samples from chimpanzees with HFV Gag and Bet proteins expressed by recombinant baculoviruses revealed that 18 out of 35 (52%) were positive for Gag antibodies. Of these, 13 (72%) showed antibodies against the Bet protein, indicating that Bet antigen is of value in serological screening for foamy virus infections.

Assay methods for retroviral proteases.

A variety of assay methods for retroviral proteases have been developed in response to different experimental requirements, such as initial identification of a protease, subsequent enzymatic characterization, and high-capacity automated screening of potential inhibitors. This chapter has reviewed a number of these methods above; most have been closely tailored to match specific experimental requirements, and some of them are described in greater detail in other chapters in this volume. They include analysis of polyprotein cleavage using SDS-PAGE, analysis of the determinants of substrate cleavage using either chromogenic peptides or reversed-phase HPLC for product separation after cleavage of unmodified peptides, and the design and utilization of quenched fluoregenic peptides for use in continuous assay.