Haruka Kamiyama

Ezrin, Radixin, and Moesin (ERM) proteins function as pleiotropic regulators of human immunodeficiency virus type 1 infection.

Ezrin, radixin, and moesin (ERM) proteins supply functional linkage between integral membrane proteins and cytoskeleton in mammalian cells to regulate membrane protein dynamisms and cytoskeleton rearrangement. To assess potential role of the ERM proteins in HIV-1 lifecycle, we examined if suppression of ERM function in human cells expressing HIV-1 infection receptors influences HIV-1 envelope (Env)-mediated HIV-1-vector transduction and cell-cell fusion. Expression of an ezrin dominant negative mutant or knockdown of ezrin, radixin, or moesin with siRNA uniformly decreased transduction titers of HIV-1 vectors having X4-tropic Env. In contrast, transduction titers of R5-tropic Env HIV-1 vectors were decreased only by radixin knockdown: ezrin knockdown had no detectable effects and moesin knockdown rather increased transduction titer. Each of the ERM suppressions had no detectable effects on cell surface expression of CD4, CCR5, and CXCR4 or VSV-Env-mediated HIV-1 vector transductions. Finally, the individual knockdown of ERM mRNAs uniformly decreased efficiency of cell-cell fusion mediated by X4- or R5-tropic Env and HIV-1 infection receptors. These results suggest that (i) the ERM proteins function as positive regulators of infection by X4-tropic HIV-1, (ii) moesin additionally functions as a negative regulator of R5-tropic HIV-1 virus infection at the early step(s) after the membrane fusion, and (iii) receptor protein dynamisms are regulated differently in R5- and X4-tropic HIV-1 infections.

Raft localization of CXCR4 is primarily required for X4-tropic human immunodeficiency virus type 1 infection.

Human immunodeficiency virus type 1 (HIV-1) infection is initiated by successive interactions of viral envelope glycoprotein gp120 with two cellular surface proteins, CD4 and chemokine receptor. The two most common chemokine receptors that allow HIV-1 entry are the CCR5 and CXCR4. The CD4 and CCR5 are mainly localized to the particular plasma membrane microdomains, termed raft, which is rich in glycolipids and cholesterol. However, the CXCR4 is localized only partially to the raft region. Although the raft domain is suggested to participate in HIV-1 infection, its role in entry of CXCR4-tropic (X4-tropic) virus is still unclear. Here, we used a combination of CD4-independent infection system and cholesterol-depletion-inducing reagent, methyl-beta-cyclodextrin (MbetaCD), to address the requirement of raft domain in the X4-tropic virus infection. Treatment of CD4-negative, CXCR4-positive human cells with MbetaCD inhibited CD4-independent infection of the X4-tropic strains. This inhibitory effect of the cholesterol depletion was observed even when the CXCR4 was over-expressed on the target cells. Soluble CD4-induced infection was also inhibited by MbetaCD. The MbetaCD had no effect on the levels of cell surface expression of CXCR4. In contrast to these infections, MbetaCD treatment did not inhibit CD4-dependent HIV-1 infection in the wild type CD4-expressing cells. This study and previous reports showing that CD4 mutants localized to non-raft domains function as HIV-1 receptor indicate that CXCR4 clustering in the raft microdomains, rather than CD4, is the key step for the HIV-1 entry.

Synthesis, structure-activity relationships, and mechanism of action of anti-HIV-1 lamellarin α 20-sulfate analogues.

Lamellarin α and six different types of lamellarin α 20-sulfate analogues were synthesized and their structure-activity relationships were investigated using a single round HIV-1 vector infection assay. All lamellarin sulfates having pentacyclic lamellarin core exhibited anti-HIV-1 activity at a 10 μM concentration range regardless of the number and position of the sulfate group. On the other hand, non-sulfated lamellarin α and ring-opened lamellarin sulfate analogues did not affect HIV-1 vector infection in similar concentrations. The lamellarin sulfates utilized in this study did not exhibit unfavorable cytotoxic effect under the concentrations tested (IC(50)>100 μM). Confocal laser scanning microscopic analysis indicated that hydrophilic lamellarin sulfates were hardly incorporated in the cell. HIV-1 Env-mediated cell-cell fusion was suppressed by lamellarin sulfates. These results suggested that lamellarin sulfates have a novel anti-HIV-1 activity besides the previously reported integrase activity inhibition, possibly at a viral entry step of HIV-1 replication.

CD4-independent human immunodeficiency virus infection involves participation of endocytosis and cathepsin B.

During a comparison of the infectivity of mNDK, a CD4-independent human immunodeficiency virus type 1 (HIV-1) strain, to various cell lines, we found that HeLa cells were much less susceptible than 293T and TE671 cells. Hybridoma cells between HeLa and 293T cells were as susceptible as 293T cells, suggesting that cellular factors enhance the mNDK infection in 293T cells. By screening a cDNA expression library in HeLa cells, cystatin C was isolated as an enhancer of the mNDK infection. Because cathepsin B protease, a natural ligand of cystatin C, was upregulated in HeLa cells, we speculated that the high levels of cathepsin B activities were inhibitory to the CD4-independent infection and that cystatin C enhanced the infection by impairing the excessive cathepsin B activity. Consistent with this idea, pretreatment of HeLa cells with 125 µM of CA-074Me, a cathepsin B inhibitor, resulted in an 8-fold enhancement of the mNDK infectivity. Because cathepsin B is activated by low pH in acidic endosomes, we further examined the potential roles of endosomes in the CD4-independent infection. Suppression of endosome acidification or endocytosis by inhibitors or by an Eps15 dominant negative mutant reduced the infectivity of mNDK in which CD4-dependent infections were not significantly impaired. Taken together, these results suggest that endocytosis, endosomal acidification, and cathepsin B activity are involved in the CD4-independent entry of HIV-1.

Cathepsin L is required for ecotropic murine leukemia virus infection in NIH3T3 cells

Abstract Recently it has been reported that a cathepsin B inhibitor, CA-074Me, attenuates ecotropic murine leukemia virus (Eco-MLV) infection in NIH3T3 cells, suggesting that cathepsin B is required for the Eco-MLV infection. However, cathepsin B activity was negative or extremely low in NIH3T3 cells. How did CA-074Me attenuate the Eco-MLV infection? The CA-074Me treatment of NIH3T3 cells inhibited cathepsin L activity, and a cathepsin L specific inhibitor, CLIK148, attenuated the Eco-MLV vector infection. These results indicate that the suppression of cathepsin L activity by CA-074Me induces the inhibition of Eco-MLV infection, suggesting that cathepsin L is required for the Eco-MLV infection in NIH3T3 cells. The CA-074Me treatment inhibited the Eco-MLV infection in human cells expressing the exogenous mouse ecotropic receptor and endogenous cathepsins B and L, but the CLIK148 treatment did not, showing that only the cathepsin L suppression by CLIK148 is not enough to prevent the Eco-MLV infection in cells expressing both of cathepsins B and L, and CA-074Me inhibits the Eco-MLV infection by suppressing both of cathepsins B and L. These results suggest that either cathepsin B or L is sufficient for the Eco-MLV infection.

Characterization of R peptide of murine leukemia virus envelope glycoproteins in syncytium formation and entry

SummaryThe C-terminal R peptide of ecotropic murine leukemia virus (MLV) envelope protein (Env) negatively controls membrane fusion activity. The R peptide cleavage during virion maturation activates its fusogenicity and is required for viral entry. We analyzed fusogenicity and transduction efficiency of mutant Env proteins of ecotropic, amphotropic, polytropic, and xenotropic MLVs. As the result, we found that the hydrophobic amino acid residues around the R peptide cleavage site are important for membrane fusion inhibition by the R peptide. In addition, we found that Env complexes with R peptide-truncated and -containing Env proteins have lower fusogenicity and transduction efficiency than those with the R-peptide-truncated Env alone, suggesting that efficient R peptide cleavage is required for efficient MLV vector transduction. The role of R peptide cleavage in amphotropic, polytropic, and xenotropic MLV infection has not been investigated. We found in this study that the R peptide cleavage is required for amphotropic, xenotropic, and polytropic MLV vector transduction, like with ecotropic MLV. The R-peptide-truncated Env proteins of the xenotropic and polytropic MLVs, however, had much lower fusogenicity than those of the ecotropic and amphotropic MLVs. These results provide valuable information for construction of efficient MLV vectors and for understanding the retroviral entry mechanism.

Infection of XC Cells by MLVs and Ebola Virus Is Endosome-Dependent but Acidification-Independent

Inhibitors of endosome acidification or cathepsin proteases attenuated infections mediated by envelope proteins of xenotropic murine leukemia virus-related virus (XMRV) and Ebola virus, as well as ecotropic, amphotropic, polytropic, and xenotropic murine leukemia viruses (MLVs), indicating that infections by these viruses occur through acidic endosomes and require cathepsin proteases in the susceptible cells such as TE671 cells. However, as previously shown, the endosome acidification inhibitors did not inhibit these viral infections in XC cells. It is generally accepted that the ecotropic MLV infection in XC cells occurs at the plasma membrane. Because cathepsin proteases are activated by low pH in acidic endosomes, the acidification inhibitors may inhibit the viral infections by suppressing cathepsin protease activation. The acidification inhibitors attenuated the activities of cathepsin proteases B and L in TE671 cells, but not in XC cells. Processing of cathepsin protease L was suppressed by the acidification inhibitor in NIH3T3 cells, but again not in XC cells. These results indicate that cathepsin proteases are activated without endosome acidification in XC cells. Treatment with an endocytosis inhibitor or knockdown of dynamin 2 expression by siRNAs suppressed MLV infections in all examined cells including XC cells. Furthermore, endosomal cathepsin proteases were required for these viral infections in XC cells as other susceptible cells. These results suggest that infections of XC cells by the MLVs and Ebola virus occur through endosomes and pH-independent cathepsin activation induces pH-independent infection in XC cells.

Fragments of Target Cells are Internalized into Retroviral Envelope Protein-Expressing Cells during Cell-Cell Fusion by Endocytosis

Retroviruses enter into host cells by fusion between viral and host cell membranes. Retroviral envelope glycoprotein (Env) induces the membrane fusion, and also mediates cell-cell fusion. There are two types of cell-cell fusions induced by the Env protein. Fusion-from-within is induced by fusion between viral fusogenic Env protein-expressing cells and susceptible cells, and virions induce fusion-from-without by fusion between adjacent cells. Although entry of ecotropic murine leukemia virus (E-MLV) requires host cell endocytosis, the involvement of endocytosis in cell fusion is unclear. By fluorescent microscopic analysis of the fusion-from-within, we found that fragments of target cells are internalized into Env-expressing cells. Treatment of the Env-expressing cells with an endocytosis inhibitor more significantly inhibited the cell fusion than that of the target cells, indicating that endocytosis in Env-expressing cells is required for the cell fusion. The endocytosis inhibitor also attenuated the fusion-from-without. Electron microscopic analysis suggested that the membrane fusion resulting in fusion-from-within initiates in endocytic membrane dents. This study shows that two types of the viral cell fusion both require endocytosis, and provides the cascade of fusion-from-within.

Androgen-independent proliferation of LNCaP prostate cancer cells infected by xenotropic murine leukemia virus-related virus

Xenotropic murine leukemia virus-related virus (XMRV) is a novel gammaretrovirus that was originally isolated from human prostate cancer. It is now believed that XMRV is not the etiologic agent of prostate cancer. An analysis of murine leukemia virus (MLV) infection in various human cell lines revealed that prostate cancer cell lines are preferentially infected by XMRV, and this suggested that XMRV infection may confer some sort of growth advantage to prostate cancer cell lines. To examine this hypothesis, androgen-dependent LNCaP cells were infected with XMRV and tested for changes in certain cell growth properties. We found that XMRV-infected LNCaP cells can proliferate in the absence of the androgen dihydrotestosterone. Moreover, androgen receptor expression is significantly reduced in XMRV-infected LNCaP cells. Such alterations were not observed in uninfected and amphotropic MLV-infected LNCaP cells. This finding explains why prostate cancer cell lines are preferentially infected with XMRV.

Role of Ezrin Phosphorylation in HIV-1 Replication

Host-cell expression of the ezrin protein is required for CXCR4 (X4)-tropic HIV-1 infection. Ezrin function is regulated by phosphorylation at threonine-567. This study investigates the role of ezrin phosphorylation in HIV-1 infection and virion release. We analyzed the effects of ezrin mutations involving substitution of threonine-567 by alanine (EZ-TA), a constitutively inactive mutant, or by aspartic acid (EZ-TD), which mimics phosphorylated threonine. We also investigated the effects of ezrin silencing on HIV-1 virion release using a specific siRNA. We observed that X4-tropic HIV-1 vector infection was inhibited by expression of the EZ-TA mutant but increased by expression of the EZ-TD mutant, suggesting that ezrin phosphorylation in target cells is required for efficient HIV-1 entry. Expression of a dominant-negative mutant of ezrin (EZ-N) and ezrin silencing in HIV-1 vector-producing cells significantly reduced the infectivity of released virions without affecting virion production. This result indicates that endogenous ezrin expression is required for virion infectivity. The EZ-TD but not the EZ-TA inhibited virion release from HIV-1 vector-producing cells. Taken together, these findings suggest that ezrin phosphorylation in target cells is required for efficient HIV-1 entry but inhibits virion release from HIV-1 vector-producing cells.