Humun herpes virus 8 (HHV-8) infects specific human cell types

This page was created as an assignment for an undergraduate course at Davidson College.

Akula, Shaw M., Naranatt Pramond, Fu-Zhang Wang, Bala Chandran. 2002. Integrin a1b2 (CD 49c/29) Is a Cellular Receptor for Kaposi’s Sarcoma-Associated Herpesvirus (KSHV/HHV-8) Entry into the Target Cells. Cell. 108: 407-419.

Human herpes virus 8 (HHV-8) infects specific human cell types. When HHV-8 infects cells, the HHV-8 proteins, glycoprotein B(gB) and gpk8.1A, interact with a heparin sulfate like complex on the host cell’s surface. Glycoprotein B(gB), expresses a RGD motif that is conserved in many HHV strains. The RGD motif of other proteins is known to interact with integrins on a cell’s surface when substances are transported into the cells. Many viruses enter their host cells via a RGD motif-integrin interaction. This study asks whether HHV-8 uses the RGD motif-integrin interaction to infect host cells.

After HHV-8 infects a cell, HHV-8 enters a latent cycle. Thus infectivity can not be measured using the traditional plaque assay. HHV-8 infectivity was assessed by measuring the fluorescence intensity of cells infected by an HHV-8 virus expressing a GFP-gB fusion protein or by measuring the staining intensity of a monoclonal antibody against the HHV-8 protein encoded by ORF73.

Cells incubated with a short RGD peptide prior to the addition of HHV-8 showed less fluorescence and staining intensity (Figure 1C,F) than cells not incubated with an RGD peptide (Figure 1B,E). Cells incubated with a mAb specific to the gB RGD motif or with fibronectin prior to infection were infected less than control cells (Figure 2B-C). Fibronection interacts with integrin molecules via the RGD motif. Infection decreased as the concentration of anti-gB-RGD antibodies, RGD peptides, and fibronectin increased (Figure 2). Increasing competition to the RGD motif of HHV-8 gB or to the RGD integrin receptor decreased the ability of HHV-8 to infect cells. HHV-8 gB-RGD motifs and RGD integrin receptors appear to be involved when HHV-8 infects cells. Interestingly, incubating cells with laminin, another protein known to interact with integrin molecules via the RGD motif, increased the HHV-8 infectivity of cells (Figure 2C).

The ability of HHV-8 to infect cells incubated with antibodies to several integrin subunits was measured. Integrin is a heterodimer composed of two subunits. Each heterodimer has a different binding specificity. Infectivity was significantly decreased in cells incubated with antibodies to the a3, b1, and a2b1 subunits when compared to cells infected with other anti integrin subunit antibodies (Figure 3A). Infectivity of cells incubated with antibodies to a2b1 and b1 did not differ significantly. Minimal inhibition occurred when cells were incubated with the an antibody to a5b1 (Figure 3A). Infectivity decreased as the concentration of anti-a3, anti-b1, and anti-a2b1 was increased (Figure 3B). Infectivity decreased the most when cells were incubated with all three antibodies, anti-a3, b1, and a2b1, prior to infection (Figure 3B). Infectivity also decreased in cells incubated with a3b1 soluble integrins but not a5b1 integrins (Figure 3C). Increasing false a3b1 integrin binding sites or increasing competition for the a3, b1, or a2b1 binding sites decreased HHV-8 infectivity.

Expression of different integrin subunits in host cells was determined using FACS. a3 and b1 had higher fluorescent intensities than IgG and aVb3 (Figure 4A-D). The percent of cells expressing the integrin was calculated for 5 different cell types known to act as hosts for HHV-8. In 3 of the cell types, the a3 subunit was expressed in more than 90 percent of the cells and the b1 subunit was expressed in more than 87 percent of all 5 of the cell types tested (Figure 4E).

Hamster CHO cells expressing low levels of a1 were transfected with human a1 subunit cDNA. Two clones expressing the human a1 subunit were created. FACS was used to measure the expression of human a3 in B3 and D5 clones. The B3 clones expressed four times as much human a3 as did the D5 clones (Figure 5A). B3 and D5 clones were infected with GFP-HHV8 and B3 cells showed more fluorescence and mAb binding to HHV-ORF73 when compared to D5 cells (Figure 5B). Both D5 and B3 cells showed less infectivity when incubated with anti-a3 Ab but not anti-b4 Ab prior to infection with HHV-8 (Figure 5C). Increasing the a3 integrin receptors on the host cells increased infectivity.

Protein from cells that were either mock infected or infected with HHV-8 were immunoprecipitated with either anti-gB or anti-IgG antibodies. Immunoprecipitation products were resolved by SDS-PAGE. A western blot of the SDS PAGE was probed with either anti-a3, b1, or a1 mAb’s. a1 was selected as a control because it is expressed at similar levels in HFF cells. No bands appeared when lanes were loaded with proteins immunoprecipitated with anti-gpk8.1 antibodies or probed with anti-a1 antibodies (Figure 6A-lane 9,12-14). Bands were only seen in lanes loaded with proteins immunoprecipitated with anti-gB and probed with anti-a3 or b1 antibodies (Figure 6A lanes 7,8,10,11). In some experiments heparinase was added to the cell-HHV8 incubation mixture to determine whether the HS interaction competed with the integrin interaction. The absence or presence of heparinase made no difference in the immunoprecipitation results (Figure 6A-lanes 7,8, 10,11). HHV-8 gB binds to the a3 and b1 subunits during infection.

Incubating cells with RGD peptides or anti-a3 and a1 subunit antibodies prior to the addition of a fixed concentration of radioactively labeled HHV-8 did not affect the ability of HHV-8 to bind to cells (Figure 6-B). Incubating the virus with soluble a3b1 subunits or anti-gB antibodies prior to infecting cells did not inhibit the ability of HHV-8 to bind to cells (Figure 6B). Thus blocking the integrin receptor on cells or the RGD binding site on the HHV-8 gB protein did not inhibit virus binding. Incubating the radioactively labeled HHV-8 with heparin before adding the mixture to cells did inhibit the ability of HHV-8 to bind to cells (Figure 6B). Heparin and not the a3b1 subunit is involved in the ability of HHV-8 to bind to cells. Thus a3b1 probably plays a role after HHV-8 binds to the cell.

The aggregation of FAK and vinculin at a focal adhesion point is one of the steps involved in the integrin mediated entrance into cells. To test if HHV-8 infection caused FAK and vinculin to aggregate, co-localization experiments on infected cells were done. Serum starved cells were either incubated with no virus, GFP-HHV-8, or LPA and then probed with a p125 FAK antibody and a vinculin antibody. LPA is known to cause FAK and vinculin to aggregate at focal adhesion points. Cells incubated with LPA showed the most co-localization of FAK and vinculin (Figure 7A). Cells incubated with GFP HHV-8 and LPA had similar co-localization patterns of FAK and vinculin (Figure 7A). Proteins of serum starved cells incubated with GFP HHV-8 were resolved on SDS-PAGE. A western blot of the SDS PAGE was probed with an anti-posphate FAK antibody. The bands on the western blot increased in intensity as the GFP-HHV8 incubation time increased (Figure 7B-lanes 3-5). In a similar experiment, GFP HHV-8 was incubated with soluble a3b1 integrin prior to infecting cells. As the concentration of soluble a3b1 in the GFP HHV-8 pretreatment decreased, phosphorylation of FAK increased (Figure 7C-lanes 2-3). Phosphorylation of FAK in cells infected with GFP HHV-8 pre-incubated with soluble a5b1 soluble integrin did not differ from cells that were infected with non pretreated GFP HHV-8 (Figure 7C-lane 5, 1). Cells infected with GFP HHV-8 pre-incubated with anti-gB antibodies showed increased phosphorylation of FAK as concentration of anti gB decreased (Figure 7D-lane2-4). HHV-8 infection induces the phosphorylation of FAK and increasing competition to either the gB or integrin binding site decreases FAK phosphorylation.

The authors correctly concluded that HHV-8 gB interacts with the a3b1 integrin after binding to the cell and that HHV-8 infectivity induces FAK phosphorylation. However the authors incorrectly concluded that HHV-8 interacts with the a3b1 integrin and not the a2b1 integrin. The authors proposed that since a2b1 results did not significantly differ from b1 results, that the inhibition caused by the a2b1 antibody was due to cross reactivity with the a1 subunit. However, the a5b1 subunit did not inhibit HHV-8 reactivity (Figure 2). Thus, the involvement of a2b1 in HHV-8 experiments can not be ruled out and further experiments should be done. To determine if a2b1 interacts with gB, proteins from infected cells should be immunoprecipitated with an anti gB antibody and resolved on SDS PAGE. The western blot of the SDS PAGE should be probed with an antibody to the a2 subunit.

Glycoprotein B and gpk8.1 interact with heparin, however, HS is not necessary for gB-a3b1 interaction (Figure 6A). The authors suggest that gB either has a separate heparin binding domain (108-117aa) or that gB interacts with heparin via another protein, such as gpk8.1. To determine if gpk and gB interact, proteins from infected cells should be immunoprecipitated with an antibody to gB and then resolved on SDS PAGE. A western blot of the SDS PAGE should be probed with an antibody to gpk8.1. A nice control would be to switch the order in which the antibodies are added to see if the same result is obtained. If gpk and gB interact then a second immunoprecipitation experiment using a gB protein that has the possible heparin binding domain mutated should be done to determine if gpk and gB interact via this domain.

Figure 6 showed that heparin but not gB inhibited HHV-8’s ability to bind to cells. The same experiment should be done to determine if gpk is involved in HHV-8 binding to cells. The authors have already begun preliminary experiments to determine if HHV-8 enters the cell endocytotically. Experiments should also be designed to determine what happens to the cell after the virus enters and FAK is phosphorylated. HIV seems to induce HHV-8 to cause cells to become cancerous. An experiments to determine if HIV and HHV-8 interact when both viruses are introduced to the same cell should also be designed.