Timothy S. Deeb
In this paper the investigators seemed to have demonstrated that dendritic cells (DC) which express both DC-SIGN, a DC-specific adhesion receptor, and CD4 "preferentially use DC-SIGN to capture HIV-1 (at low virus titer) via its (DC-SIGN's) high affinity for HIV-1 gp120." The investigators also demonstrated that DC-SIGN could retain HIV-1 in its native infectious state for prolonged periods of time in which DC travel to lymph nodes from mucosa via lymphatics. The investigators went on to show that DC-SIGN could subsequently enhance trans-infection of HIV-1 to replication-permissive T cells expressing CD4 and chemokine receptors (such as CCR5).
Figure 1 is composed of a set of figures which demonstrate that DC-SIGN is a DC-specific surface receptor for the HIV-1 gp120 envelope glycoprotein. Figure 1A consists of results for a representative experiment showing histograms for FACScan analysis of peripheral blood lymphocytes (PBL), monocytes and immature DC with anti-DC-SIGN antibodies (AZN-D1). The histograms indicate that PBL and monocytes express little or no DC-SIGN, while immature DC clearly express high levels of DC-SIGN.
Figure 1B shows the results for a representative experiment of a flow cytometric adhesion analysis which measured the competence of HIV-1 gp120-coated fluorescent beads to bind to immature DC. Immature DC from the medium bound well to the HIV-1 gp120-coated fluorescent beads. Anti-CD4 antibodies did not inhibit the binding of the HIV-1 gp120-coated beads to immature DC; in the presence of neutralizing anti-CD4 antibodies, the gp120-coated beads bound to immature DC slightly more efficiently (~1% more) than in the absence of anti-CD4 antibodies. Anti-DC-SIGN antibodies (AZN-D1 and AZN-D2) significantly blocked binding of the HIV-1 gp120-coated beads to immature DC. EGTA (~10%) and mannan (~6%) also blocked adhesion of the HIV-1 gp120-coated beads to immature DC, but to a lesser extent than AZN-D1 and AZN-D2. (Percentages refer to HIV-1 gp120 binding to immature DC).
Figure 1C shows histograms for flow cytometric analysis of CD4, CCR5, and DC-SIGN expression on THP-1 cells transfected with DC-SIGN (THP-DC-SIGN) and on immature DC. Immature DC (right three panels) express low levels of CD4 and CCR5, while expressing high levels of DC-SIGN. THP-DC-SIGN cells do not express CD4 or CCR5; THP-DC-SIGN does express high levels of DC-SIGN.
Figure 1D shows the results of a flow cytometric adhesion analysis which measured the ability of HIV-1 gp120-coated fluorescent beads to bind to THP-DC-SIGN and THP-1 MOCK cells. The HIV-1 gp120-coated beads bound efficiently to THP-DC-SIGN, and did not bind to THP-1 MOCK (negative control). Anti-CD4 antibodies did not inhibit the binding of the HIV-1 gp120-coated beads to THP-DC-SIGN. Anti-DC-SIGN antibodies blocked binding of the HIV-1 gp120-coated beads to THP-DC-SIGN. Adhesion was also inhibited by EGTA.
From figure1, we can conclude that DC-SIGN does bind to the HIV-1 gp120 beads without co-expression of CD4 and CCR5, that HIV-1 gp120 preferentially binds to DC-SIGN over CD4, that anti-DC-SIGN antibodies as well as EGTA and mannan effectively inhibit HIV-1 gp120 binding to DC and THP-DC-SIGN, and that anti-CD4 antibodies do not inhibit HIV-1 gp120 binding to DC and THP-DC-SIGN. I question whether CD4 and CCR5 coexpression with DC-SIGN increase DC-SIGN's competence to bind to HIV-1 gp120. I would test this by transfecting THP-1 cells with 1)CD4, 2)CD4 and CCR5, 3)CD4 and DC-SIGN, 4)CD4, CCR5, and DC-SIGN, 5) CCR5, 6) CCR5 and DC-SIGN, 7) DC-SIGN (again to compare), and 8) THP-1 control (no transfectants); I would then perform a fluorescent bead adhesion assay as described by Geijtenbeek et al. to measure HIV-1 gp120 binding. I would then compare HIV-1 gp120 binding between the transfected THP-1 cells and determine if CD4 and/or CCR5 affect DC-SIGN's ability to bind to HIV-1 gp120. Additional experiments can be done to measure the effects of anti-DC-SIGN and anti-CD4 antibodies as well as CCR5-specific chemokines on HIV-1 gp120 binding of the hypothetical THP-1 transfectants (again measured by fluorescent bead adhesion assay).
Figure 2 examines the role of DC-SIGN in HIV-1 infection in DC-T cell cocultures. Figure 2A shows the results of an ELISA (enzyme-linked immunosorbent assay) for DC pre-infected with anti-DC-SIGN antibodies, anti-CD4 antibodies, and/or CCR5-specific chemokines, in inhibiting HIV-1 infection measured in a DC-T cell coculture. Inhibition effects were compared to DC lacking antibodies and chemokines (control medium). Anti-CD4 antibodies alone did not block HIV-1 replication. The CCR5-specific chemokine trio (RANTES, MIP-1alpha, and MIP-1beta) alone did not inhibit HIV-1 infection. Anti-DC-SIGN antibodies alone did inhibit HIV-1 infection. The CCR5-specific chemokine trio and anti-CD4 antibodies did block HIV-1 infection.
Figure 2B shows the results of an analysis for DC pre-infected with different combinations of anti-DC-SIGN antibodies, anti-CD4 antibodies and the CCR5-specific chemokine trio in inhibiting HIV-1 infection measured in a DC-T cell coculture. Inhibition effects were compared to DC with no antibodies or chemokines (control). Anti-CD4 antibodies alone did not significantly inhibit HIV-1 infection. The CCR5-specific chemokine trio alone did not significantly block HIV-1 infection. Anti-DC-SIGN antibodies alone did significantly did significantly inhibit HIV-1 infection. Anti-CD4 antibodies in combination with the CCR5-specific chemokine trio blocked HIV-1 infection most effectively. Anti-DC-SIGN antibodies in combination with the CCR5-specific chemokine trio, anti-CD4 antibodies, or anti-CD4antibody-CCR5-specific chemokine trio inhibited HIV-1 infection.
Figure 2C shows the results of an assay for DC post-infected with anti-DC-SIGN antibodies, anti-CD4 antibodies, or the CCR5-specific chemokine trio in inhibiting HIV-1 infection. Inhibition effects were compared to DC lacking antibodies and chemokines (control). Anti-CD4 antibodies effectively inhibited HIV-1 infection. The CCR5-specific chemokine trio also inhibited HIV-1 infection. Anti-DC-SIGN antibodies did not effectively inhibit HIV-1 infection of activated T cells.
In figure 2, anti-DC-SIGN antibodies were efficient in inhibiting HIV-1 infection in the DC-T cell coculture, when added prior to HIV-1 infection. Anti-DC-SIGN antibodies did not block HIV-1 infection in the DC-T cell coculture, when added post-HIV-1 infection of DC, whereas the combination of anti-CD4 antibodies and CCR5-specific chemokines did. Figure 2 suggests that DC-SIGN facilitates the transmission of HIV-1 from HIV-1 pre-infected DC to T cells, but does not facilitate HIV-1 transmission to T lymphocytes from HIV-1 post-infected DC. The investigators indicate from previous experiments that DC-SIGN binds to ICAM-3 on T lymphocytes. The investigators suggest that it is DC-SIGN's ability to bind to gp120, not the adhesive interactions of DC-SIGN with ICAM-3, that is responsible for the propagation of HIV-1 in DC-T cell cocultures (independent of the presence of ICAM-3). To determine whether ICAM-3 plays an important role in HIV-1 infection of a DC-T cell coculture, I would repeat the experiments performed in figure 2 (measuring HIV-1 production using p24 antigen ELISA) with the following modification: substitute the coculture T cells with ICAM-3 negative T cells (if possible) or introduce an ICAM-3-specific inhibitor which prevents adhesion of ICAM-3 to the DC. From the results of such an experiment, the contribution, if any, of ICAM-3's adhesive properties between DC and T cells to HIV-1 infection could be determined. If ICAM-3 proved to be significant in facilitating HIV-1 infection in a DC-T cell coculture, I would perform similar experiments measuring other adhesion molecules such as LFA-1 and ICAM-1.
The investigators then went on to determine if DC-SIGN facilitates HIV-1 entry into 293T cells. Figure 3A shows the results of an ELISA measuring HIV-1 entry into 293T cells transfected with DC-SIGN or CD4 and CCR5. 293 T cells lacking DC-SIGN, CD4 and CCR5 did not show a significant increase in p24 antigen levels during the days following the HIV-1 pulse. 293T cells expressing CD4 and CCR5 showed a significant gradual increase in p24 antigen levels, thus indicating the 293T-CD4-CCR5 cells were infected. 293T cells expressing DC-SIGN were not readily infected.
Figure 3B shows the amount of luciferase activity of transfected 293T cells which were infected with a replication defective HIV-1 virus encoding a luciferase reporter gene. 293T cells not expressing DC-SIGN, CD4, or CCR5 were used as a control (negative) showing low levels of luciferase activity. 293T-DC-SIGN, 293T-CD4, 293T-CCR5, and 293T-CCR5-DC-SIGN cells exhibited low levels of luciferase activity. 293T cells expressing CD4 and DC-SIGN showed a slight increase in the level of luciferase activity, indicating HIV-1 infection occurred. 293T-CD4-CCR5 cells and 293T-CD4-CCR5-DC-SIGN cells showed high levels of luciferase activity, indicating HIV-1 infection occurred.
Figure 3 demonstrates that HIV-1 entry is not detected in 293T cells which express only DC-SIGN; therefore, DC-SIGN expressed alone does not appear to mediate HIV-1 entry into 293T cells. The investigators suggest that coexpression of DC-SIGN with either CCR5 or CD4 on 293T cells does not result in infection. However, the data indicate HIV-1 infection occurred in 293T-CCR5-DC-SIGN cells (slight increase in luciferase activity is visible) and 293T-CD4-DC-SIGN. If, as the investigators suggest, DC-SIGN does not facilitate HIV-1 entry into 293T cells by working in conjunction with either CCR5 or CD4, then the luciferase activity levels of DC-SIGN coexpressed with either CD4 or CCR5 should not differ from the luciferase activity levels of 293T cells expressing only CD4 or CCR5. The data indicate that DC-SIGN does work in conjunction with CCR5 (slightly), CD4, or CD4-CCR5 (slightly) to facilitate HIV-1 entry into 293T cells. The data also indicate that HIV-1 entry into 293T cells can occur independently of DC-SIGN, as shown by 293T-CD4-CCR5.
The investigators hypothesized that "in a DC-T cell coculture DC-SIGN might facilitate both capture of HIV-1 on DC, independent from CD4 and CCR5, and subsequent transmission of HIV-1 to the CD4/CCR5-positive T cells." Figure 4A shows the level of HIV-1 infection by measuring luciferase activity (cells were infected with HIV-luciferase virus and washed after 2 hours). DC-SIGN negative THP-1 cells (control) did not "capture" HIV-1 and "transmit HIV-1" to either 293T-CD4-CCR5 cells or activated primary T cells. Antibodies against DC-SIGN (AZN-D1 and AZN-D2) blocked HIV-1 infection. THP-1 cells transfected with DC-SIGN successfully captured and transmitted HIV-1 to both 293T-CD4-CCR5 and activated primary T cells. CD4 positive T cells did not capture and transmit HIV-1 to activated primary T cells.
Figure 4B is a representative experiment showing the level of HIV-1 infection by measuring luciferase activity of "HIV-luciferase virus pseudotyped with the CCR5-specific HIV-1 envelopes from JRFL and JRCSF and from primary viruses 92US715.6, 92BR020.4, and 93TH966.8 (R5 isolates of HIV-1)." DC-SIGN negative THP-1 cells (control) did not capture HIV-1 and transmit HIV-1 to 293T-CD4-CCR5. THP-1 cells expressing DC-SIGN were successful in capturing and transmitting the HIV-1 luciferase viruses pseudotyped with R5isolates' envelopes.
Figure 4C is a density plot of a representative experiment in which THP-1 cells transfected with DC-SIGN were incubated with HIV-eGFP (green fluorescent protein reporter gene, a pseudotyped HIV-1 vector) and then cocultured with activated T cells. CD3 negative THP-DC-SIGN cells did not express the HIV-1-encoded GFP, indicating they were not infected by HIV-1. CD3 positive T cells which expressed eGFP were infected.
From figure 4, we can conclude, in accordance with the investigators, that HIV-1 capture is DC-SIGN dependent, cells expressing only DC-SIGN were competent to capture and transfer HIV-1 to cells expressing CD4 and CCR5, and only CD3 positive T cells were infected with HIV-1. I can agree with the investigators that DC-SIGN facilitated HIV-1 infection is not dependent on ICAM-3, but I do not agree that ICAM-3 cannot facilitate DC-SIGN mediated HIV-1 infection of target cells in trans. To determine whether ICAM-3 plays an important role in HIV-1 infection of target T cells, I would repeat the experiment done in 4 A with the following modifications: 1) don't add infected cells to 293T-CD4-CCR5 cells, and 2) perform an additional experiment in which an ICAM-3-specific inhibitor would be introduced to the activated primary T cells before adding HIV-1 infected (then washed) THP-DC-SIGN cells. I would compare the results of the THP-DC-SIGN cells added to the activated primary T cells in the presence of an ICAM-3 inhibitor to the results from THP-DC-SIGN cells added to activated primary T cells in the absence of the ICAM-3 inhibitor.
The investigators then went on to test DC-SIGN's ability to capture HIV-1 and present HIV-1 to target cells, when the virus is present in low levels. Figure 5A is a representative experiment showing the level of trans HIV-1 infection by measuring luciferase activity; cells were infected with low titers of pseudotyped HIV-1 and subsequently added (without washing) to either 293T-CD4-CCR5 or activated T cells. In the absence of THP-1, THP-DC-SIGN, and CD4 positive T cells, the HIV-1 virus (control showing low levels of luciferase activity) was not efficiently transmitted to either 293T-CD4-CCR5 or activated T cells. Antibodies against DC-SIGN slightly blocked HIV-1 transmission from THP-1 cells to either 293T-CD4-CCR5 or activated T cells. Anti-DC-SIGN antibodies inhibited THP-DC-SIGN from transmitting HIV-1 to activated T cells. Antibodies against DC-SIGN partially inhibited THP-DC-SIGN from transmitting HIV-1 to 293T-CD4-CCR5 cells. THP-1 cells transmitted low levels low levels of HIV-1 to 293T-CD4-CCR5 and to activated T cells. THP-DC-SIGN efficiently captured and transmitted HIV-1 to 293T-CD4-CCR5 and to activated T cells. CD4 positive T cells did not effectively transmit HIV-1 to activated T cells.
Figure 5B is a representative experiment showing the level of HIV-1 by measuring luciferase activity of R5 isolates of HIV-1 (JRFL, JRCSF, 93TH966.8, 92US715.6, 92BR020.4). In the absence of THP-1 cells and THP-DC-SIGN, the pseudotyped HIV-1 viruses were not transmitted to primary T cells. THP-1 cells did not transmit high levels of the pseudotyped HIV-1 viruses to primary T cells. THP-DC-SIGN cells effectively transmitted (high levels of luciferase activity) the pseudotyped HIV-1 viruses to primary T cells.
From figure 5, the investigators indicated that "DC-SIGN not only sequesters HIV-1 but also enhances CD4-CCR5-mediated HIV-1 entry by presentation in trans to the HIV-1 receptor complex." The investigators also suggest that anti-DC-SIGN antibodies "completely inhibited infection." However, the data indicate that anti-DC-SIGN antibodies only partially inhibited THP-DC-SIGN from transmitting HIV-1 to 293T-CD4-CCR5 cells. Anti-DC-SIGN antibodies, as indicated above, almost completely inhibited transmission of HIV-1 from THP-DC-SIGN to activated T cells. Although THP-1 cells seem to have some ability to facilitate transmission of HIV-1 cells to CD4/CCR5 positive T cells, we can conclude from figure 5 that DC-SIGN effectively enhances HIV-1 entry into CD4/CCR5-positive, as the investigators mentioned.
Figure 6 shows the results of immunohistochemical analysis to determine whether DC expressing DC-SIGN are active in viral infection in vivo. Figure 6A shows that the mucosal tissues of the cervix, rectum, and uterus possess DC expressing DC-SIGN, in the lamina propria.
Figure 6B is an immunohistochemical analysis comparing expression of DC-SIGN, CD4 and CCR5 on DC in the mucosal tissues of the rectum and the uterus. Most of the DC expressing DC-SIGN coexpressed CD4 and only a minority of DC coexpressed CCR5.
From figure 6 we learn that DC-SIGN is abundantly expressed in the lamina propria of HIV-1-related mucosal tissue, and that DC-SIGN is coexpressed with CD4 on DC in the HIV-1 related mucosal tissue. To determine whether CCR5 expressed on DC plays a role in facilitating the capture and transmission of HIV-1 with DC-SIGN, I would repeat the experiments performed in figure 2 (measuring HIV-1 production using p24 antigen ELISA) with the following modification: use CCR5 negative DC and see if similar results to those found in figure 2 are obtained.
Figure 7 shows the results of experiments done to determine how long DC-SIGN-bound HIV-1 retains infectivity. Figure 7A shows the results of a FACScan analysis used the time course of HIV-1 gp120 binding to THP-1 cells expressing DC-SIGN. HIV-1 gp120 beads exhibited high levels of binding for approximately 60 hours in the absence of anti-DC-SIGN antibodies. HIV-1 gp120 beads exhibited low levels of binding for approximately 60 hours in the presence of anti-DC-SIGN antibodies.
Figure 7B shows examination of luciferase activity to measure the time duration for which pseudotyped HIV-1 pulsed THP-DC-SIGN cells retained infectious HIV-1. As a control, virus in the absence of cells, showed a dramatic decrease in infectivity after 1 day; HIV-1 infectivity (control) continued to decrease. DC-SIGN negative THP-1 cells did not show significant levels of luciferase activity (HIV-1 infectivity in 293T-CD4-CCR5 cells) throughout the seven day incubation period. THP-DC-SIGN cells in the presence of anti-DC-SIGN antibodies did not efficiently infect 293T-CD4-CCR5 cells during the seven-day incubation period. THP-1 cells expressing DC-SIGN in the absence of anti-DC-SIGN antibodies, were efficiently able to infect 293T-CD4-CCR5 cells up until the fifth day of the seven day incubation period.
Figure 7C is a model showing HIV-1 capture and in trans infection of T cells by DC-SIGN.
Figure 7 demonstrates that DC-SIGN is able to capture and bind HIV-1 efficiently for more than 4 days. The investigators suggest that "this long-term preservation of HIV-1 in an infectious state" allows "sufficient time for HIV-1 to be transported by DC trafficking from mucosal surfaces to lymphoid compartments, where virus can be transmitted."
The investigators suggest that DC utilize DC-SIGN to capture HIV-1 present in low levels in the mucosal tissues; after traveling to lymphoid compartments, the investigators suggest DC utilize DC-SIGN for presentation of HIV-1 to CD4/CCR5 complexes, and enhancement of HIV-1 entry into T cells expressing CD4/CCR5 complexes. The in vitro experiments performed in this paper in addition to the pattern of expression of receptors found in the mucosal tissues, indicate that DC-SIGN plays a major role in the early stages of viral infection. The actual mechanism of DC-SIGN capture and enhancement of HIV-1 infectivity is difficult to test in vivo in humans, since this process most likely occurs within four days of HIV-1 penetration into the mucosal tissues. Perhaps, when the appropriate procedures are designed, in vivo testing in animals may illuminate the mechanism of enhanced infectivity by DC-SIGN in trans. Future in vivo experiments may also reveal the efficacy of antibodies which inhibit DC-SIGN binding to HIV-1 envelopes as well as HIV-1 entry into target cells through CD4 and CCR5 during early stages of infection.