Cell, Vol. 100, 587-597, March 3, 2000

This paper examined the role of dentritic cells in transmission of HIV-1 virus.  The authors describe in detail the role of DC-SIGN, a DC-specific adhesion receptor that binds to the HIV-1 envelope glycoprotein gp120 and is highly expressed on DC present in mucosal tissues.  They argue that DC-SIGN recruits HIV-1 and uses a new in trans mechanism to catalyze the entry of HIV-1 into cells that express CD4 and chemokine receptors.  HIV-1 may then use the migratory capacity of DC to gain access to the T cell compartment in lymphoid tissues.  These findings are important because they may lead to strategies for preventing or blocking HIV-1 infection.

Figure 1. DC-SIGN is a DC-Specific Receptor for HIV-1 gp120: 1A is used to demonstrate that DC-SIGN is preferentially expressed on immature, in vitro cultured DC but not on other leukocytes, such as peripheral blood lymphocytes (PBL) or monocytes.  Expression of DC-SIGN was determined by FACScan analysis.  In 1B, a flow cytometric adhesion assay (showing the amount of HIV-1 gp120 that binds to DC) demonstrates that DC-SIGN (and not CD4) mediates binding of HIV-1 gp120 to DC.  The gp120-coated beads bound efficiently to the DC, and the binding was completely blocked by anti-DC-SIGN antibodies AZN-D1 and AZN-D2.  Anti-CD4 antibodies had no effect on gp120 binding to DC.  1C shows that immature DC express low levels of CD4 and CCR5 and high levels of DC-SIGN.  When THP-1 cells (monocytic cells that lack expression of CD4 and CCR5) are stably transfected with DC-SIGN, they express high levels of DC-SIGN and do not express CD4 or CCR5.  1D shows the binding of DC-SIGN transfectants (THP-DC-SIGN from figure 1C) to HIV-1 gp120.  Adhesion was blocked by anti-DC-SIGN antibodies and EGTA, but wasn’t blocked by anti-CD4 antibodies.  In 1B and 1D, EGTA and mannan are used as controls for the blocking of adhesion.
Bottom Line from Figure 1:  The authors show that, among the cell types they examined, DC-SIGN is expressed only by dentritic cells.  HIV-1 preferentially binds to DC-SIGN, as anti-DC-SIGN blocks adhesions much more strongly than anti-CD4 or anti-CCR5.  This preference is observed in DC cells and in DC-SIGN transfectants (1B and 1D, respectively).  A problem with this figure occurs in 1B; they do not show the effect of anti-CCR5 on binding.  They seem to assume it is the same as anti-CD4, but we cannot know for certain without the data.  Although the authors do not explain why immature DC express levels of CD4 and CCR5 higher than the control (as determined by isotyped matching of antibodies against them), the data support their claims, as will be discussed later in this paper.

Figure 2. DC-SIGN Mediates HIV-1 Infection in a DC-T Cell Coculture: 2A shows that antibodies against DC-SIGN inhibit HIV-1 infection in a DC-T cell coculture.  p24 antigen levels were measured by ELISA.  We see that anti-DC-SIGN as well as anti-CD4+CCR5 trio block infection of HIV-1 into DC, while anti-CD4 and CCR5 trio alone do not.  p24 values were measured daily to quantify the blocking ability of the antibodies.  Under these conditions, however, infections still occurred in the presence of antibodies, though they were delayed about 2 days.  2B represents the Day 5 results from Figure 2A.  2C investigates the possibility that, due to the affinity of DC-SIGN for ICAM-3 on T cells, DC-SIGN could interfere with the DC-T interaction and thereby prevent HIV-1 transmission.  The figure shows that DC-SIGN had no effect on transmission of DC-bound HIV-1 to T cells.  However, anti-CD4 and CCR5-specific chemokines strongly inhibited HIV-1 infection of activated T cells.
Bottom Line from Figure 2:  DC-SIGN has an important role in the propagation of HIV-1 in DC-T cell cocultures.  This function is related to the ability of DC-SIGN to bind to gp120 and not to its interaction with ICAM-3.  A weakness in this figure is that that authors fail to explain the purpose of 2B.  It seems to only be a slightly altered reproduction of the findings shown in 2A.

Figure 3. DC-SIGN Expressed on Target Cells Does Not Mediate HIV-1 Entry:  In 3A, ELISA and p24 were again used to show that in cells expressing DC-SIGN, HIV-1 did not enter the cells.  In cells expressing CD4 and CCR5, entry did occur. 3B employs luciferase to show that DC-SIGN does not form a complex with CD4 and/or CCR5 to permit viral entry.  High luciferase activity was obtained with infection of 293T cells with CD4 + CCR5, but not with any other combinations.  The addition of DC-SIGN to this combination did not affect the level of luciferase activity, so it is unlikely that a complex was formed among the three proteins.
Bottom Line from Figure 3:  DC-SIGN does not act as a receptor that permits HIV-1 entry.  It cannot substitute for CD4 or CCR5 in the process of HIV-1 entry.  There are no real problems with this figure.

Figure 4. DC-SIGN Captures HIV-1 that Retains Infectivity for CD4+ T Cells: THP-DC-SIGN transfectants do not express CD4 or CCR5 and cannot be infected by HIV-1 (data not shown).  In 4A, THP-1 and THP-DC-SIGN cells were pulsed with HIV-luciferase virus pseudotyped with HIV-1 envelope glycoprotein.  The cells were cocultured with cells permissive for HIV-1 infection.  Only the THP-DC-SIGN cells captured HIV-1 and transmitted it to target cells expressing receptors necessary for viral entry.  This was true whether the infected cells were added to 293T-CD4-CCR5 cells or to activated T cells.  In all cases, HIV-1 capture was completely blocked by anti-DC-SIGN.  In 4B, DC-SIGN was able to mediate capture of HIV-1 viruses from different isolates.  In both 4A and 4B, DC-SIGN-negative THP-1 cells were unable to capture or transmit HIV-1.  In 4C, the authors used a pseudotyped HIV-1 vector with GFP gene to demonstrate that cells expressing CD4/CCR5 were infected in cocultures, while those expressing DC-SIGN were not infected.  The CD3-negative THP-DC-SIGN cells were not infected by HIV-1, while the CD3-positive T cells expressed virus-encoded GFP.
Bottom Line from Figure 4:  This figure shows that cells which cannot be infected by HIV-1 (THP-1 cells) can still capture and transport HIV-1 to cells that can be infected (cells carrying CD4 + CCR5 and activated T cells), but only if DC-SIGN is expressed on the carrier cells.  HIV-1 will still retain its ability to infect HIV-1-permissive cells.  I would like to have seen another antibody used as a positive control (4A), perhaps anti-CD4+CCR5.  The antibody should not block the effects of DC-SIGN (as evidenced in Figure 2B).  They do not include it, likely believing Figure 2 is enough support, but I would still like to have seen it for consistency.

Figure 5.  DC-SIGN Enhances HIV-1 Infection of T Cells by Acting In trans:  5A and 5B are very similar in design to 4A and 4B.  They show that the same pattern of HIV-1 infection as seen in 4A and 4B, except the cells were infected with very low amounts of pseudotyped HIV-1 virus (and isolates).
Bottom Line from Figure 5:  This is evidence that DC-SIGN enhances CD4-CCR5-mediated HIV-1 entry by presentation in trans to the HIV-1 receptor complex.   Again, DC-SIGN is necessary for the CD4+CCR5-positive cells to be infected.

Figure 6. DC-SIGN Is Expressed on DC Present in Mucosal Tissue that Do Not Express CCR5:  In 6A, immunohistochemical analyses (using tissues stained with anti-DC-SIGN antibodies) of mucosal tissues were used to determine if cells containing DC-SIGN were present in vivo.  DC-SIGN was expressed on DC-like cells present in mucosal tissues, including cervix (a), rectum (b), and uterus (c).  DC-SIGN-expressing cells were distinct from T cells, B cells, monocytes, and macrophages (data not shown).  6B shows immunohistochemical staining of serial sections of rectum (a-c) and uterus (d-f) using antibodies against DC-SIGN.  This figure compared the expression of DC-SIGN, CD4, and CCR5 on DC, and found that the majority of DC-SIGN-positive DC in these tissues coexpressed CD4 but lacked CCR5.
Bottom Line from Figure 6:  The authors suggest these data demonstrate that DC present at mucosal sites (which first contact HIV-1 during sexual transmission) are not infected with HIV-1 through the use of CD4/CCR5.  Concerning this figure, I would like to see the data that weren’t shown which demonstrate that DC-SIGN-expressing cells were distinct from the other cell types.

Figure 7. CD-SIGN captures HIV-1 and Retains Long-Term Infectivity: 7A shows a time-course of HIV-1MN gp120 binding to THP-DC-SIGN.  We see that gp120-coated beads remained bound to DC-SIGN for more than 60 hours.  In the presence of anti-DC-SIGN, very little HIV-1-gp120 bound to DC.  7B shows the length of time during which HIV-1-pulsed THP-DC-SIGN cells could retain infectious virus.  As demonstrated by luciferase activity, infectivity declined after 3 days, but HIV-1-pulsed cells were still able to efficiently infect target cells 4 days after binding.  Meanwhile, the virus itself could only infect target cells within 1 day.  As shown by the lack of luciferase activity at any time, DC-SIGN antibodies blocked infection, while THP-1 cells never became infected.  7C is not actual data, it is a model of HIV-1 coopting DC-SIGN as a trans receptor after initial exposure.  In this model, HIV-1 targets DC cells first during mucosal exposure and are DC-SIGN positive.  DC-SIGN binds HIV-1, and the immature DC carries HIV-1 to the lymphoid tissue.  When they arrive, DC clusters with T cells, enhancing HIV-1 infection of T cells in trans.
Bottom Line from Figure 7:  The data from A and B indicate that the presence of DC-SIGN strongly increases the binding ability of DC cells for HIV-1.  Once bound, the virus can infect other cells up to 4 days later, much longer than the virus can alone.  It appears HIV-1 gp120 remains bound to DC-SIGN for more than 60 hours.  However, infectivity lasts up to about 96 hours (4 days).  It would have been interesting for the authors to see if HIV-1 gp120 can remain bound to DC-SIGN for 96 hours.  If not, they would have to devise a model to explain how the virus could infect in trans if it wasn’t bound to its carrier DC.

Overall, these authors propose a new model of HIV-1 infection.  They summarize their findings by saying that DC-SIGN captures HIV-1 in the periphery and facilitates its transport to secondary lymphoid organs rich in T cells, to enhance infection in trans of these target cells.  They do an excellent job of pairing their data with their claims, and each step of their model can be critiqued.

Claim 1:  DC-SIGN is a DC-specific HIV-1 binding protein.  Figure 1 is very clean.  The differences used for arguments are very clear.  We can see clearly that DC-SIGN is expressed in DC (A) more so than other leukocytes, and that anti-DC-SIGN antibodies block HIV-1 gp120 binding to DC while anti-CD4 antibodies do not (B).  The data in C concerning CD4 and CCR5 counts in immature DC cells offers evidence for a claim they make later, that DC-SIGN can interact with CD4 and CCR5.  All three are present in DC, so it is possible they interact.  In D, it is very clear that anti-CD4 does not block binding, while anti-DC-SIGN blocks it very well.  DC-SIGN certainly seems to be DC-specific, and it seems to be important for DC to bind to HIV-1 gp120.

Claim 2:  DC-SIGN is required for efficient HIV-1 infection in DC-T cell cocultures.  To support this claim, the authors say anti-DC-SIGN blocks HIV-1 infection.  However, infection under these conditions still occurs, as shown in Figure 2A.  Cells containing anti-CD4+CD5 trio and those containing anti-DC-SIGN reached the same level of infection as medium and anti-CD4 or CCR5 trio cells did.  For example, the “infected” cells had p24 levels of about 11,000 on day 7, while the “inhibited” cells reached 11,000 on day 9.  Infection still occurred, it was just delayed.  The authors do not directly examine this possibility.  However, they claim only that DC-SIGN is required for efficient HIV-1 infection, and this appears to be true.
The authors’ claims are better supported in Figure 2C.  It is apparent that anti-DC-SIGN has very little effect on ICAM-3, while anti-CD4 and CCR5 trio almost completely inhibited infection.  (Since 2B is the same data as 2A, it will not be further discussed.)

Claim 3:  DC-SIGN does not mediate HIV-1 entry.  They are saying that DC-SIGN does not behave like CD4 or CCR5 in acting as a receptor to permit HIV-1 entry.  It may mediate infection, but it isn’t the critical element that actually allows the virus to enter.  These data are very clear.  As is shown in 3A, 293T cells did not allow HIV-1 infection, nor did 293T cells transfected with DC-SIGN.  293T cells transfected with CD4+CCR5 did allow HIV-1 entry.  Since no other variables were present, it is clear that DC-SIGN is not responsible for HIV-1 entry under those conditions.  In 3B, the presence of DC-SIGN did not affect the amount of infection allowed by CCR5 or CD4+CCR5, while there was a slight increase in infection when CD4 alone was transfected by DC-SIGN.  However, this increase (equal to a luciferase activity count of about 4000) is small compared to the amount of infection obtained by CD4+CCR5-transfected cells (luciferase activity is well over 10^6).  The figures certainly seem to support the authors’ claims.

Claim 4:  DC-SIGN captures HIV-1 and facilitates infection of HIV-1 permissive cells in trans.  If this is true, it means that HIV-1 binds to DC-SIGN-positive cells while infecting other cells.  They have already shown that HIV-1 binds to DC-SIGN-positive cells, so these data focus on demonstrating that interactions with those cells can cause other cells to become infected with HIV-1.  Since THP-DC-SIGN cells cannot be infected by HIV-1 (data not shown), any infection detected after incubation must therefore occur in the “recipient” cells, and thus the “donor” cells must act as infection facilitators.  In 4A, recipients are infected only when DC-SIGN-positive cells are used, and this infection is blocked by anti-DC-SIGN, offering evidence that DC-SIGN facilitates HIV-1 viral infection.  CD4+ T cells did not facilitate infection.  This was not surprising because although CD4 is a viral receptor, it is not thought to be a carrier like DC-SIGN.  As shown by the clear data from 4B, DC-SIGN can mediate capture of HIV-1 viruses from different primary isolates.  This shows that the particular isolate from which the virus comes is not critical in its binding to DC-SIGN.  In 4C, the authors cocultured virus-pulsed THP-DC-SIGN cells with T cells, and only the CD3+ T cells expressed virus-encoded GFP.  This appears to be correct.
 Figure 5 is very similar to Figure 4 (A and B), except low titers of pseudotyped HIV-1 were cocultured with HIV-1-permissive cells.  Interestingly, the authors did not wash away unbound virus, and they do not explain their rationale.  Perhaps it was to ensure there was enough virus to show up in the assay.  At any rate, their results are very clear.  Efficient HIV-1 infection occurred only when THP cells were transfected with DC-SIGN, suggesting that infection occurs in trans.  Like 4B, 5B shows that HIV-1 infection of primary T cells by DC-SIGN occurs with different virus isolates and R5 envelopes.  The authors’ arguments seem to hold up.

Claim 5:  DC present in mucosal tissues at sites of HIV-1 exposure express DC-SIGN and are CCR5 negative.  This is to show that DC with characteristics necessary for HIV-1 capture and transmission are present in vivo.  Immunohistochemical analyses are used to examine mucosal tissues, the sites of first HIV-1 viral exposure during sexual transmission.  The authors state that all mucosal tissues (cervix, rectum, and uterus) contain DC-SIGN positive cells in the lamina propria.  They also say that staining of serial sections demonstrate that these DC-SIGN-positive cells do not express CD3, CD20, CD14, or CD68, but the data isn’t shown.  Their comparison of expression of DC-SIGN, CD4, and CCR5 on DC in the mucosa of the uterus and rectum seems to show that the majority of DC-SIGN-positive DC in those tissues coexpressed CD4 but lacked CCR5.  This agrees with their model.

Claim 6:  DC-SIGN-bound HIV-1 retains infectivity after long-term culture.  To support this claim, the authors perform experiments whose results are shown in Figure 7.  These data are very clean.  As shown in 7A, when DC-SIGN-positive cells are incubated with gp120-coated beads, binding occurs for more than 60 hours.  In the presence of anti-DC-SIGN, binding was almost completely blocked.  7B demonstrates clearly that THP-DC-SIGN cells could retain infectious virus for a much longer time than the virus itself.  These data lend credibility to the author’s model, as DC-SIGN-positive cells (such as DC) could sequester HIV-1 for later catalyzation of entry into cells expressing CD4 and chemokine receptors.

The authors offer a very credible model for how DC cells containing DC-SIGN function to facilitate the infection of cells expressing CD4 and chemokine receptors.  HIV-1 could bind to DC present in mucosal tissues, exploit the migratory capacity of DC to gain access to lymphoid tissues, and infect those T cells that contain CD4+CCR5.  This proposed mechanism may require the internalization of HIV-1 in DC-SIGN in order to protect it during the time required for the DC cells to complete their journey from mucosal membranes to the lymph nodes.  In order to examine this possibility, a good first question would be, “Can DC-SIGN-positive DC cells internalize HIV-1?”

I would address this question by incubating DC-SIGN-positive DC cells with small amounts of HIV-1 for 24 hours.  (A small amount of virus relative to DC cells would ensure that all virus was internalized, while 24 hours should be ½ the time required for DC to migrate from the periphery to draining lymph nodes—ample enough for internalization to take place.)  I would then introduce an agent that removed all extra-cellular debris and look there for HIV-1.  I would not expect to see any HIV-1 in the debris, as it should all be internalized.  As a control, I would detect for a known extra-cellular protein that could be detected using the same method as employed to find HIV-1.  If HIV-1 were present, the model would have to be adjusted.  Perhaps the virus is carried around extra-cellularly or is integrated into the cellular membrane (where it could be detected).  Although this experiment is based on a negative result, any sort of positive result would shed light on the mechanism (HIV-1 wouldn’t be internalized).  If no HIV-1 were found, I would then use increasing concentrations of HIV-1 in my incubation to see if I could find a saturation point for HIV-1.

Here is a brief dictionary of terms used in the paper which may be unfamiliar to undergraduates.  Most definitions come from the online medical dictionary found at this address: http://www.graylab.ac.uk/omd/index.html

CD4:  55-kd glycoproteins originally defined as differentiation antigens on T-lymphocytes, but also found on other cells including monocytes/macrophages.  CD4 antigens also serve as HIV receptors, binding directly to the envelope protein gp120 on HIV.

chelating agents:  Organic chemicals that form two or more coordination bonds with a central metal ion. Heterocyclic rings are formed with the central metal atom as part of the ring. Some biological systems form metal chelates, e.g., the iron-binding porphyrin group of hemoglobin and the magnesium-binding chlorophyll of plants. They are used chemically to remove ions from solutions, medicinally against microorganisms, to treat metal poisoning, and in chemotherapy protocols.

chemokine:  Cytokines that are chemotactic for leukocytes

chemokine receptor:  A molecule that receives a chemokine and a chemokine dock. Several chemokine receptors are essential co-receptors for HIV. CCR5 is an example of one.

chemotaxis:  A response of motile cells or organisms in which the direction of movement is affected by the gradient of a diffusible substance

cytokines:  Non-antibody proteins secreted by inflammatory leukocytes and some non-leukocytic cells, that act as intercellular mediators. They differ from classical hormones in that they are produced by a number of tissue or cell types rather than by specialized glands.

dentritic cells:  heterogeneous population of cells that are present in minute numbers in various tissues just beneath the dermis or mucosal layer and form a first-line defense against viruses and other pathogens.  DC localized in the skin and mucosal tissues (such as the rectum, uterus, and cervix) have been proposed to play a role in HIV-1 infection.

EGTA (egtazic acid): A chelating agent relatively more specific for calcium and less toxic than edetic acid (edta).  Chemical name: 6,9-Dioxa-3,12-diazatetradecanedioic acid, 3,12-bis(carboxymethyl)

ELISA> enzyme-linked immunoabsorbent assay:  A serologic test used as a general screening tool for the detection of antibodies to the HIV virus. Reported as positive or negative. Since false positive tests due occur (for example recent flu shot), positives will require further evaluation using the western blot. ELISA technology links a measurable enzyme to either an antigen or antibody. In this way, it can then measure the presence of an antibody or an antigen in the bloodstream.

gp120:  A glycoprotein from HIV's envelope that binds to the CD4 molecules on cells outside membrane. Free gp120 in the body may be toxic to cells on its own, causing CD4 cell depletion in the immune system through apoptosis and neurological damage leading to AIDS dementia complex.

ICAM:  Glycoproteins that are present on a wide range of human cells, essential to the mechanism by which cells recognize each other, and thus important in inflammatory responses.

mannan:  contains mannose:

mannose:  A hexose or fermentable monosaccharide and isomer of glucose.
in trans: a genetic term to mean that while the virus binds to one cell, it infects another cell

monocyte:  Mononuclear phagocyte circulating in blood that will later emigrate into tissue and differentiate into a macrophage.

p24:  Protein antigen from HIV's core that can be measured in blood and other body fluids. Measurement of p24 levels in the blood have been used to monitor viral activity, although this is not considered a very accurate method due to the existence of p24 antibody that binds with the antigen and makes it undetectable.

serologic test:  A blood test that detects the presence of antibodies to a particular antigen (for example rheumatoid factor, HIV test).

Back to Molecular Biology Homepage
Back to my main page.
email me at erferguson@davidson.edu