Researchers Dissect the Mechanisms of HIV Infection

Data derived from the Science Watch/Hot Papers database and the Web of Science (ISI, Philadelphia) show that Hot Papers are cited 50 to 100 times more often than the average paper of the same type and age. For all that is known about human immunodeficiency virus (HIV), some remarkably fundamental questions remain. One of the most notable, perhaps, is just how HIV manages to infect its primary target, CD4+-T cells, when so few of those cells can be found at the virus' typical entry points: the v

May 27, 2002
Jeffrey Perkel
Data derived from the Science Watch/Hot Papers database and the Web of Science (ISI, Philadelphia) show that Hot Papers are cited 50 to 100 times more often than the average paper of the same type and age.

For all that is known about human immunodeficiency virus (HIV), some remarkably fundamental questions remain. One of the most notable, perhaps, is just how HIV manages to infect its primary target, CD4+-T cells, when so few of those cells can be found at the virus' typical entry points: the vagina, uterus, cervix, and rectum. It is conceivable that the virus may happen upon a stray macrophage or T cell, but for the most part, these cells are hard to come by.

Courtesy of Yvette van Kooyk

Yvette van Kooyk

Two years ago, Yvette van Kooyk, professor of molecular cell biology at Vrije University Medical Center, Amsterdam, and postdoc Teunis Geijtenbeek published back-to-back reports in Cell about a dendritic cell (DC)-specific receptor that appears to help DCs engage naïve T cells.1-3 But HIV's envelope protein can also bind this receptor, giving the virus a beachhead within the body. Together, these reports suggest a simple, elegant model for both T cell activation and virus infection. But some researchers question whether the final story will be as neat and tidy a package as the model van Kooyk and colleagues put forward.

Dendritic Cell Function

Scientists have long recognized that DCs exhibit two interesting properties. First, as described in the first van Kooyk paper, DCs are "professional antigen-presenting cells." Present in low numbers in the skin and mucosal tissues where they patrol for foreign pathogens, these cells constitute a first line of defense. As such, they possess many sticky fingers—adhesion molecules designed to find all the invaders they can. Once a DC has captured something suspicious, it migrates from the periphery to the draining lymph nodes, where it presents its findings to CD4+-helper T cells. The helper T cells then orchestrate an appropriate immune challenge.

Yet dendritic cells have a dark side, too. For several years, researchers have suspected that the cells are HIV's first victims, and that the virus somehow subverts the same features that make DCs such effective sentinels, thus creating a kind of cellular Trojan horse. In this model, HIV-infected DCs migrate to the lymph nodes and infect the helper T cells, amplifying the infection.

Hijacking Cellular Function

van Kooyk was interested in cellular adhesion receptors; naïve T cells are coated with one called intracellular adhesion molecule 3 (ICAM3). Since DCs interact with ICAM3-bearing T cells, van Kyook wanted to determine which DC receptor engages this ligand. Antibodies against ICAM3's primary receptors, the b2-integrins LFA-1 and kDb2, failed to block DC-ICAM3 interactions. Instead, ICAM3's partner on the DC was something new, which the researchers dubbed DC-SIGN, for dendritic cell-specific, ICAM3-grabbing nonintegrin. (van Kooyk says her lab has since shown that ICAM2 is also a DC-SIGN ligand.) At the time, DC-SIGN was the only DC-specific receptor known.

Courtesy of Dan Littman

Dan Littman

Yet DC-SIGN was not new. Sequence analysis indicated that DC-SIGN matched a 44-kDa C-type lectin, or carbohydrate-binding protein, which Bristol-Myers Squibb had cloned from a placental cDNA library eight years earlier because it could bind gp120, HIV's viral envelope glycoprotein.4 "That was the moment when we made the link" between DC-SIGN and HIV, says van Kooyk. The Dutch re-searchers, who are not virologists, made an initial foray into the virology, but negative reviews of their manuscript prompted them to seek out an HIV expert; they chose Dan Littman, Howard Hughes Medical Institute investigator, Helen L. and Martin S. Kimmel Professor of Molecular Immunology, and professor of pathology and microbiology at the Skirball Institute of Biomolecular Medicine, New York University Medical Center.

With Littman's help, Geijtenbeek demonstrated that gp120 binds DC-SIGN with high affinity. Despite the fact that these cells also express the primary HIV receptors, CD4 and CCR5, the cell is never actually infected, say van Kooyk and Littman. They suggest that the cell acts like a viral taxi, taking HIV to a T cell-rich lymph node. This has two consequences. First, it amplifies an otherwise weak HIV inoculation, and during a natural exposure to virus, pathogen titers are generally low. Second, by sequestering the virus, DCs keep HIV stable during the trek to the lymph nodes.

Though extrapolating from in vitro culture conditions to the vastly more complex conditions inside the body is a tricky business, van Kooyk and Littman offer a compelling theory that addresses a number of previously unanswered questions. But researchers are divided over whether it is actually correct. "The model ..., I think, is going to be overly simplistic," says DC-SIGN researcher Robert Doms, chair, department of microbiology, University of Pennsylvania. Key to Doms' complaint is a technical—but critical—issue: just what is a true dendritic cell?

Courtesy of Yvette van Kooyk

Cellular Trojan Horse: In the van Kooyk-Littman model of HIV-1 infection, virus-bearing dendritic cells migrate to the T-cell-rich draining lymph nodes, where they can infect CD4+-T helper cells. (Reprinted with permission from Elsevier Science, © 2000.)

Not All Dendritic Cells Are Alike

Researchers debate whether HIV does, or does not, infect dendritic cells. According to Littman, the DC "is basically used ... as a way for the virus to hitch a ride to the draining lymph nodes." But Doms counters that at a recent meeting on DCs, "Most people I talked to said, 'well yeah, they are infected by HIV,' and then some people say, 'well, they're not.'"

The problem, Doms explains, is that, like macrophages and T cells, "there are probably lots of different flavors" of dendritic cells. So, whereas some DCs, like the type van Kooyk used, cannot be infected by HIV, others can be. Because they are so rare, researchers typically produce DCs in the lab by culturing monocytes with interleukin (IL)-4 and granulocyte-monocyte colony-stimulating factor (GM-CSF). The resulting cells have "dendritic cell-like properties," says Doms. But how similar these monocyte-derived DCs are to the ones patrolling the front lines of the human body is anybody's guess. There is no doubt that DCs found in vivo express DC-SIGN. But whether this particular molecule is the key HIV receptor is another issue.

The problem may boil down to the particular complement of adhesion molecules present on a given DC. Evidence from Doms, Vineet KewalRamani at the National Cancer Institute, and Anthony Cunningham at the Westmead Millennium Institute in Sydney, Australia, suggests that DC-SIGN is not the only adhesion molecule on a DC that can capture HIV.5,6 "There are other players, and those other players could be quite important," says Doms.

Cunningham, for example, demonstrated that at least three receptors, DC-SIGN among them, can bind gp120 onto in vitro cultured DCs. However, ex vivo DCs do not appear to express DC-SIGN; instead, they bind gp120 through CD4.6

Doms wonders whether it is reasonable to assume that dendritic cells always mediate HIV infection. It is likely, he says, that sometimes the virus will encounter a DC, and other times, a T cell or macrophage. "I suspect it's at some level a stochastic process," he says. In fact, researchers discovered an endothelial cell-specific DC-SIGN homolog, called DC-SIGNR, not long after the publication of the DC-SIGN data.7 Like DC-SIGN, DC-SIGNR appears to enhance trans-infection of T cells.8

Routes of Infection

Littman poses a question: Is this DC-SIGN-mediated process an alternative or replacement for infection by naked virus in the bloodstream? Littman deems the latter possibility unlikely. "It's not clear if virus in the bloodstream ever infects T cells directly, and that's maybe the one thing that one needs to question at this point." Instead, he says, the virus is internalized and kept viable inside DCs. Recently, Littman's lab reported that DCs must internalize HIV into a low-pH nonlysosomal compartment to efficiently capture virus and trans-infect T cells.9

Doms poses another question: "Might other pathogens have evolved to take advantage of the adhesive and cell-migratory properties of dendritic cells to assist in their transmission or dissemination with a host?" His lab, for example, has shown that DC-SIGN can bind Ebola virus "wonderfully," and greatly increase its infection efficiency.

van Kooyk's work has caused researchers to reexamine the role dendritic cells play in viral life cycles, says Doms. If DC-SIGN is a critical HIV receptor in vivo, then it becomes an attractive therapeutic target because it functions so early in the infection. He adds, "I think DC-SIGN serves as a really great example of what can happen when you have an efficient virus-attachment factor, [and] what impact that might have on virus tropism in pathogenesis."

Jeffrey M. Perkel can be contacted at

1. T.B.H. Geijtenbeek et al., "Identification of DC-SIGN, a novel dendritic cell-specific ICAM-3 receptor that supports primary immune responses," Cell, 100:575-85, March 3, 2000. (Cited in 108 papers)

2. T.B.H. Geijtenbeek et al., "DC-SIGN, a dendritic cell-specific HIV-1-binding protein that enhances trans-infection of T cells," Cell, 100:587-97, March 3, 2000. (Cited in 126 papers)

3. 7. D. Steinberg, "Receptor boosts HIV infection," The Scientist, 14[6]:12, March 20, 2000.

4. B.M. Curtis et al., "Sequence and expression of a membrane-associated C-type lectin that exhibits CD4-independent binding of human immunodeficiency virus envelope glycoprotein gp120," Proceedings of the National Academy of Sciences (PNAS), 89:8356-60, 1992.

5. L. Wu et al., "Rhesus macaque dendritic cells efficiently transmit primate lentiviruses independently of DC-SIGN," PNAS, 99:1568-73, Feb. 5, 2002.

6. S.G. Turville et al., "HIV gp120 receptors on human dendritic cells," Blood, 98:2482-8, Oct. 15, 2001.

7. E.J. Soilleux et al., "Cutting edge: DC-SIGN; a related gene, DC-SIGNR; and CD23 form a cluster on 19p13," Journal of Immunology, 165:2937-42, 2000.

8. S. Pohlmann et al., "DC-SIGNR, a DC-SIGN homologue expressed in endothelial cells, binds to human and simian immunodeficiency viruses and activates infection in trans," PNAS, 98:2670-5, 2001.

9. D.S. Kwon et al., "DC-SIGN-mediated internalization of HIV is required for trans-enhancement of T cell infection," Immunity, 16:135-44, January 2002.