Early Damage to the Gut Environment B Cells Need to Mature
While T-cell depletion in the gut-associated lymphoid tissue and in blood has been well documented, not much is known about the effect of early HIV infection on B cells. A study has now taken a closer look, and found a possible explanation as to why the body produces HIV-specific neutralizing antibodies much later after infection than in response to other pathogens (1).
Researchers found that just a few weeks after HIV infection, more B cells have matured into antibody-producing memory B cells in blood and in the gut of HIV-infected individuals than in uninfected people. These B cells are often not HIV specific, and instead include influenza and autoreactive antibodies. As a result, HIV-infected people have fewer naive B cells available to mature into B cells that produce HIV-specific antibodies. To make matters worse, the study also found that just weeks after HIV infection, the environment in the gut that could induce the maturation of naive B cells into cells that produce HIV-specific antibodies is damaged.
“People have identified damage to the T-cell arm of the immune system,” says Anthony Moody, an assistant professor of pediatrics at Duke University and one of the lead authors of the study. “I think our paper added that damage is also occurring to the environment that supports the B cells. It adds another face onto the picture of acute HIV infection.”
The researchers analyzed B cells and their environment from biopsies of the terminal ilium section of the gut as early as 47 days after HIV infection. The terminal ilium is the last part of the small intestine and contains a large amount of lymphoid tissue. They found that HIV-infected people had damage to the gut follicles where B cells collect and mature to become antibody-producing cells. This damage included cell death and disruption of the arrangement of dendritic cells, which present antigen to the B cells and support the different cell types in the follicle.
In addition, fewer B cells in the gut follicles were dividing in HIV-infected people, although there did not seem to be fewer B cells overall in the gut than in uninfected people. “This data is suggesting that the environment of the B cells to mature into antibody secreting cells is being damaged,” says Moody. “That may be part of the reason why patients infected with HIV don’t make a good antibody response to HIV.” However, what leads to the damage is still an open question. “It’s a bit like looking at a car wreck and saying, I know that a wreck has occurred, but exactly how did it happen?”
Researchers also found that the terminal ilium of HIV-infected people also contained fewer naive B cells, and more memory and plasma B cells as a percentage of all B cells. The same was observed in blood samples taken as early as 17 days after infection. However, many of these memory and plasma B cells were not HIV specific, and instead included antibodies to the body’s own antibodies usually seen in patients with rheumatoid arthritis. “They are just making antibodies to everything,” Moody says. “It’s not a response the body generally would want to make. Any specific HIV response is probably being drowned out by all of the non-specific responses that are being driven through.”
One possible cause for this early unspecific B cell maturation, Moody says, might be an early production of cytokines in HIV-infected people such as interleukin 15 and interferon-α. This might be the earliest sign of the chronic inflammation associated with HIV infection that eventually leads to AIDS.
Taken together, these observations could explain why HIV-infected people only show neutralizing antibodies to HIV 40 or more days after infection, much longer than the week it often takes for the body to mount such a specific response to other pathogens, Moody says.
“While B cell dysfunction is known to be a feature of chronic HIV infection, this is the first study to demonstrate B cell pathology in gut tissues and peripheral blood in acute and early HIV infection,” says Savita Pahwa, a professor at the University of Miami School of Medicine who was not involved in the study. “These findings highlight the need for a vaccine that at the very least has to act early enough so that it can prevent the virus from causing the damage to the immune system that ensues rapidly after HIV infection.”
Next, Moody plans to study the bone marrow of HIV-infected people to see if there is any damage to the generative environment of B cells. “The bone marrow is where [B] cells originate, and if there is damage, that would be another reason why people may have damage to their immune systems by HIV,” he says. —Andreas von Bubnoff
Lower Antibody Levels Can Protect from Low-Dose Challenge
Previous studies have found that high serum concentrations of broadly neutralizing antibodies are required to completely protect rhesus macaques from a high-dose intravenous challenge. But according to a recent study, much lower concentrations of the broadly neutralizing antibody b12 can significantly lower the risk of infection in rhesus macaques exposed to a repeat low-dose vaginal challenge regimen with a simian immunodeficiency virus (SIV)/HIV hybrid virus (1). Such a regimen is considered more similar to HIV transmission than a single high-dose challenge.
This is promising news for vaccine researchers as it suggests that an HIV vaccine would have to induce lower levels of antibodies than previously thought to significantly decrease the risk of infection.
Researchers infused rhesus macaques once a week with a dose of 1 mg/kg of the broadly neutralizing antibody b12, and then challenged them vaginally twice a week with a low dose of SHIV162P3. This dose of the challenge virus was at least 30-times lower than what is usually used in high-dose intravenous challenge studies in macaques. At this dose, it took 10 challenges to infect four macaques that were untreated or treated with a mock antibody, while it took 108 challenges to infect the same number of b12-treated macaques.
Before the new study, it was perceived that the levels of elicited neutralizing antibodies needed to protect against HIV challenge may be unachievable, says Ann Hessell, a senior research assistant at The Scripps Research Institute and first author of the study. For example, one previous study used a 25-fold higher dose of b12 antibody than what was used in the new study to protect all macaques from an intravenous high-dose challenge with a different challenge virus (2). The fact that the two studies used different challenge viruses makes it difficult to compare the antibody doses, Hessell says. Still, the new study suggests that a much lower dose of antibody can decrease the risk of infection from a repeat low-dose challenge than the antibody dose needed to protect from a high-dose challenge.
“It has often been said that it will take supraphysiologic levels of neutralizing antibodies to protect against HIV-1 infection,” says John Mascola of the Vaccine Research Center at the National Institute of Allergy and Infectious Diseases, who was not involved in the study. “These data support a more encouraging model suggesting that modest levels of vaccine-elicited neutralizing antibodies could have a major impact on acquisition of HIV-1 infection.”
“If you had this amount of antibody on board [in humans], then you could decrease the risk of becoming infected, and that’s a benefit,” says Dennis Burton, a professor at The Scripps Research Institute, who led the study. “[But] this is a model and it’s one antibody and it’s one virus and so it should be taken very carefully.” —Andreas von Bubnoff
Among the handful of broadly neutralizing antibodies to HIV that have so far been isolated, 2G12 is unique because it recognizes the sugar coat of HIV, as opposed to the viral proteins. A study in 2000 showed 2G12 is also exceptional because even though it neutralizes virus relatively poorly in vitro compared with other broadly neutralizing antibodies, it protects animals similarly well (1).
In a recent study, Ann Hessell, a senior research assistant at The Scripps Research Institute, and colleagues confirmed these earlier findings that 2G12 could protect rhesus macaques from infection better than what would be expected from its poor neutralization ability in vitro (2). 2G12 completely protected three out of five rhesus macaques from high-dose vaginal challenge with the hybrid simian immunodeficiency virus (SIV)/HIV challenge strain SHIV162P3.
2G12 protected at about twice the dose of the b12 antibody dose that had previously been shown to protect all animals from a similar high-dose challenge (3). While it is difficult to compare the antibody doses in the two studies because they used a different challenge virus, this suggests that 2G12 can protect from high dose challenge at a similar dose as what was previously shown for the b12 antibody. However, the new study also showed that 2G12 fared about 100-times worse than b12 at neutralizing the challenge virus in vitro.
“Generally we tend to think that neutralization correlates with protection,” says Dennis Burton, a professor at The Scripps Research Institute, who led the study. “[But 2G12] boxes above its weight in the sense that it protects [animals] at relatively low neutralization titers.”
John Mascola, of the Vaccine Research Center at the National Institute of Allergy and Infectious Diseases, who led the 2000 Nature Medicine study, says the new 2G12 study is “more definitive and clearly shows that relatively modest levels of 2G12 have a major in vivo impact.”
However, just why 2G12 does so much better in vivo than what would have been expected from its in vitroneutralization ability is still unclear. The new research ruled out the possibility that 2G12 may be better transported from blood to mucosal tissues, such as those in the vagina, than other antibodies like b12.
Another explanation could be that there are many biological components that are present in vivo but not in vitro, Hessell says. “In vitro you don’t have all of the mechanisms of the human immune system. You don’t have other cells and things like complement that can be brought in,” she says. “In vivo you bring in these other aspects that possibly influence the mechanism by which the antibody is able to protect.”
Whatever the case, there might be something advantageous to antibodies like 2G12 that are directed against the sugar coat of HIV as opposed to protein targets, Burton says. Perhaps 2G12 acts especially strongly on HIV transmission in that it inhibits interactions between the sugars on the HIV surface and proteins called lectins on host cells that pick up the virus during transmission, according to Burton.
Hessell is also the first author of another single high-dose challenge study submitted by Burton’s lab that came to similar, although less dramatic, conclusions for the broadly neutralizing antibodies 2F5 and 4E10, suggesting that these two antibodies are also better at in vivo protection than what would be expected from their in vitroneutralization ability. —Andreas von Bubnoff
Two Studies Add to Understanding of HIV Assembly
The exact way HIV assembles during and after budding from an infected cell is still not completely understood, but two recent studies give more detailed insight into this process, which might eventually lead to the development of new drugs that inhibit HIV assembly.
One study shows, at a higher resolution than previously reported, how full-length Gag proteins assemble to form a honeycomb-shaped lattice of hexamers under the plasma membrane of immature HIV particles as they are released from the host cell (1; see step A in Fig. 2). The study used a method called cryoelectron tomography, where HIV samples are frozen extremely quickly and then analyzed by electron microscopy. This avoids the formation of ice crystals that could destroy biological structures.
Figure 2. Simplified Representation of HIV's Life Cycle (counter-clockwise). Image courtesy of John Briggs/European Molecular Biology Laboratory.
John Briggs, a researcher at the European Molecular Biology Laboratory in Heidelberg, Germany, and lead author of the study, says it was already known from previous studies that the spheres formed by the lattice are incomplete, with sometimes as much as one third missing, similar to an egg with its top cut off (2,3).
But it was unclear how the continuous part of the honeycomb-shaped Gag lattice could be curved because a lattice of hexamers should be flat. “We always knew that [the lattice] had to have something other than hexamers going in there because otherwise [it] couldn’t be a curved structure,” Briggs says. “The question was, what are those things?” One possibility, he says, was that it contained pentamers just like a soccer ball, although it was also considered possible that the lattice contains holes.
Researchers found evidence for holes and not pentamers (see Fig. 3). “This, I think, is the first time that it has been shown convincingly that those are irregular shaped defects rather than anything regular,” says Briggs.
Figure 3. Lattice Maps for Immature HIV Particles. The 3D computer reconstruction shows the immature Gag lattice of HIV that matures to form the protein shell of the infectious virus. Maps are shown in perspective such that hexamers on the rear surface of the particle appear smaller. The side of the particle toward the viewer lacks ordered Gag. Image courtesy of John Briggs/European Molecular Biology Laboratory.
Elizabeth Wright, an assistant professor at Emory University School of Medicine and the first author of the 2007EMBO Journal study, says the new study complements previous studies in its demonstration that immature HIV virions and immature in vitro assembled particles are not composed of a completely enclosed protein shell. In addition, she says, it “unequivocally [illustrates] that the hexagonal lattice appears to be a continuous sheet into which irregular defects are included to generate the curvature necessary to make an enclosed particle.”
Another recent study took a detailed look at the cone-shaped capsid of mature HIV particles, which forms at a later stage in the HIV life cycle and encloses the genetic material of the virus (4; see step B in Fig. 2). The study, led by Mark Yeager of The Scripps Research Institute and the University of Virginia, is the first X-ray crystallographic analysis of the capsid protein CA, which is generated when HIV protease cleaves Gag in the immature Gag protein lattice studied by Briggs. The CA proteins then form the cone-shaped capsid by assembling as about 250 hexamers and 12 pentamers.
The CA proteins arrange in a noncovalent way to form the hexamers and pentamers, but exactly how they are arranged has been unclear. This was in part because the capsid of each HIV particle is slightly different, making it impossible to grow crystals of intact capsids to determine the structure at high resolution using X-ray crystallography, Yeager says. In addition, the CA protein itself has two domains that are connected with a flexible linker that impedes growth of protein crystals.
So the researchers used a trick. Guided by a model of the hexameric CA lattice based on a lower resolution structure determined using electron cryomicroscopy, they engineered CA hexamers that were stable enough to grow crystals. This enabled the determination of the X-ray structure of the CA hexamer at an unprecedented atomic resolution of two Ångstrom (see cover image).
The study for the first time identifies the atomic structure of the so-called NTD-CTD interface that holds the N-terminal and C-terminal domains, or opposite ends, of adjacent CA proteins in the hexamers together. It shows that water molecules make the surface of the interacting proteins somewhat slippery. This, Yeager says, could explain how the CA protein is flexible enough to both arrange in the form of hexamers as well as pentamers.
“We had little clue about the molecular details of the NTD-CTD interface,” says Ian Taylor of the National Institute of Medical Research in London, who was not involved in the study. “This paper now provides these details and furthermore confirms the idea that interdomain flexibility is the driver of shell curvature.” Knowing the NTD-CTD interface could help develop drugs that inhibit assembly and maturation of HIV. —Andreas von Bubnoff
Public Database of Viral Vectors Released
CompuVac, a consortium of 140 scientists from 11 countries that was established in 2005, held a day-long symposium on June 29 at the Collège de France in Paris to commemorate the public release of an interactive database and bioinformatics tool that allows researchers to compare different viral vectors and virus-like particles (VLPs) to predict the best potential immunization strategy for the development of novel vaccines. This database was the culmination of an €8 million grant from the European Union under its Sixth Framework Programme.
CompuVac was created to help standardize the evaluation of viral vectors and aid the rational development of novel recombinant vaccines. “Many novel technologies for development of new vaccines exist but their design or improvement profoundly lacks a reliable evaluation system,” said David Klatzmann, CompuVac’s coordinator. This is because it is difficult to compare results from different studies because researchers often use different antigens or methods to evaluate vaccine vectors or VLPs.
Cedrik Britten, a T-cell immunologist from Johannes Gutenberg University in Germany, spoke at the symposium about efforts to encourage standardization in cancer research. He is pioneering an effort that would require any publication of immunology data to provide an explanation of methods. Although this “reporting standard,” as he called it, falls short of requiring use of standardized assays, it is a first step. Britten joked that a legal body would be required to realistically impose standardization because researchers are so unwilling to accept another’s protocol, but then cited an example involving a Grand Challenges in Global Health grant from the Bill & Melinda Gates Foundation that successfully promoted standardization. This grant provided free reagents to researchers as a way to motivate them to use the same reagents.
Peter Piot, chairman of the board of the Global HIV Vaccine Enterprise, who also spoke at the symposium, said that the biggest resistance to standardization and sharing data comes from academic researchers, and not industry. He said that many of the “organizational challenges” in the AIDS vaccine field are “as big as the scientific ones.”
In an effort to standardize evaluation, CompuVac set out to study several vaccine vectors, including adenovirus, herpes simplex virus-1, measles virus, modified vaccinia Ankara, Bacillus Calmette-Guérin, and DNA, as well as many VLPs, using standardized antigens—the GP33-41 epitope from lymphocytic choriomeningitis virus to evaluate T-cell responses and the envelope protein from vesicular stomatitis virus to evaluate B-cell responses. In all, CompuVac researchers fully assessed the immunogenicity of 51 different vectors with these so-called “gold-standard” antigens, based on their ability to induce T- and B-cell responses, as well as their ability to protect against an infectious challenge in mice. CompuVac researchers also assessed the gene-expression profiles induced by the different vectors and VLPs expressing the standard antigens. The results of these assessments are available in the Genetic Vaccine Decision Support system (GeVaDSs), the online database created by the consortium that stores this data and allows users to create associations and develop algorithms to compare new vaccine vectors or VLPs to those previously analyzed (available at www.compuvac.org). The premise of GeVaDSs is that researchers can then predict, based on the data sets for each of these single vectors, the best homologous or heterologous prime-boost combinations to evaluate further. The GeVaDSs system is what Klatzmann refers to as a “systems vaccinology approach.” —Kristen Jill Kresge