February 8, 2007
About the Palm Project Interviews
TAG has conducted a series of interviews with leading scientists about the underlying pathogenesis of AIDS to gain insights into emerging lines of research and observations about the current research funding environment.
Barbara Shacklett began studying HIV in the late 1980s and has become a highly respected expert in the under-studied area of mucosal immunity. Shacklett’s laboratory at UC Davis has recently published novel findings regarding the role of mucosal immune responses in individuals who control HIV replication in the absence of therapy. Barbara Shacklett generously took time out to discuss her research with TAG’s BSVP coordinator, Richard Jefferys.
Gut Immune Responses in Elite Controllers
Q. Can you describe how you became interested in the elite control phenomenon and your recent research findings?
We were studying two or three individuals here in Sacramento – not quite elite controllers but with very low viral loads – whom we had biopsied as part of our ongoing studies on mucosal immunity. We had taken rectal biopsies from these individuals, as well as peripheral blood, and we found that they had really strikingly high CD8 responses to stimulation with HIV peptides in the rectal mucosa; much higher and more polyfunctional than in peripheral blood. So we wondered if this was a phenomenon that would be generalizable, if not to all elite controllers then at least a subset, and if it could perhaps be considered a correlate of non-progression. I started talking with Steve Deeks at a Keystone meeting around three years ago and we then went on to collaborate and obtain IRB approval for a study in which we were allowed to obtain rectal biopsies from elite controllers.
The definition of elite controllers that we are using is very similar to Bruce Walker’s: untreated individuals who have viral loads below 50 or 75 copies/mL and have maintained this status for at least two years, with three separate blood samples confirming. And in this study we also included what we’re calling viremic controllers. These are individuals whose viral loads are between the detectable limit and 2,000 copies/mL. We enrolled 17 in the elite group and 11 in the viremic group and compared them to what we call “non controllers” with viral loads of 10,000 or more.
What were the results?
What we found is that the elite controllers, as a group, have very strong CD8 responses and CD4 responses to HIV peptides in mucosal tissues. And I should make sure that it’s clear that I’m talking about rectal mucosa, as there are differences between the upper GI tract and lower GI tract, and the GI tract and the other mucosal tissues, so those of us that do mucosal work have to be very careful to specify which compartment we’re talking about!
As far as CD4 count, we found that the rectal mucosa seems not to be as depleted of CD4 cells as in non-controllers. There was a statistically significant difference: the elite and viremic controllers tended to retain CD4 cells in rectal mucosa whereas the non-controllers did not.
Something else that we found out in this process came from another control group of individuals on long-term (> 5 yrs) ART. There have been several studies about the lack of adequate gut CD4 repopulation on ART in the jejunum and terminal ileum. Well, it turns out that in the rectum this is not so much the case, and we see very similar CD4 percentages in rectal mucosa and blood and those tend to be relatively high – not within the normal range but on the order of 40-50% of normal, and in some cases even higher – in people on long-term ART. So that’s a difference between rectal mucosa and jejunum, it does seem that there is better repopulation of CD4 cells. That’s another one of the reasons why it’s important to clarify which part of the mucosa you’re talking about.
Looking at mucosal CD8 responses in controllers, our study picked up on the theme of previous work by Mike Betts. He published a paper in the journal Blood in 2006, looking at long-term non-progressors (LTNP) and finding that LTNP had stronger and more polyfunctional CD8 responses in blood compared to progressors. It turns out we didn’t see the same thing in blood, in fact we see a very broad range of CD8 responses amongst our elite controllers. Some have very low and almost undetectable responses in blood and others have very high responses, but the median is not significantly different from what we see in our non-controllers. And I think the distinction may be that we’re looking at a very well-defined a group, a group that’s virologically defined as having low or undetectable viral load, so the criteria that are used are important (see glossary for more information on the differences between LTNP and elite controllers). Also, the make-up of the cohort in terms of MHC class I alleles, in terms of genetics, may be slightly different as well, just due to the individuals we have selected.
But what we did find is a statistically significantly higher CD8 response in the rectal mucosa of our controllers compared to non-controllers and individuals on long-term ART. In terms of the polyfunctionality, we used a flow cytometry panel which is essentially the VRC panel that has been looked at by a number of people now. This is TNFa, IFNg, MIP-1beta, IL-2 and CD107. We see a preponderance of largely 4-function, not 5-function CD8 cells because the fifth function is IL-2, which is not produced at high levels by gut CD8 cells to begin with. But what we do see is a high magnitude of 4-function CD8 cells in the gut. Taken together, these findings suggest that there is an active CD8 T-cell response in rectal mucosa that may be contributing to immune control. Although one can always make the opposite argument, that these individuals may have an intact immune system because they’re elite controllers. In other words, this could be more an effect rather than a cause of elite controller status.
What kind of breadth of response do you see?
The breadth does not appear to be significantly different between groups. We’ve done extensive epitope mapping to Gag, Env and Nef regions. There are a few differences; we tend to see a greater breadth of Gag-specific responses in rectal mucosa compared to blood, so that’s interesting. The Gag responses appear to be important, Christian Brander is finding that as well, but we don’t see a significant difference between our controllers and non-controllers just in terms of breadth. But we do see a difference in terms of magnitude and polyfunctionality.
And what about CD4 responses?
This is interesting, too. The CD4 responses are low, even in most of our elite controllers. But there are just a few individuals who have overwhelmingly high CD4 responses in the gut, something on the order of 6-12% of CD4 cells producing cytokines. But that’s all non-overlapping functions combined, and when that’s broken down into different functional categories the individual categories are obviously lower. But it’s only five or six individuals who have responses like that, the rest of them are much more what one would be predict, but what’s interesting for us is that they have any CD4 responses to HIV at all in the gut, where typically CD4 cells are depleted in HIV infection. Just that observation by itself is quite striking and is significantly different in the controllers from the non-controllers.
Is there anything else that has emerged from this study?
Yes, there are a couple of things. We’re seeing relevant differences related to the MHC class I and II alleles that have been identified by others as potentially associated with non-progression. So controllers with “protective” class I alleles including B*57 and B*27, and a few others that are less well known such as B*58, B*81, B*13, tend to have higher CD8 responses than individuals in the controller group who lack those alleles. That’s actually a statistical trend, there is one outlier and the p value is only significant when that individual is not included in the analysis. So these protective alleles are important, although they’re not the whole story.
And with the MHC class II alleles and CD4 responses there’s something very interesting, which has been alluded to by other groups but hasn’t been emphasized to the level that the B*57 relationship has. There is a potentially protective class II haplotype involving HLA DR*13 and DQ*6, this was reported by group at the University of Washington as well as a few others. We’re finding that we have a very surprising number amongst our controllers, 6 out of 28 have the DR*13 and DQ*6 haplotype and another 11 have DQ*6 alone. Interestingly, many of the individuals who have this haplotype for class II also have the protective class I alleles so there may be some interaction between the class I and class II so-called “protective” alleles.
This DR*13/DQ*6 relationship was described in a paper by Uma Malhotra in the Journal of Clinical Investigation a few years back and there have been two or three other references to this in the literature, but it hasn’t received much attention and we were actually quite surprised to see the association. If you do the polyfunctionality analysis and break our patients down into groups based on the presence or absence of this haplotype, you see that the polyfunctional CD4 responses in the gut tend to be higher in the individuals who possess it. And when you think about the tie-in with the protective class I alleles it could suggest that the relationship between helper (CD4) responses and CD8 responses may be important. We are still in the process of mapping CD4 responses in our controllers, and we hope to have more information on the MHC class II-restricted responses very soon.
Do you think it might be an idea to pay a little more attention to the role of class II MHC?
I think so, yes. One of the reasons people haven’t is that the CD4 responses tend to be of low magnitude and hard to measure – in our case we have a high background in rectal mucosa of cytokine production by CD4s and so we have to use statistical tests to sort out the background from the real antigen-specific responses. In addition, people have been able to rely on MHC class I tetramers for CD8 responses while class II tetramers are hard to get hold of, expensive, and don’t work all that well. And the actual responses are a bit tricky to map as well, the peptides tend to be longer and the genetics complex so people have tended to steer away from that a little bit, but it is an important area to look at.
Do you think it might be productive to put some effort into improving those reagents?
Yes, absolutely, and making them more readily available, too.
Are there any elite controllers where you don’t see any HIV-specific responses?
Well this is the really interesting thing, we have one famous patient who has shuttled back and forth between the West Coast and East Coast and is also a part of Bruce Walker’s cohort. She’s a good example of somebody who, as many times as we have biopsied her – two or three times now – we’ve never been able to see a really strong antigen-specific CD4 or CD8 response and neither has Bruce Walker. We’ve tried recently to isolate virus from PBMC and gut and have not had success with that. I think this does bring up the point that elite controllers may not be a homogenous group and there are certainly suggestions from people like this that some people may be infected with attenuated virus or they may have genetic factors that prevent their HIV from replicating to high levels that are completely outside the purview of classical immune responses. I like to cite the paper in Science, the siRNA paper by Brass et al, all the HIV dependency factors that were identified as areas of potential further study. Because really we don’t know enough the various proteins at the cellular level that interact with HIV as it’s entering, uncoating, reverse transcribing and then finally reproducing and budding out. Many of the cellular proteins that interact with HIV throughout that whole process may indeed have genetic polymorphisms that can limit virus replication independently of the immune system, so those are going to be very important to study.
What else are you looking at in your cohort?
Because of these strong responses in rectal mucosa, it makes you wonder if there is a low level of percolating virus that’s in the mucosal tissue being kept under local control and not really making it out into plasma. So that’s something we really need to study in more detail. We have a long-term project with Steve Deeks to obtain more biopsies from the elite controllers and a collaboration with Peter Anton at UCLA who has mastered the art of measuring virus in tiny rectal biopsies. So we’re in the process of looking at viral load in the mucosa to see whether there may be foci of viral replication that are being kept under local immunological control. We’re also interested in the role of innate immunity, natural killer cells, I think Mary Carrington’s work on KIR genes and interactions between KIR gene products and MHC class I molecules will be very important to investigate at mucosal surfaces. Also T-regulatory (Treg) cells, we’re looking at Tregs in the gut.
But the elite controller studies are our main story right now, the importance of doing these studies of mucosal immunity and also the importance of being very precise when investigating mucosal immunity in terms of what location is being looked at.
In terms of your readership, what I’d especially like to express is the overwhelming gratitude I feel toward everybody that comes in for our studies, it involves taking time off work and it’s not the most pleasant thing to be biopsied multiple times, I have so much respect for our volunteers.
I’d also like to put in a plug for the elite controller study, I think Bruce Walker is making a wonderful effort to raise the awareness of everyone in the HIV research community and in the HIV-affected community regarding the importance of studying elite controllers. I’d encourage everyone in either the elite controller or viremic controller group to talk to their clinician about contributing a blood sample to that repository.
CD4 T-cells, or helper T-cells: These T cells are present in blood and tissues and are the major cells to become infected by HIV. Their role in the immune system is to provide “help” to other types of cells such as macrophages and B-cells, in fighting infections. The CD4 receptor present on the surface of CD4 T-cells is the major receptor for the HIV envelope glycoprotein.
CD8 T-cells, or cytotoxic T-cells (CTL): These T cells fight viral and bacterial infections by identifying and killing the infected cells. By eliminating infected cells, CD8 T-cells can reduce the number of viruses being produced and limit, if not fully eradicate, the infection.
Long-Term Non-Progressors (LTNPs): Definitions can vary somewhat from study to study, but LTNP are typically defined as individuals who maintain CD4 T cell counts in the normal range and show no clinical signs of disease progression despite long-term (e.g. >10-15 years) HIV infection. Not all LTNP maintain undetectable viral loads and therefore not all LTNP fit the definition of elite controller. Conversely, while many elite controllers also fit the definition of LTNP, some do not (due to experiencing CD4 T cell declines and clinical progression).
Polyfunctional: A term used to describe the ability of T cells to perform more than one function, such as production of multiple cytokines (e.g. interleukin-2, interferon-gamma and TNF-alpha) and chemokines, and release of cytolytic granules.
Flow cytometry: A technique for studying cells that allows different subsets of cells to be sorted and analyzed based on the use of lasers to excite fluorescent dyes. Researchers use these dyes as markers to “tag” proteins expressed by certain cells of the immune system (for example, the CD4 and CD8 markers expressed by T cells). This allows researchers to identify, count and sort any subset of cells that they are interested in. Flow cytometers are machines designed to identify and count cells that have bound fluorescent marker dyes. The number of lasers that the machine has governs how many different subsets of cells can be simultaneously analyzed.
VRC: Vaccine Research Center. A center at the National Institutes of Health created by Bill Clinton’s administration. The VRC possesses high-performance flow cytometers capable of analyzing multiple cell functions and researchers at VRC have utilized this resource to pioneer the study of polyfunctional T cell responses in HIV infection.
MHC: The Major Histocompatibility Complex, which is an array of genes that encode cell surface molecules that allow T cells to distinguish “self” from “non-self”. MHC molecules are also used by the immune system to “present” pieces of foreign pathogens (individual peptides) to T-cells. This process, known as antigen presentation, is critical for the development of T-cell immune responses that fight HIV and other viral infections. Class I MHC proteins present peptides to CD8 T cells and class II MHC genes present peptides to CD4 T cells. In humans, MHC is also called the Human Leukocyte Antigen (HLA) system. These genes are the most variable in the human genome. Certain MHC/HLA variants are particularly good at presenting important HIV peptides to the immune system, and people with these variants (examples include HLA B*57 and HLA B*27) are more likely to become elite controllers than people who lack these genes.
T cell receptor (TCR): All T-cells (including both CD4 and CD8 T cells) have a unique molecule on their surface, called the T-cell receptor, that allows them to recognize and bind to short peptides from pathogens such as viruses or bacteria. Some T-cell receptors can recognize peptides from HIV proteins. During antigen presentation (see MHC), the T-cell receptor recognizes an MHC class I or II molecule bound to a peptide from a virus, bacterium or other pathogen.
Tetramers: These are laboratory reagents that allow T cells to be measured based on the specific part of HIV (or other antigen) that the T cell receptor recognizes. They are called tetramers because they are composed of four MHC class I (or II) molecules that have each been folded around a peptide from HIV (or another pathogen). The four molecules are then chemically linked to one another and labeled with a fluorescent marker dye. The marker dye allows T cells that bind the tetramers to be detected in a flow cytometer. Due to differences in structure, MHC class I tetramers for measuring CD8 T cell responses are much easier to make and use than MHC class II tetramers for measuring CD4 T cell responses.
Natural Killer (NK) cells: Lymphocytes that respond non-specifically to pathogens. NK cells are part of the innate immune system. http://www.immunecentral.com/immune-system/iss9.cfm
KIR: Killer Immunoglobulin-like Receptor. These receptors are expressed on natural killer cells, which are part of the innate immune system. Natural killer cells don’t target pathogens specifically, but can recognize certain features common to infected cells. KIRs interact with the receptors made by HLA genes, because one sign of an infected cell or distressed cell can be the loss of HLA receptors.
Links & Resources:
Barbara Shacklett’s laboratory at UC Davis:
The International HIV Controllers Study:
The Zephyr Foundation for elite controllers and LTNPs:
The published paper in the journal Blood:
Blood First Edition Paper, prepublished online December 23, 2008; DOI 10.1182/blood-2008-10-182709.
April L Ferre, Peter W. Hunt, J. William Critchfield, Delandy H Young, Megan M Morris, Juan C. Garcia, Richard B Pollard, Hal F Yee Jr., Jeffrey N Martin, Steven G Deeks, and Barbara L Shacklett
There exists a unique group of individuals who are able to durably control HIV in the absence of therapy. The mechanisms of control in these individuals remain poorly defined. In this study we examined CD8+ T-cell responses in blood and rectal mucosa from 17 “elite controllers” (viral load < 75 copies/ml), 11 “viremic controllers” (75-2,000 copies/mL), 14 non-controllers (>10,000 copies/mL), and 10 antiretroviral-treated individuals (<75 copies/mL). Production of IFN-, IL-2, TNF-, MIP-1, and CD107a by CD8+ T-cells in response to HIV-1 Gag stimulation was measured using flow cytometry. Our hypothesis was that ‘polyfunctional’ T-cells producing multiple antiviral factors would be most abundant in mucosal tissues of HIV controllers. Mucosal CD8+ T-cell responses were significantly stronger and more complex in controllers than in antiretroviral-suppressed individuals (P=0.0004). The frequency of 4-function responses in rectal mucosa was higher in controllers than in non-controllers and patients on therapy (P<0.0001). Mucosal responses in controllers were frequently stronger and more complex than blood responses. These findings demonstrate that many controllers mount strong, complex HIV-specific T-cell responses in rectal mucosa. These responses may play an important role in mucosal immune surveillance, as suggested by their relative enrichment among individuals who control HIV in the absence of therapy.
Additional recent publications:
J Intern Med. 2009 Jan;265(1):58-66. Mucosal T-cell responses to HIV: responding at the front lines. Shacklett BL, Critchfield JW, Ferre AL, Hayes TL.
Methods Mol Biol. 2009;485:359-74. Quantifying HIV-1-Specific CD8 (+) T-Cell Responses Using ELISPOT and Cytokine Flow Cytometry. Shacklett BL, Critchfield JW, Lemongello D.
Methods Mol Biol. 2009;485:347-56. Isolating mucosal lymphocytes from biopsy tissue for cellular immunology assays. Shacklett BL, Critchfield JW, Lemongello D.
PLoS ONE. 2008;3(10):e3577. Epub 2008 Oct 30. Magnitude and complexity of rectal mucosa HIV-1-specific CD8+ T-cell responses during chronic infection reflect clinical status. Critchfield JW, Young DH, Hayes TL, Braun JV, Garcia JC, Pollard RB, Shacklett BL.
J Virol. 2008 Jun;82(11):5398-407. Epub 2008 Mar 19. HLA class I-restricted T-cell responses may contribute to the control of human immunodeficiency virus infection, but such responses are not always necessary for long-term virus control. Emu B, Sinclair E, Hatano H, Ferre A, Shacklett B, Martin JN, McCune JM, Deeks SG.
Curr HIV/AIDS Rep. 2006 Feb;3(1):26-31. Review. Understanding the “lucky few”: the conundrum of HIV-exposed, seronegative individuals. Shacklett BL.
J Immunol. 2004 Jul 1;173(1):641-8. Abundant expression of granzyme A, but not perforin, in granules of CD8+ T cells in GALT: implications for immune control of HIV-1 infection. Shacklett BL, Cox CA, Quigley MF, Kreis C, Stollman NH, Jacobson MA, Andersson J, Sandberg JK, Nixon DF.