THE ROLE OF DISEASE STAGE, PLASMA VIRAL LOAD AND Tregs ON AUTO-ANTIBODY PRODUCTION IN THE SIV INFECTED NON-HUMAN PRIMATES

Abstract

Auto-antibodies appear in the sera of rhesus macaques following SIV infection. The present study was conducted to examine the role of viral load, anti-viral chemotherapy and stage of disease on the titers of such auto-antibodies and the spectrum of auto-antigens that become the target of such auto-immune responses. In addition, the role of regulatory T cells (Tregs) was also examined. Results of these studies showed that the highest auto-antibody titers were noted in animals with lower relative plasma viral loads with a wider spectrum of auto-antigens that are the target of such responses as compared with lower auto-antibody titers in animals with relatively higher plasma viral loads and a narrower spectrum of auto-antigens. Short term anti-viral chemotherapy did not influence the titers of auto-antibodies. While there was a gradual decrease in the frequency and absolute number of Tregs, the levels of Tregs was inversely correlated with viral load and lower autoantibody titers. The mechanisms for these differences remain unknown and suggest complex relationships exist between levels of immuno-suppression, auto-immune response, homeostatic proliferation and the spectrum of auto-antigens that become the target of such auto-immune responses.

INTRODUCTION

It is now widely recognized that HIV-1 infection in humans and SIV infection in rhesus macaques results in a massive depletion of CD4+ T cells primarily in the gut [1–4] during the acute infection period which is accompanied by rapid virus replication. This event is followed by a readily recognizable decrease in plasma levels of virus and a steady state of plasma virus levels referred to achievement of viral load “set point”. In most cases, this is followed by a chronic stage of disease which is a variable time period during which there is a gradual decreasing population of CD4+ T cells, a steady state of plasma viremia and a progressive loss of immune function [5]. While there is considerable variability in terms of numbers of total lymphoid cells during this entire acute and chronic phase of the disease, the peripheral lymphoid cell pool is to a large degree maintained. To a certain degree such depletion during the acute and early chronic stage induces the differentiation and maturation of T cell progenitors which require a functioning thymus gland. However, both the degree of depletion and the kinetics of depletion is such that it cannot replenish the number of cells needed to maintain the lymphoid cell pool to pre-infection levels. In addition, the infection also affects thymic tissue which also becomes gradually dysfunctional. It is reasoned that this void in the maintenance of T cells is compensated for by a process known as homeostatic proliferation and expansion (HPE) and since such HPE has been incriminated in the induction of autoimmune responses [6, 7] a summary of this issue is in order. Thus, as shown in other animal models, lymphopenia induces the remaining cells to proliferate in efforts to fill the void left by the depleting cells [8, 9]. It is also a normal physiologic function that occurs during the process of aging which accompanies failing thymic function. The precise mechanisms that lead to HPE continues to be defined but it is recognized that such proliferation and expansion require cytokines/growth factors and for select subsets of CD4+ and CD8+ T cells the presence of their cognate peptide bearing MHC molecules on antigen presenting cells (APC’s) [10, 11]. A number of studies have shown that differences exist in the requirements for naïve and memory CD4+ and CD8+ T cells to undergo HPE [12–16]. Results from such studies show that the differences include the competition that exists between naïve as compared with memory T cells for critical cytokines in concert with the expression of appropriate high versus low affinity receptors for the specific cytokines and peptide bearing MHC molecules and the relative avidity of the TCR’s for the specific peptide-MHC bearing molecules. Thus, the physiologic, immunologic and molecular understanding of the requirements and the mechanisms by which HPE occurs in hosts that become lymphopenic due to HIV in humans and SIV in non-human primates remains far from clear. This is an important issue since strategies aimed at reversing these pathological manifestations following HIV infection including full immune reconstitution is critical since anti-viral chemotherapy is clearly not an option for an increasing group of infected individuals [17] and even successful cART does not lead to full immune reconstitution [18] specially in the mucosal tissues [19].

Our lab has been studying the SIV infected non-human primate model of human AIDS for the past few decades and have previously shown that SIV infected rhesus macaques develop the same pathological manifestations as does HIV-1 in humans [20]. Germane to the present study, our lab has previously shown that SIV infected rhesus macaques develop significant levels of auto-immune antibodies [21] which seem to occur primarily during the chronic viremia period, which is a period during which considerable HPE is occurring. The spectrum and the number of self antigens that are the target of such auto-immune response have also been characterized but to a limited extent [22]. The purpose of the present study was to study in more detail, the role of viral load, regulatory T cells (Tregs), and the effect of anti-viral chemotherapy on the quality and quantity of auto-immune response in SIV infected rhesus macaques. Results of these studies show that higher viral loads are associated with lower auto-antibody titers but the spectrum of the target auto-antigens are narrower and focused as compared with lower viral load. There appears to be a gradual decrease in the frequency of Tregs and the number of Tregs were inversely correlated with viral load and lower autoantibody titers. Short term anti-viral chemotherapy had no detectable effect on autoantibody titers. It should be kept in mind that while high titers of auto-antibodies are noted in such monkeys, there is no clinical signs of auto-immune disease and thus immune deficiency leads to the breakdown of self tolerance but the clinical implications of the generation of such auto-antibodies remains to be defined. Results of these studies constitute the basis of this report.

MATERIALS & METHODS

Non-Human Primates

Heparinized blood samples from various groups of rhesus macaques (Macaca mulatta) and sooty mangabeys (Cercocebus atys) were the source of plasma samples utilized for the studies reported herein. A group of 26 rhesus macaques who were infected intravenously with 200 TCID50 of SIVmac251 (a lot of virus prepared in the CEM cell line) was used for the initial series of studies. A sub-group of these rhesus macaques following infection and during the chronic infection period were treated with PMPA (20 mg/kg) subQ daily for 28 days. The samples were from the monkeys that responded to the anti-viral chemotherapy with a decrease in viral load to below 100 viral copies/ml of plasma. The SHIV89.6P virus was initially obtained from Dr. Keith Reimann (Harvard Medical School, Boston, MA) and used to infect adult rhesus macaques. Lymph node cells from an acutely infected rhesus macaque was cultured in vitro with con-A and a virus stock of SHIV 89.6P prepared and used for the studies reported herein. The group of long term non-progressors (LTNP) were rhesus macaques that were infected with SIVmac239. This stock of SIVmac239 was prepared using primary rhesus macaque Con-A blasts. In addition, there were also blood samples obtained from SM that were experimentally infected with an SIV isolate termed FUo (a kind gift from Dr. S. Staprans, Emory University) that replicates efficiently in CD4+ T cells from sooty mangabeys. Each of the SIV infected animal used in the studies reported herein was housed in the BSL-2/3 non-human primate facility of the main Station of the Yerkes National Primate Research Center of Emory University. All animals were housed and cared for at the YNPRC in conformance to the guidelines of the Committee on the Care and Use of Laboratory Animals of the Institute of Laboratory Animal Resources, National Research Council and the Health and Human Services guidelines "Guide for the Care and Use of Laboratory Animals" and all protocols were reviewed and approved by the Emory IACUC.

Viral load determination

Plasma viral load was routinely monitored in each of the SIV infected monkey species by the NIAID, NIH CFAR sponsored Virology Core Laboratory of Emory University School of Medicine.

Flow cytometric analysis

Our laboratory routinely monitors each monkey for the frequency and absolute numbers of total T (CD3), CD4+ and CD8+ T cells and its naïve, central and effector memory T cells, monocytes, and NK cells using a panel of monoclonal antibodies [L. Pereira et al, in preparation]. For the present study, aliquots of PBMC samples were first incubated with 1 ug/ml of anti-FcR I monoclonal antibody (clone 2.4G2, courtesy of Emory CFAR core, Dr. R. Mittler, Emory University) for 30 min at 4C to block FcR. The cells were washed and then incubated with FITC-conjugated anti-CD20 antibody (B-D, Mountain View, CA) for 30 min at 4C washed and then incubated with 0.1 ml of varying dilutions of the monkey plasma to be screened for the presence of auto-antibodies for 30 min at 4C. the cells were once again washed and then incubated with a PE-conjugated goat anti-monkey IgG (Fisher Scientific ) for 30 min at 4C , washed and re-suspended in 0.1 ml of PBS, pH 7.4 containing 1% fetal bovine serum and subjected to flow cytometric analysis. All washes were performed using PBS pH 7.4 buffer. All samples were analyzed using a FACSCaliber (B-D, Mountain View, CA). Following gating for lymphocytes using forward and side scatter profiles and gating out the CD20 positive cells, the frequency and mean fluorescent intensity of the PE-positive cells was recorded using Cell Quest and FlowJo (TreeStar, Ashland. OR) software. The dilution of plasma that retained the maximal reactivity profile (highest MFI) was utilized as a yardstick for comparative evaluation.

Western Blot analysis

A pool of lymphoid cells from rhesus macaques were utilized to prepare cell lysates which were used to screen sera for the nature of the target antigens of the auto-antibodies present in the sera from the monkeys. Only sera that were positive for cell surface staining by flow cytometry were analyzed by Western Blots because in a preliminary study none of the sera that gave bands by Western Blots were negative by the flow cytometry assay. Briefly, cells were washed X3 in PBS, centrifuged and the PBS aspirated and the cell pellet suspended in lysis solution (1% SDS, 10 mM Tris pH 7.4) and heat denatured. The sample was subjected to boiling for 5 min, passage through a 26 g needle, centrifugation and the removal of the insoluble material. The protein content of the soluble material was determined using BCA reagent and aliquots crypreserved at –85C until use for Western Blot analysis. For Western Blot analysis, 5 ug of the cell lysate was loaded on each well of a 12 % SDS-PAGE and electrophoresed, transferred to a nylon membrane overnight and the strips cut. After blocking in 5% milk, the plasma sample to be analyzed was diluted 1/50 in the 5% milk and incubated for 1 hr. The strips were then washed 5 times with PBS containing 0.5% Tween 20 (PBS-T). This step was followed by incubation of the strip for 1 hr with goat anti-rhesus IgG diluted 1/500 (Fisher Scientific), washed and developed with an alkaline phosphatase conjugated rabbit anti-goat antibody diluted 1/1000. Following addition of the NBT substrate the bands were visualized and the image scanned. Controls consisted of a strip that was incubated with normal rhesus plasma (negative control) at the same dilution and a strip that was incubated with an aliquot of a pool of serum from rhesus macaques that were experimentally immunized with allogeneic cells for another study [23] that was known to contain antibodies against rhesus macaque cell surface proteins (positive control). The number of readily detectable bands were enumerated. In select experiments, positively reactive sera were analyzed at varying concentrations in efforts to determine the relative titers of the reactive antibodies.

Indirect immunofluorescence (IFA) for analysis of tissue specific reactivity

Reactivity of sera against various rhesus macaque tissues were determined by our laboratory established standard IFA technique. Briefly, heart, kidney, liver, lung, skeletal muscle, spleen tissue sections from a normal rhesus macaque (euthanized for physical trauma related injury ) were incubated with blocking antibody and following washing, incubated with 1 : 50 dilution of the sera to be analyzed for auto-antibody reactivity. After incubation, the slides were washed and then incubated with a 1/ 500 dilution of FITC-conjugated goat anti-rhesus IgG antibody (Fisher Scientific). The slides were then washed and incubated with slow fade reagent and then the fluorescent pattern of reactivity visualized using confocal microscopy (Orthoplan, Leitz-Wetzlar, Germany). Negative controls consisted of normal rhesus sera.

Enumeration of Tregs

Aliquots of PBMC were first incubated with 1 μg/ml of anti-FcR antibody (clone 2.4G2) for 15 min at 4° C, washed and then surface-stained for 15 min at 4° C with a pre-determined optimal concentrations of CD4-PerCP, CD95-FITC and CD25-PE or CD127-PE. The antibody clones utilized included: CD4-PerCP (clone L200), CD127-PE (hIL-7R-M21), CD95-FITC (clone DX-2) , all purchased from BD Pharmingen (San Diego, CA), CD25-PE (clone 4E3, Miltenyi Biotec, Auburn, CA), and FoxP3-APC (clone PCH101 or 236A/E7, E-Bioscience, San Diego, CA). Fixation and intracellular staining to detect FoxP3 was performed according to E-Biosciences protocol. Appropriate mAb isotype controls were included. Flow cytometric acquisition of at least 100,000 events from each sample was performed on a FACSCalibur flow cytometer. Samples were also analyzed on the LSR II system using the following panel: CD4 AmCyan (L200), CD95 PECy7 (DX2), both from BD Biosciences, CD25 Biotin (4E3, Miltenyi), Streptavidin ECD (Beckman-Coulter, Miami, FL) and FoxP3 PE (206D, BioLegend, San Diego, CA). Data acquisition and analysis was done using CellQuest (BD Biosciences) and FlowJo (TreeStar, Ashland, OR) software, respectively. The absolute number of Tregs were determined based on CBC counts performed on an aliquot of the blood sample from the same bleed.

RESULTS

Stage of disease and level of auto-antibody production in SIV infected rhesus macaques

The strategy utilized for these studies was to first screen the sera for the presence of auto-antibodies against cell surface molecules using standard flow cytometric procedures standardized in our lab using a FITC-conjugated anti-monkey IgG reagent. If the sera showed positive reactivity, an aliquot of the sera was then analyzed against lymphoid cell lysates using standard Western Blot techniques and for reactivity against tissue sections from a variety of organs from rhesus macaques utilizing standard indirect immunofluorescence techniques. This strategy was based on our previous findings which showed that none of the sera we had tested that were positive by Western Blot or by IFA were negative for cell surface staining which allowed us to follow this paradigm.

Sera from male monkeys prior to SIV infection rarely contained any significant levels of auto-antibodies at least in the group of young adolescent male animals utilized in the present study (denoted as titers of < 1 : 20). It should be noted that select sera from female rhesus macaques do show readily detectable allo-antibodies which are likely a result of pregnancy related allo-immunization and their significance on SIV infection is discussed below. As stated above while male rhesus macaques do not show detectable auto-antibodies prior to infection, shortly after infection with SIV, there was a clear production of readily detectable auto-antibodies, initially screened using flow cytometry and autologous PBMC samples. Sera from each of a total of 26 rhesus macaques obtained prior to and during the acute infection period (6–12 weeks), during the chronic infection period (n = 21, 16–72 weeks ) and during AIDS (n = 7 ) were screened for auto-antibody titers using the flow cytometric assay. As seen in Fig. 1, while there is considerable variability, a general trend in auto-antibody titers at least as measured by this assay is evident. Thus, relatively lower auto-antibody titers are noted during the acute phase. However, this is followed by an increase in the number of animals that show higher titers of auotantibodies. Sera from the monkeys that developed clinical AIDS, clearly had the highest autoantibody titers. Sera obtained from rhesus macaques who were either fast progressors (post infection with SHIV 89.6P) and died < 6 months post infection or long term non-progressors were also analyzed for titers of auto-antibodies. Whereas sera from the fast progressors (n = 6 ) did not contain any detectable auto-antibodies against cell surface antigens (data not shown), sera from 10 SIV infected long term non-progressor (LTNP) rhesus macaques (alive with undetectable viral load and asymptomatic > 2 years p.i. ) showed low levels of auto-antibodies (Fig. 1). In addition, none of the sera so far tested from uninfected (n = 12) , naturally (n = 22 ) or experimentally SIV infected sooty mangabeys (n = 6) prior to and following infection showed any detectable auto-antibodies (data not shown). Sera from 2 of the 7 monkeys with clinical AIDS had the highest titers of auto-antibodies. Both of these monkeys had chronic diarrhea and opportunistic infections and it is reasoned that such O.I.’s led to the exacerbation of the autoimmune response.

FIG. 1.

Plasma samples from 26 adult rhesus macaques obtained prior to infection and following 6–12 weeks post intravenous infection with 200 TCID50 SIVmac251 (acute infection period), from 21 of these infected but clinically asymptomatic monkeys post 16–72 weeks of infection (chronic infection period), 7 monkeys that developed clinical AIDS at varying time intervals post infection and a set of 10 long term SIV infected non-progressor (LTNP) rhesus macaques who were asymptomatic > 2 years post infection were assayed for titers of auto-antibodies using flow cytometry. Data shown reflects the highest dilution of the plasma that retained the maximal MFI seen with each plasma sample.

Role of viral loads on titers on autoantibody titers

When the titers of auto-antibodies detected during the chronic phase were analyzed as a function of viral loads, an interesting fact emerged. Monkeys with lower plasma viral loads appear to have relatively higher auto-antibody titers than monkeys with higher viral loads (see Fig. 2). The reason for this relationship are unclear at present since they did not appear to be related to the frequency or absolute number of CD4+ T cells (data not shown) and/or the levels noted in the same monkeys.

Fig. 2.

Plasma samples from SIVmac251 infected rhesus macaques that had plasma viral loads < 10,000 viral copies/ml (n = 9 ) and those that had > 10,000 viral copies/ml were analyzed for auto-antibody titers as described under Fig. 1.

Role of regulatory T cells (Tregs) on auto-antibody responses

The frequency and absolute numbers of Tregs in samples from 18 SIV infected rhesus macaques and for comparison 12 non-infected rhesus macaque monkeys and 6 non-infected and 6 SIV infected sooty mangabeys were also analyzed. As seen in Fig. 3 A and B, besides slight transient increases in the frequency of Tregs during the acute infection period, the SIV infected rhesus macaques showed a general trend towards a decrease in the frequency and absolute numbers of circulating Tregs as a function of time post infection. In marked contrast, while the frequency and absolute numbers of circulating Tregs in the uninfected sooty mangabeys is lower than uninfected rhesus macaques (a species specific difference) the frequency and absolute numbers remain pretty constant following experimental SIV infection. When data on the frequency and absolute numbers of Tregs were analyzed as a function of auto-antibody titers in the SIV infected rhesus macaques, there did not appear to be any clear correlation (data not shown). Thus, sera from monkeys that had relative high titers of auto-antibodies (n = 9, Fig. 2) had both low (n = 4 ) and higher relative levels (n = 5 ) of Tregs and monkeys with low titers of auto-antibodies (n = 9, Fig. 2) similarly had low (n = 6 ) and high relative levels (n = 3 ) of Tregs. These findings reflect studies on samples during the chronic viremia period. Thus, a more thorough study on the kinetics of autoantibody titers and levels of Tregs needs to be performed to arrive at any meaningful conclusions.

Fig. 3.

The frequency (Fig. 3A) and absolute numbers (Fig. 3B) of Tregs in the PBMC samples from 18 SIV infected rhesus macaques (9 with low and 9 with high viral loads) , 12 non-infected rhesus macaques, 6 non-infected and 6 experimentally SIV infected sooty mangabeys were determined at the identified time intervals as outlined in the methods section. The data at 0 weeks reflect mean values pre-infection and the rest post infection as indicated. The S.D. of the values was < 20%.

Spectrum of auto-antibody responses post SIV infection

Aliquots of the sera that were positive by flow cytometry were subsequently analyzed for the spectrum of host antigens that served as the target of the auto-antibodies using both standard Western Blot analysis and standard IFA techniques. Lysates of PBMC’s were utilized for the Western Blot studies and sections of heart, liver, lung, kidney, spleen, intestine, and muscle tissues were utilized for the IFA studies. In general, there appeared to be a broader tissue reactivity with increasing time post infection and with sera that were classified as high relative titers by the flow analysis. Patterns of reactivity of sera from chronically SIV infected monkeys when analyzed for reactivity showed that sera with lower global auto-antibody titers by flow had a lower spectrum of reactivity as compared with sera with high auto-antibody titers by flow analysis (see Table 1). However, upon closer examination, it appeared that the sera with a lower spectrum of target antigens reacted with the individual auto-antigens with a high relative degree of reactivity than sera that had a broader spectrum of reactivity. Thus, sera classified as low titer sera by flow showed positive reactivity against MHC class I and class II antigens by Western Blots (titers ranging from 1 : 256 to 1: 1024) as compared with sera classified as high titer sera by flow analysis (titers ranging from 1: 64 : 1 : 256). The identification of the MHC class I and class II antigens on Western Blots was made on the basis of reactivity of rhesus macaque MHC class I and II reactive monoclonal antibodies and thus needs to be interpreted with caution as other proteins with similar molecular weight could be present in the identified band, Of the target antigens identified besides MHC class I and II, there appeared to be reactivity against the viral receptor/co-receptor CD4, CCR5, molecules involved in complement fixation which included CD 55 and CD 59, cell adhesion molecules such as ICAM-1, LFA-1, VLA-4, cardiolipin, denatured DNA, a series of structural protein such as alpha-actinin, ezrin, moesin. co-filin, thrombospondin, vimentin, laminin etc. Of interest is the finding of IgG antibodies against the Fab chain of Ig only in sera from monkeys with clinical AIDS. The reason for this unique reactivity remains to be determined.

Table 1.

Spectrum of Auto-Antibody reactivity
Serum auto- Antibody titer	Western blots* bands	Heart	Spleen	IFA # Kidney	Lung	Liver	Intest.	Muscle
1:40	4 – 7	+	+	−	−	+	+	−
1:160	9 – 12	+	+	+	−	+	+	−
1:640	> 12	+	+	+	+	+	+	+

^*Number of bands detected using lymphocyte lysates, sera tested at a 1:50 dilution

^#Reactivity denoted as positive fluorescence with sera diluted 1:200

Effect of anti-viral chemotherapy on titers of auto-antibodies

Aliquots of sera from a set of monkeys that were positive by flow analysis prior to the institution of anti-viral chemotherapy (PMPA ) were also examined for titers following a short term 28 daily subQ dose of 20 mg/Kg. None of the post therapy sera examined (n = 14 ) showed any difference in titers when analyzed in parallel with sera obtained prior to therapy (data not shown). These data suggest that at least short term chemotherapy does not appear to influence titers of auto-antibody although such chemotherapy was successful in decreasing plasma viral loads to undetectable levels.

DISCUSSION

Lymphopenia which occurs following HIV infection in humans and SIV infection in certain species of non-human primates also commonly occurs in a number of other clinical conditions. Thus lymphopenia occurs secondary to a number of other viral infections, following chemotherapy and/or radiation therapy of certain malignancies, in select auto-immune diseases and allograft rejection [22, 24–28]. One of the immediate manifestation of such lymphopenia is the triggering of the bone marrow and thymus progenitor cell pool to induce the maturation of the depleted lymphoid cell lineage (s), the mobilization of cells from existing cellular depots and most prominently the proliferation of the residual cells to fill the space that is vacated by the depletion of the lymphoid cells. This latter phenomenon of replenishment by lymphoid cell proliferation is termed “homeostatic proliferation” [29]. Thus, there is considerable physiological pressure to ensure that the size of the lymphoid cell population and to certain extent the repertoire is maintained. While the thymus plays a major role in the maturation of the pre-thymic T cells and in the relative proportions of the various subsets of T cells, with aging such function diminishes, which is compensated in part by peripheral homeostatic mechanisms. There is evidence that HIV infection in humans and SIV infection in rhesus macaques affects thymic function thus limiting its ability to re-populate. Mobilization of lymphoid cells from depots also contributes to the re-population of the peripheral lymphoid cell pool but has limited capacity. Thus, following the initial phase of mobilization, there is limited ability of such depots to supply more lymphoid cells. It thus becomes clear that hyper-proliferation expansion (HPE ) is likely to be one of the major sources for replenishment of the peripheral lymphoid cell pool and the involvement of HPE in the genesis of autoimmunity has been previously described [6, 7]. There are basically two distinct and opposing factors that govern the degree, quality and kinetics of homeostatic proliferation, those that provide positive and those that provide negative signaling. The re-circulating pool of T cells are physiologically influenced by such factors continuously and as such, compete for such factors which translates into T cells that continue to proliferate, die and/or are replaced. Among the plethora of positive signals are the interaction of the TCR with its cognate peptide bearing MHC molecules, a set of co-stimulatory molecules such as B7 for CD4+ T cells and among factors are the well known cytokines and to some extent chemokines such as IL-2, IL7 and IL-15, IL-12, IL-21, CCL21 [30–34]. Among the negative signals are those generated by Tregs, PD-1/PD-L1 interactions, CTLA-4/CD80–86 interactions, role of BTLA and select cytokines [35, 36]. It is also well recognized that those T cells that have a high avidity for self peptide-MHC ligands have a greater potential to undergo homeostatic proliferation than those with low avidity. Given the degree of depletion of CD4+T cells that occurs during HIV and SIV infection, it seems that the remaining cells with the highest avidity contribute to HPE and as such are contributing to a skewing of the repertoire of T cells. It is thus reasoned that such perturbation of the clones of T cells during HIV infection in humans and SIV infection in non-human primates is one of the contributing factors in the induction of auto-immune antibodies. This view is strongly supported by findings in animal models. Thus, the finding that thymectomy of 1–2 day old mice leads to a variety of autoimmune disease affecting multiple organs such as the stomach, thyroid, pancreatic islets, adrenal glands, gonads) which is also seen in rats following thymectomy and sub-lethal irradiation [37, 38] provides support to this view.

Our laboratory has previously identified a series of overlapping phases during lentiviral infection during which auto-immune responses have been documented [22] with specificity for different target antigens and likely a result of distinct mechanisms. These phases include the pre-infection stage, the acute, the chronic, the post anti-viral chemotherapy stages and the stage of clinical AIDS. As far as the pre-infection stage, while sera from male monkeys do not normally contain auto and/or allo-antibodies, sera from multiparous female macaques contain readily detectable levels of allo-antibodies. Our lab has similarly noted allo-antibodies in the sera from non-HIV infected IV drug abuse patients and select cohorts of HIV negative female sex workers. Since such antibodies may have specificity for CD4 or CXCR4/CCR5, it is possible that such antibodies can play a role in the pre-infection period in terms of transmission resistance. Similarly, allo-sensitization could result in the synthesis of chemokines that can play a role in protection from viral infection and/or maintaining low viral loads [39]. During the acute stages of infection, there is a considerable degree of pathology in the gut associated lymphoid tissues [1–4] which induces mobilization of lymphoid cells to traffic to the gut and induction of homeostatic proliferation. This stage is also marked with hyper-gamma-globulinemia, T cell dysregulation and the induction of auto-antibodies. There is also evidence that the institution of anti-retroviral therapy which reduces viral loads also leads to increases in CD4 levels and this gives rise to a syndrome termed immune reconstitution syndrome (IRS) and one of the side effects of such IRS once again is associated with the induction of auto-antibodies. Finally, during clinical AIDS, there appears to be a complete breakdown of immuno-regulatory mechanisms which in turn leads to the generation of a wide spectrum of auto-antibodies. The key to the analysis of data from such different phases of the disease is to not only define the precise mechanisms that are at the basis of such auto-immune responses but also to determine whether such auto-antibodies have any pathological effect or mere markers of immune dysregulation. Most of our studies have failed to show any clinical significance for the presence of such auto-antibodies in the SIV infected rhesus macaque model. It has always been the dogma that immune-suppression in general mutes the level of auto-immune disease such as that seen in SLE patients during pregnancy. Thus, if HIV-1 and SIV infection induces immune suppression, how does lead to auto-immunity? Perhaps the reasons for the lack of clinical disease in the presence of such auto-immune antibody responses in the SIV infected rhesus macaques may shed light on this issue.

The findings from the present study suggest that SIV infection of rhesus macaques clearly leads to the generation of auto-immune antibodies which is likely initiated right after the acute infection period and is prominent during the chronic viremia period. What is difficult to understand is the precise role of viral load. Thus, the data appear to indicate that in fact the autoantibody titers are higher in monkeys with relatively lower plasma viral loads than monkeys with relatively higher viral loads at similar times post infection. It is possible that higher viral loads are sub-optimal for the CD4+ T cell helper responses that are required for the induction of auto-immune responses and that a balance between viral load induced immuno-suppression and homeostatic proliferation must operate to regulate such levels of auto-immune antibodies. In this regard, it is important to note that viral load per se cannot by itself account for the induction of auto-antibodies, since none of the SIV infected disease resistant mangabeys to date have demonstrated any significant levels of auto-immune responses. Thus, it is not the level of viremia and the virus that is the basis for the induction of auto-immune antibodies but it is the level of immune dysfunction induced in rhesus macaques but not sooty mangabeys which is the basis of the induction of the auto-immune antibody response.

Since a prominent role of Tregs has been advanced in a number of auto-immune disease [40], it was reasoned appropriate to examine the role of this lineage in both the disease susceptible and disease resistant SIV infected rhesus macaques and sooty mangabeys, respectively. SIV infection did appear to perturb the frequency (during the acute infection period) and the gradual reduction in the absolute number of Tregs as a function of time post infection (see Fig. 3A and 3B). However, there was no relationship with either titers of auto-antibody or the spectrum of auto-antibodies and levels of Tregs. The reasons for this lack of relationship are not clear at present but it is likely that it is the functional status of the Tregs and not their mere presence that dictates this lack of correlation. It is also possible that peripheral blood evaluation for this lineage of Tregs may not be the most informative. There does appear a relationship between viral loads and absolute numbers of Tregs [22]. Thus the higher the viral load, the lower the numbers of Tregs but such higher viral loads led to lower relative levels of auto-antibodies. It is important to keep in mind that SIV much like HIV may decrease the levels of Tregs as a general function of viral depletion of the CD4+ T cell lineage and as such the decrease of Tregs noted in this study may be secondary to the general depletion of CD4+ T cells and not specific targeting of this sub-lineage which may account for the lack of correlation with levels of auto-antibody as seen in this study. Lastly, a role for other lineages of cells with immune-regulatory potential that may play a more important role cannot be discounted. The lack of an effect of chemotherapy in this context is difficult to assess. Thus, while no effect of anti-viral chemotherapy on levels of auto-antibodies was noted in the present study, a previous study on the role of anti-viral chemotherapy on levels of Tregs showed that there indeed was a significant effect of such chemotherapy on levels of Tregs (L. Pereira et al., in preparation). Thus, there seems to be a paradox in that while anti-viral chemotherapy reduces viral load to undetectable levels and leads to increases in levels of Tregs, there was no correlation of such viral load reductions with levels of auto-antibodies. If Tregs indeed control levels of auto-antibodies, these data can only be explained by the fact that such increased levels of Tregs were non-functional increases and/or not sufficient time had elapsed between the decreases of viral load and effect on auto-antibody production. Functional studies of the increased levels of Tregs post antiviral chemotherapy is needed to address this issue.

What is of interest is our findings is the differences noted in the spectrum of auto-antibodies induced. As seen in Table 1, increasing titers of auto-antibodies was correlated with a broader spectrum of auto-antigens being recognized. However, since the increased titers of auto-antibodies were seen in monkeys with lower viral loads, it seems logical that lower viral loads leads to not only higher titers of auto-antibodies but a broader spectrum of auto-immune responses. A further detailed study, however, showed, that the levels of auto-antibodies against specific targets such as MHC class I/II molecules was significantly higher (p< 0.0001) in the plasma samples that had relatively lower overall titers of auto-antibodies. These data indicate that increased viral loads lead to lower overall titers of auto-antibodies (fewer target auto-antigens) but that the auto-antibody responses in these animals is more focused to select auto-antigens. Conversely, lower overall auto-antibody titers are directed against a wider number of self antigens but the titers against select auto-antigens is lower. More specific analysis of these sera is currently underway using recombinant auto-antigens to address this issue. One possible explanation for these findings could be that that higher viral load leads to greater depletion of CD4+ T cells and/or greater degree of immune dysfunction causing a higher degree of stress on homeostatic proliferation and skewing of the repertoire which may result in the expansion of a more select clones of auto-reactive T cells and vice versa. A more detailed study of this phenomenon is required before conclusions can be made.