It has been clearly established that the state of cellular activation is pivotal for successful replication of HIV . In vitro, cell–cell interaction during antigen presentation has been demonstrated to enhance HIV replication in CD4+ T cells and antigen-presenting cells (APC) alike. Lymphocyte proliferation in response to antigen presentation is essential for HIV reverse transcription and incorporation of proviral DNA , whereas engagement of adhesion molecules and release of activating cytokines during interaction with antigen-specific T cells enhances HIV replication in APC .
HIV expression and immune activation, in response to infection or immunization, demonstrate a similar correlation in vivo. Previous studies have demonstrated that opportunistic  and bacterial infections  or parenteral immunization [6–8] induce transient bursts of plasma viraemia in HIV-infected subjects. Demonstration that immunization with tetanus toxoid renders peripheral blood mononuclear cells from non-infected individuals more susceptible to HIV infection in vitro  suggests that such immune activation may enhance HIV replication in vivo by increasing the pool of susceptible target cells. Thus, levels of immune activation may potentially affect disease progression, a possible explanation for the faster disease progression in sub-Saharan Africa, which is associated with a higher prevalence of infections than in developed countries .
Progression to AIDS is associated with characteristic immunopathological changes in tissue lymph nodes, and HIV replication is active in these sites throughout the period of clinical latency . However, the contribution of gut-associated lymphoid tissue (GALT), which contains 60% of the total body's T-lymphocyte pool , to the pathogenesis of HIV infection both in terms of virus production and disease progression has not been clearly defined. Constant stimulation of this large immune cell population by exposure to infectious agents is likely to provide significant contribution to the patients' total viral load by providing a continuous source of permissive cellular targets.
Gut immune response to combined whole cell–toxin B subunit oral cholera vaccine has been well studied [12,13]. Oral cholera vaccine induces a CD4 T-lymphocyte-dependent immune response in Peyer's patches and mesenteric lymph nodes , after which primed B and T lymphocytes migrate in peripheral circulation [15,16]. Although cholera is not a clinically important infection in the context of HIV disease, nor a major health problem in Brazil, oral cholera vaccine was used in this study since it provides a reproducible, safe and well-defined model of gut immune stimulation for both HIV-infected and uninfected individuals .
The aim of this investigation was to evaluate the effects of intestinal mucosal immune stimulation on HIV plasma load using oral cholera vaccine as a potent defined oral antigen.
Patients and methods
Patients and ethical approval
Twelve HIV-1-infected men aged 30–60 years (mean, 42.3 years) from the Casa de Apoio Santo Antonio in Rio de Janeiro, Brazil, volunteered for the study. The Casa de Apoio Santo Antonio is a home for HIV-infected homeless individuals, where the patients receive medical, nutritional, and psychological counselling and care as well as social support and accommodation . HIV clinical status was assessed by experienced clinicians using Centers for Disease Control and Prevention criteria . Since the majority of the volunteers were illiterate, verbal consent was obtained from all subjects after explanation of the risks and benefits of the trial, before entry into the study, which was approved by the Ethical Committee of FIOCRUZ (Fundação Oswaldo Cruz). Patients were free to withdraw from the study at any point with no constraint to their further clinical follow-up.
A single dose of combined whole cell–toxin B subunit oral cholera vaccine (SBL Vaccin AB, Stockholm, Sweden) was given, after fasting for 1 h, in 100 ml of 2% sodium bicarbonate solution to buffer gastric acid. Blood was collected before immunization (day 0) and on days 2, 4, 6, 10 and 15 thereafter. Five HIV-1-infected men with matched ages (range, 30–58 years; mean, 44.8 years), clinical and socioeconomical status were recruited as controls, and were given bicarbonate solution alone. Blood was collected on the same schedule as the study group. Plasma aliquots were frozen at −70°C until assayed. All patients were followed by clinicians, and were counselled to attend the unit for any adverse effects related to HIV disease or intestinal symptoms. In addition, all patients except one had been on antiretroviral monotherapy (zidovudine or didanosine) for at least 6 months prior to the study, and this was continued during the 3-month study period for ethical reasons. After conclusion of the study period, patients continued routine clinical follow-up.
CD4 lymphocyte counts
Lymphocyte subset counts (CD3, CD4, CD8) were performed by FACScan (Beckton-Dickinson, San Jose, California, USA) using anti-CD3/CD4 and anti-CD3/CD8 conjugated monoclonal antibody beads on the day of immunization.
β2-Microblubulin assays were performed by competition enzyme-linked immunosorbent assay (ELISA) using commercially available kits (Pharmacia, Uppsala, Sweden) and following the manufacturer's instructions.
Cholera toxin-specific antibody response
Plasma antitoxin-specific IgG and IgA antibodies were assayed by ELISA as described previously . Briefly, cholera toxin-coated 96-well plates (Costar, Cambridge, Massachusetts, USA) were sequentially incubated with skimmed milk block, test serum and goat anti-human IgG or IgA alkaline phosphatase conjugate (Sigma, St Louis, Missouri, USA). Antitoxin levels were expressed in nominal ELISA units determined by the pooled slope method relative to reference sera of known activity included on each plate.
Quantification of plasma viral load
Plasma samples were assayed by nucleic acid sequence-based amplification (Organon Teknika, Boxtel, the Netherlands) according to manufacturer's protocols. The lower detection limit of this assay was 400 copies/ml.
The Wilcoxon matched pairs signed rank-sum test was used to evaluate viral load changes within each group at day 0 compared with days 2, 4, 6, 10 and 15, respectively. A correlation test was performed between viral load on different days and CD4 cell counts.
Clinical evaluation, CD4 cell counts and β2-microglobulin assay
The clinical status and peripheral blood CD4 cell counts of subjects are listed in Table 1. There were no adverse clinical effects following immunization and no deterioration in clinical outcome during 3 months of study follow-up. Thirteen out of 17 subjects were clinically followed up for more than 15 months and clinically improved under triple therapy, whereas three subjects (two immunized and one control) died from HIV-related causes not linked to immunization 7–15 months after the study. There was no variation in the plasma β2-microglobulin levels in HIV-infected immunized or control subjects (data not shown) at the timepoints studied.
Cholera toxin-specific antibody response
High pre-immunization anti-cholera toxin titres were observed in both study groups, independent of clinical status and CD4 cell counts. Only four subjects (patients 2, 4, 8 and 10) had significant plasma antibody response to immunization (more than twofold increase in specific IgG levels; Table 1). These four subjects had CD4 cell counts above 200 × 106/l. There was no increase in IgG ELISA activity in any of the four control subjects tested who did not receive oral cholera vaccine. Anti-cholera toxin IgA was tested only in those individuals who had an increase in IgG levels, as previously demonstrated in HIV-infected subjects where plasma IgA responses were only detected in those who mounted an increase in specific IgG titres .
Quantification of HIV load
Immunization with oral cholera vaccine induced transient increase in viral load in all HIV-infected subjects (range, 2–60-fold; P = 0.005; Fig. 1). Baseline viral load ranged from below 400 to 40 000 copies/ml (medians, 6300 and 7200 copies/ml), and the peaks ranged from 3900 to 2 400 000 copies/ml (medians, 23 000 and 46 000 copies/ml). There was some variability among peak responses, which occurred between days 2 and 15, with a tendency to cluster around day 6 (Table 2). The increase was statistically significant (α = 0.05) on days 2 (P = 0.017), 6 (P = 0.025), and 10 (P = 0.021). There was no observed correlation with peripheral blood CD4 cell counts (P > 0.10 on all days) or with clinical status. High viral load increases (more than fivefold) were observed in symptomatic patients with CD4 cell counts below 50 × 106/l (patients 11 and 12; Table 2) as well as in those whose CD4 cell counts were above 200 × 106/l (patients 2, 5 and 10). No difference in viral load increase was observed between patients receiving different antiretroviral regimens. Fig. 2 shows viral load changes in 12 immunized subjects and five controls, with minimum, maximum and median values, as well as the range between the 25th and 75th percentiles for each day. The baseline viral load of control subjects ranged from below 400 to 120 000 copies/ml (mean, 22 680 copies/ml) but, in contrast with the immunized group, there was no pattern of increased viraemia observed after immunization (P = 0.5; Figs 1 and 2).
Data presented in this study indicate that immunization with oral cholera vaccine induces a transient increase in HIV viraemia of up to 60-fold, regardless of clinical stage of HIV infection, peripheral blood CD4 cell count and immune response, as assessed by specific antibody titre. These data are in strong agreement with previous studies, which demonstrated that parenteral immunization increased HIV levels in antiretroviral-naive subjects . However, this is the first reported observation that oral immunization stimulates plasma HIV load. Previous parenteral immunization studies in HIV-infected patients were carried out with recall antigens delivered subcutaneously [6–8]. No previous studies have reported changes in response to immunization by the mucosal route, which is the mode of naturally acquired infections.
We have previously characterized the immune response to oral cholera vaccine in terms of kinetics of specific IgG and IgA following immunizations in both normal volunteers and HIV-infected subjects . This study demonstrated that immune response to the B subunit of cholera toxin is reduced in advanced HIV infection, but has proved to be a clinically safe and potent stimulus of the immune response in HIV-infected individuals with established immunodeficiency . The lack of IgG and IgA response in most of the subjects in the present study is to be expected because cholera vaccine may require two doses to stimulate response in patients with advanced HIV infection, particularly those who respond in a recall manner . The high baseline ELISA cholera toxin-specific antibody titres observed in study and control subjects is likely to reflect a cross-reaction with shared epitopes of cholera toxin and enterotoxigenic Escherichia coli , a frequent cause of diarrhoea in Brazil. As with our previous studies in Kenya , such serological findings probably indicate that cholera toxin is a recall antigen in this population.
All except one subject in this study had been on anti-retroviral monotherapy for at least 6 months prior to the study and such treatment was continued for ethical reasons during the entire investigation period. While levels of resistance to antiretroviral therapy in the study population are unknown, antiviral activity within individual subjects may have masked the full potential of oral cholera immunization to enhance HIV viraemia, as has been reported in previous studies . Furthermore, potential differences in antiretroviral resistance patterns may account for some of the observed variations in the magnitude of enhanced plasma HIV levels.
Variability in timepoints of peak HIV increase in plasma observed in our study is consistent with similar findings after immunization with parenteral tetanus toxoid . Such common intrapatient variability may reflect the heterogeneity in immune response between individuals. However, in this study, all responses occurred within 15 days of immunization and there was a tendency to cluster around the mean of 6 days. In contrast to parenteral vaccine studies, there was no apparent correlation between time and duration of HIV peak increase with peripheral blood CD4 cell levels [6,7]. Indeed, an increase in HIV viraemia was observed in some patients with CD4 cells below 50 × 106/l.
The major contribution to plasma viraemia may have been provided by release of virus from HIV-infected macrophages within GALT, which may explain the increased HIV load despite lack of specific antibody response, or from lymphocytes proliferating in response to cholera toxin presentation by APC also within GALT, or shortly after lymphocyte migration to the blood. Such a finding is supported by our previous observation that cholera toxin-specific IgG and IgA antibody-forming cells appear in peripheral blood between 5 and 7 days post-immunization  and T cells at around 10 days .
Parenteral vaccination has been previously shown to stimulate a transient increase in plasma viraemia without having adverse effects on disease progression [7,8]. Our study is in agreement with these findings and oral immunization had no clinical impact on health or disease progression in these patients over the study period (3 months).
These data clearly demonstrate that a defined acute mucosal immunogen has profound but transient effects on HIV viral load and suggests that GALT under constant immunostimulation is likely to provide a major contribution to the total body levels of HIV. Such data may explain the observation that CD4 T lymphocytes decrease faster in number in intestinal mucosa than in peripheral blood . In addition, these data provide a possible mechanism to support the clinically observed accelerated HIV disease progression in African patients with high levels of chronic intestinal parasitic infestation and frequent infections with other enteropathogens.
The authors thank the patients and the nursing staff of the Casa de Apoio Santo Antonio for their cooperation and willingness to take part in this study, Dr Sarah Galpin and Dr Mounir Ait-Khaled of the Department of Genito-Urinary Medicine and Communicable Diseases, Imperial College School of Medicine at St Mary's Hospital, London, UK, for technical advice on viral load (nucleic acid sequence-based amplification) assays, and Eduardo André M.M. Costa for computer assistance.
1. Zagury D, Bernard J, Leonard R: Long-term cultures of HTLVIII-infected T cells: a model of cytopathology of T-cell depletion in AIDS
2. Bukrinsky M, Stanwick T, Dempsey M, Stevenson M: Quiescent T lymphocytes as an inducible virus reservoir in HIV-1 infection
3. Shattock R, Burger D, Dayer J-M, Griffin GE: Enhanced HIV replication in monocytic cells following engagement of adhesion molecules and contact with stimulated T cells
. Res Virol
4. Donovan RM, Bush CM, Markowitz NP, Baxa DM, Saravolatz LD: Changes in virus load markers during AIDS-associated opportunistic diseases in human immunodeficiency virus-infected persons
. J Infect Dis
5. Bush CM, Donovan RM, Markowitz NP, Kvale P, Saravolatz LD: A study of HIV RNA viral load in AIDS patients with bacterial pneumonia
. J Acquir Immune Defic Syndr Hum Retrovirol
6. Staprans SI, Hamilton BL, Follansbee SE, et al.
: Activation of virus replication after vaccination of HIV-1-infected individuals
. J Exp Med
7. Stanley SK, Ostrowski MA, Justement JS, et al.
: Effect of immunization with a common recall antigen on viral expression in patients infected with human immunodeficiency virus type-1
. N Engl J Med
8. Cheeseman SH, Davaro RE, Ellison RT III: Hepatitis B vaccination and plasma HIV-1 RNA [letter]
. N Engl J Med
9. Gilks CF: The clinical challenge of the HIV epidemic in the developing world
10. Pantaleo G, Grazosi C, Demarest JF, et al.
: HIV infection is active and progressive in lymphoid tissue during the clinically latent stage of disease
11. Mowat AM, Viney JL: The anatomical basis of intestinal immunity
. Immunol Rev
12. Svennerholm A-M, Jertborn M, Gothefors L, Karim AM, Sack DA, Holmgren J: Mucosal antitoxic and antibacterial immunity after cholera disease and after immunisation with a combined B subunit-whole cell vaccine
. J Infect Dis
13. McGhee JR, Mestecky J, Dertzbaugh MT, Eldridge JH, Hirasawa M, Kiyono H: The mucosal immune system: from fundamental concepts to vaccine development
14. Hörnqvist E, Goldschmidt TJ, Holmdahl R, Lycke N: Host defence against cholera toxin is strongly CD4+ T cell dependent
. Infect Immun
15. Lewis DJM, Novotny P, Dougan G, Griffin GE: The early cellular and humoral immune response to primary and booster oral immunization with oral cholera toxin B subunit
. Eur J Immunol
16. Castello-Branco LRR, Griffin GE, Poulton TA, Dougan G, Lewis DJM: Characterization of the circulating T-cell response after oral immunization of human volunteers with cholera toxin B subunit
17. Lewis DJM, Gilks CF, Ojoo S, et al.
: Immune response following oral administration of cholera toxin B subunit to HIV-1-infected UK and Kenyan subjects
18. Castello-Branco LR, Linhares-de-Carvalho MI, Ponce de Leon AC: AIDS therapy in Brazil [letter]
19. Centers for Disease Control: 1993 Revised classification system for HIV infection and expanded surveillance case definition for AIDS among adolescents and adults
20. Holmgren J, Svennerholm A-M, Gothefors L, Jertborn M, Stoll B: EnterotoxigenicEscherichia colidiarrhea in an endemic area prepares the intestine for an anamnestic immunoglobulin A antitoxin response to oral cholera B subunit vaccination
. Infect Immun
21. Kroon FP, van Furth R, Bruisten SM: The effects of immunization in human immunodeficiency virus type-1 infection
. N Engl J Med
22. Lim SG, Condez A, Lee CA, Johnson MA, Elias C, Poulter LW: Loss of mucosal CD4 lymphocytes is an early feature of HIV infection
. Clin Exp Immunol