Haissman, Judith M BM*; Vestergaard, Lasse S MD, PhD†‡; Sembuche, Samuel MD, MPH§; Erikstrup, Christian MD, PhD‖; Mmbando, Bruno MSc§; Mtullu, Samuel MD¶; Lemnge, Martha M MSc, PhD§; Gerstoft, Jan MD, DMSc†; Ullum, Henrik MD, PhD*
From the *Department of Clinical Immunology, Copenhagen University Hospital, Copenhagen, Denmark; and †Department of Infectious Diseases, Copenhagen University Hospital (Rigshospitalet), Copenhagen, Denmark; ‡Department of Bacteriology, Mycology and Parasitology, Statens Serum Institut, Copenhagen, Denmark; §National Institute for Medical Research, Tanga Centre, Tanga, Tanzania; ‖Department of Clinical Immunology, Aarhus University Hospital, Skejby, Denmark; and ¶Tanga Aids Working Group, Tanga, Tanzania.
Received for publication October 14, 2008; accepted June 17, 2009.
Presented at ImmunoRio 2007: 13th International Congress of Immunology, August 21-25, 2007, in Rio de Janeiro, Brazil.
The authors have no commercial or other association that might pose a conflict of interest.
This study was funded by grants from the Danish Medical Research Council (271-06-0134); the Novo Nordic Foundation; AP Møller (Fonden for Lægevidenskabens Fremme); the Danish Research Agency (2117-05-0147); and the Cluster of International Health, University of Copenhagen.
Correspondence to: Henrik Ullum, MD, PhD, Department of Clinical Immunology, 2031, Center of Clinical Investigation, Copenhagen University Hospital, Blegdamsvej 9, dk-2100 Copenhagen, Denmark (e-mail: email@example.com).
Inflammation has repeatedly been shown to cause increased viral replication, viral entry, and immune dysfunction in HIV infection, indicating that immune activation is a driving factor in immune impairment and HIV/AIDS progression.1 Studies conducted in Europe and North America further indicate that viral replication itself is the main cause of the chronic inflammatory state found in HIV-infected individuals.1 The HIV epidemic in these settings differs from the epidemic in sub-Saharan Africa in several aspects. Studies have shown that abnormal immune activation is present in HIV infection in African individuals2,3 and that such activation is present even in HIV-seronegative controls.3 The immune activation has been hypothesized to be due to diverse dietary and environmental conditions, especially the abundance of infectious agents found in sub-Saharan Africa. Therefore, it has been suggested that coinfections play a primary role in the HIV epidemic in Africa and that the treatment of coinfection could decrease immune activation and retard the progression to AIDS in the HIV-infected African population.4 However, studies have shown that the impact of antiretroviral treatment (ART) on immune recovery is strongly related to a reduction in immune activation through the decrease in viral load.5 With the present effort to roll out ART programs in sub-Saharan Africa, it is important to gain insight into the driving factors of immune activation in sub-Saharan HIV-infected individuals. We have therefore conducted a study to investigate the level of immune activation in HIV-uninfected individuals and HIV-infected individuals after treatment with ART.
Study participants were recruited from the Tanga Aids Working Group (TAWG), a local nongovernmental organization conducting HIV testing and care in Tanga, northeastern Tanzania. Eligible study participants were adult (≥18 years) males and nonpregnant females attending TAWG during the period from June to August 2006. Only HIV-infected individuals naive to treatment with antiretroviral drugs were included in the study. HIV-uninfected participants, serving as negative controls, were recruited among individuals attending TAWG for HIV testing. HIV-infected individuals, who at baseline assessment were discovered to be clinically eligible to begin ART, were referred to the Care and Treatment Clinic at Bombo Regional Hospital in Tanga and offered ART according to national guidelines. Study patients started on ART after the baseline assessment and were invited for additional follow-up visits after 2 and 4 months of treatment.
The Medical Research Coordinating Committee of the National Institute for Medical Research, the Ministry of Health in Dar es Salaam, Tanzania, approved the study. Study information was provided, and written informed consent was obtained from all study participants.
Blood for cytokine and HIV RNA measurements was kept on ice until centrifugation at 5°C for a maximum of 20 minutes after sampling and stored at −80°C. Plasma HIV RNA was assessed by a reverse transcriptase polymerase chain reaction assay (Roche Amplicor HIV Monitor Test v1.5; F. Hoffmann-La Roche, Basel, Switzerland).
Plasma concentrations of the cytokines, interleukin (IL)-1 receptor antagonist (ra), IL-6, IL-8, IL-10, tumor necrosis factor (TNF)-α, and monocyte chemotactic protein (MCP)-1, were measured on the Luminex 100 platform (Luminex Corporation, Austin, TX, with assay Fluorokine MAP, R&D Systems, Cary, NC).
Levels of HIV RNA and cytokines IL-1ra, IL-8, TNF-α, and MCP-1 were log10 transformed to approximate normal distribution. For IL-6 and IL-10, a number of samples were below the level of detection and a normal distribution could not be obtained.
Subjects were stratified according to HIV status, and HIV-infected individuals were further stratified according to CD4 count (above or below 200 cells/μL). Mean plasma concentrations of IL-1ra, IL-8, TNF-α, and MCP-1 were compared using analysis of covariance adjusting for age and sex. Plasma concentrations of IL-6 and IL-10 were compared by the Mann-Whitney U test.
Univariate and multivariate linear regression analyses were performed to investigate associations between cytokine levels (IL-1ra, IL-8, TNF-α, and MCP-1) and viral load and hemoglobin and CD4 cell count, respectively. Spearman correlation coefficient was used to examine paired correlations between levels of cytokines IL-6 and IL-10 and the 3 following markers: CD4 T-cell count, viral load, and hemoglobin. The effect of ART on cytokine levels was assessed by the paired t test for IL-1ra, IL-8, TNF-α, and MCP-1 and by the Wilcoxon signed rank test for IL-6 and IL-10.
Characteristics of Study Participants
A total of 283 participants were recruited for our study, of which 229 were ART naive and HIV antibody seropositive, whereas 54 were HIV antibody seronegative (Table 1). HIV-infected individuals included a smaller proportion of male participants [63 (28%)] than the HIV-uninfected participants [29 (54%)] (P = 0.01) (Fisher exact test). The age distribution among HIV-uninfected male and female participants was similar [mean age of males was 38 years and standard error of the mean (SEM) was 2.7 years, and mean age of females was 38 years and SEM was 3.4 years), but males were older than females [mean age of males was 42 years (SEM 1.1 years), and mean age of females was 36 years (SEM 0.8 year), P < 0.001 (Student t test)] among HIV-infected individuals. Therefore, adjustment for age and sex was done statistically where appropriate. The clinical profiles of the uninfected participants, all HIV-infected participants, and HIV-infected participants after 2 and 4 months of ART are summarized in Table 1.
The analysis of ART response after 2 months of follow-up is based on 35 patients with a median treatment period of 62 days (range: 43-91 days), and the analysis after 4 months is based on 22 patients with a median treatment period of 123 days (range: 92-152 days). There were no statistically significant differences in age, gender distribution, body mass index, viral load, CD4 cell count, white cell blood count, and cell subsets or in baseline cytokine levels between the patients retained for analysis and the patients lost to follow-up (data not shown). Therefore, only baseline characteristics for all patients with CD4 cell count less than 200 cells per microliter are shown in Table 1.
Viral load decreased after 2 months of ART, and a further decrease in viral load after 4 months of treatment was observed. The decrease in viral load was observed for all patients. Eleven (31%) patients had HIV RNA levels below detection level after 2 months of treatment, and 13 (59%) patients had HIV RNA levels below detection level after 4 months of treatment. An increase in CD4 count was observed after 2 months of treatment, but no further increase was observed after 4 months. There was no apparent effect of ART on hemoglobin levels, but a significant increase in body mass index was observed after 4 months of treatment.
Plasma concentrations of IL-1ra, IL6, IL-8, IL-10, TNF-α, and MCP-1 were measured in plasma samples from HIV-uninfected controls and HIV-infected individuals stratified on the basis of CD4 cell count and in patients included in the follow-up after 2 and 4 months of ART (Fig. 1). HIV-infected individuals showed elevated levels of the cytokines TNF-α, IL6, IL-8, MCP-1, and IL-1ra. HIV progression had an effect on cytokines, with significantly higher levels of TNF-α, IL-6, IL-8, MCP-1, and IL-1ra in the group of patients with CD4 counts less than 200 cells per microliter. All cytokine levels decreased after 2 months of ART. A further decrease in cytokine levels of the cytokines TNF-α, IL-6, and IL-1ra was observed after 4 months of treatment.
Plasma cytokine levels in HIV-infected individuals after 4 months of ART were found to be lower than cytokine levels in HIV-infected individuals with high CD4 cell count and approaching cytokine levels of HIV-uninfected individuals (Fig. 1). Cytokine levels were only slightly lower in HIV uninfected vs. 4 month of ART: TNF-α mean 1.3 vs. 1.52 pg/μL, P = 0.4; IL-1ra mean 110 vs. 152 pg/μL, P = 0.6; IL-8 mean 1.0 vs. 1.5 pg/μL, P = 0.3; MCP-1 mean 33 vs. 31 pg/μL, P = 0.8; IL-10 median 0.56 vs. 0.33 pg/μL, P = 0.7; and IL-6 median 0.35 vs. 0 pg/μL, P = 0.9.
Associations Between Cytokines, HIV RNA, Hemoglobin, and CD4 Cells
Univariate and multivariate linear regression analyses in addition to Spearman correlations are presented in Table 2. In univariate analysis, all cytokines correlated positively with viral load and all cytokines, but IL-10, correlated negatively with CD4 cell count. IL-6 and IL-1ra correlated negatively with hemoglobin. In multivariate analysis including sex and age adjustments, higher viral load predicted higher levels of TNF-a, IL-8, and IL-1ra. Lower CD4 cell counts and hemoglobin levels were also significant predictors for higher IL-1ra levels (Table 2).
We found that circulating levels of the cytokines TNF-α, IL-6, and IL-1ra and chemokines MCP-1 and IL-8 were elevated in HIV-infected individuals. This is in accordance with data from studies conducted in Europe, North America, and Africa1,6-9 showing elevated cytokine and chemokine levels in HIV-infected patients. In line with the results from 2 cross-sectional6,8 and 1 longitudinal studies7 investigating the effect of HIV progression on cytokines levels in African patients, we found a strong correlation between HIV progression and elevated levels of cytokines. However, one cross-sectional study reported lower cytokine levels in AIDS patients compared with less progressed HIV patients.9
We found that all cytokines correlated positively with viral load. Furthermore, viral load was the strongest predictor of cytokine levels. Indeed, viral load was an independent predictor of all analyzed cytokines in multivariate analysis. One study conducted in an African setting showed a positive univariate correlation between soluble TNF receptor-rII and HIV RNA.8 However, to our knowledge, this is the first time that a strong positive association between both pro- and anti-inflammatory circulating cytokines and viral load has been found in African HIV-infected individuals.
Our study showed that ART efficiently reduced plasma levels of circulating cytokines. Similar results have been reported from high income countries, where ART has been shown to decrease levels of pro-inflammatory cytokines.10,11 However, this has not previously been shown in sub-Saharan African individuals. In addition, our study indicates that ART reduced circulating cytokines to levels close to those in HIV uninfected patients. As ongoing inflammation is believed to be a driving factor for HIV-related immunodeficiency, our findings suggest that immune reconstitution can be achieved in highly immune-activated African patients. A study conducted in a sub-Saharan African setting showed no decline in viral load after antituberculosis treatment, and systemic levels of TNF-α remained high.12 This is in line with our previous13 and present results that suggest that viral replication itself is the primary driving factor of immune activation, even in an environment with an abundance of coinfections. Our study suggests that ART is very efficient in decreasing immune activation, which may possibly be a more rational approach than eradication of coinfection. However, because of the small sample size after 4 months of ART and a possible differential loss to follow-up of patients receiving ART in our study, we cannot definitively exclude clinically significant elevations in inflammatory cytokines after 4 months of ART. Further studies are needed to investigate the level of immune reconstitution after ART.
In conclusion, findings from this study demonstrated high levels of cytokines among HIV-infected patients with highest levels occurring in progressed individuals. The data support that HIV replication by itself causes increased cytokine levels. A rapid decrease of elevated cytokine levels after ART is reassuring for the long-term immune reconstitution during ART in an African environment where individuals are exposed to an abundance of other infectious agents.
All study participants are thanked deeply for their participation. Furthermore, we wish to acknowledge the counselors, laboratory, and clinical staff at TAWG; The National Institute for Medical Research, The Tanga Centre, The Care and Treatment Clinic, The Bombo Regional Hospital, and The Department of Immunology as well as the AIDS laboratory.
1. Appay V, Sauce D. Immune activation and inflammation in HIV-1 infection: causes and consequences. J Pathol
2. Bentwich Z, Kalinkovich A, Weisman Z. Immune activation is a dominant factor in the pathogenesis of African AIDS. Immunol Today
3. Rizzardini G, Trabattoni D, Saresella M, et al. Immune activation in HIV-infected African individuals. Italian-Ugandan AIDS cooperation program. AIDS
4. Lawn SD. AIDS in Africa: the impact of coinfections on the pathogenesis of HIV-1 infection. J Infect
5. Autran B, Carcelain G, Debre P. Immune reconstitution after highly active anti-retroviral treatment of HIV infection. Adv Exp Med Biol
6. Ayehunie S, Sonnerborg A, Yemane-Berhan T, et al. Raised levels of tumour necrosis factor-alpha and neopterin, but not interferon-alpha, in serum of HIV-1-infected patients from Ethiopia. Clin Exp Immunol
7. Chollet-Martin S, Simon F, Matheron S, et al. Comparison of plasma cytokine levels in African patients with HIV-1 and HIV-2 infection. AIDS
8. Messele T, Brouwer M, Girma M, et al. Plasma levels of viro-immunological markers in HIV-infected and non-infected Ethiopians: correlation with cell surface activation markers. Clin Immunol
9. Thea DM, Porat R, Nagimbi K, et al. Plasma cytokines, cytokine antagonists, and disease progression in African women infected with HIV-1. Ann Intern Med
10. Bisset LR, Rothen M, Joller-Jemelka HI, et al. Change in circulating levels of the chemokines macrophage inflammatory proteins 1 alpha and 11 beta, RANTES, monocyte chemotactic protein-1 and interleukin-16 following treatment of severely immunodeficient HIV-infected individuals with indinavir. AIDS
11. Stylianou E, Aukrust P, Kvale D, et al. IL-10 in HIV infection: increasing serum IL-10 levels with disease progression-down-regulatory effect of potent anti-retroviral therapy. Clin Exp Immunol
12. Lawn SD, Shattock RJ, Acheampong JW, et al. Sustained plasma TNF-alpha and HIV-1 load despite resolution of other parameters of immune activation during treatment of tuberculosis in Africans. AIDS
13. Erikstrup C, Kallestrup P, Zinyama R, et al. Predictors of mortality in a cohort of HIV-1-infected adults in rural Africa. J Acquir Immune Defic Syndr
© 2009 Lippincott Williams & Wilkins, Inc.