SEARCH

SEARCH BY CITATION

Keywords:

  • anti-retroviral therapy;
  • highly active;
  • chemokine CCL19;
  • HIV;
  • Mycobacterium avium

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Disclosure
  9. References

Based on the ability to recruit lymphocytes and dendritic cells to lymphoid tissue and to promote inflammation, we hypothesized a role for dysregulated CCL19 and CCL21 levels in human immunodeficiency virus (HIV)-infected patients with advanced immunodeficiency, and in particular in those with accompanying Mycobacterium avium complex (MAC) infection. The hypothesis was explored by studies in HIV-infected patients with and without MAC infection, as well as in vitro, examining the ability of proteins from MAC to promote CCL19 and CCL21 responses in peripheral blood mononuclear cells (PBMC) during highly active anti-retroviral therapy (HAART). Our main findings were: (i) raised serum levels of CCL19 in HIV-infected patients with CD4+ T cell count <50 cells/µl compared with HIV-infected patients with CD4+ T cell count >500 cells/µl and healthy controls, with particularly high levels in those with MAC infection; (ii) elevated plasma levels of CCL19 predicted a higher mortality in acquired immune deficiency syndrome (AIDS)-patients, independent of ongoing MAC infection; and (iii) marked production of CCL19 in MAC-stimulated peripheral blood mononuclear cells (PBMC) and pronounced disturbances in MAC-induced CCL19 production in PBMC from HIV patients that was partly reversed during HAART. Our findings suggest the involvement of CCL19 in AIDS patients with advanced immunodeficiency, potentially mediating both adaptive and maladaptive responses.


Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Disclosure
  9. References

The homeostatic chemokines CCL19 and CCL21 and their cognate receptor CCR7 are expressed constitutively within secondary lymphoid tissue, regulating lymphocyte and dendritic cell (DC) homing [1,2]. However, more recent studies have revealed roles for CCR7 and its ligands in inflammation and T cell homing into non-lymphoid tissue, implying a more complex role of CCR7 in immune responses [3,4]. An imbalanced regulation of CCL19 and CCL21 seems to be involved in the pathogenesis of various inflammatory disorders, including rheumatoid arthritis, inflammatory bowel diseases and atherosclerosis [5–7].

There are some studies on CCL19/CCL21/CCR7 expression during HIV infection. Choi et al. have reported that simian immunodeficiency virus (SIV) infection altered the expression of these chemokines in lymphoid tissue [8]. CCL19 and CCR7 expression are increased in lymph nodes during acute infection before reduction to levels found in overt immunodeficiency, whereas CCL21 expression decreased progressively throughout the infection. Moreover, we have recently shown increased serum levels of CCL19 and CCL21 in relation to disease progression in human immunodeficiency virus (HIV)-infected patients [9]. However, there are limited data on CCL19/CCL21 regulation during opportunistic infections in HIV-infected patients with advanced immunodeficiency.

Mycobacterium avium complex (MAC) causes disseminated infection in immunocompromised hosts [10]. Key defence mechanisms against MAC and other mycobacteria include generation of antigen-specific T cells and recruitment of monocytes and macrophages, which form granulomatous lesions at the site of mycobacterial replication, representing an essential mechanism for control of pathogen growth and persistence [11]. HIV-infected patients who develop disseminated MAC infection usually have markedly decreased CD4+ T cell counts (i.e. below 50 cells/µl) [10]. However, while the participation of CD4+ T cells in preventing rapid progression of MAC-related disease is well established, little is known regarding the role of various soluble mediators such as cytokines and chemokines in disease progression of MAC infection in HIV-infected patients.

Based on their ability to recruit lymphocytes and DCs to lymphoid tissue as well as to promote inflammation, we hypothesized a role for dysregulated CCL19 and CCL21 levels in HIV-infected patients with advanced immunodeficiency, and in particular in those with accompanying MAC infection. In the present study, this hypothesis was explored by various experimental approaches including clinical studies in HIV-infected patients with and without accompanying MAC infection, as well as in vitro studies, examining the ability of proteins from MAC to promote CCL19 and CCL21 responses in peripheral blood mononuclear cells (PBMC) before and after initiating highly active anti-retroviral therapy (HAART).

Methods

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Disclosure
  9. References

Patients and controls

In the pre-HAART era a cohort of 446 HIV-infected patients was followed during a period of 6 years with repeated blood sampling and clinical evaluation (mean follow-up 13·4 months, four samples per patient). From this cohort, we present data from 38 acquired immune deficiency syndrome (AIDS) patients with MAC infection (29 males and nine females, 35 ± 8 years). All these patients had a minimum of one blood culture that was positive for MAC. We present data from serum samples taken prior to diagnosis (3–6 months, baseline), from the time of MAC diagnosis and from follow-up samples (3–6 months after diagnosis). Mean survival time after diagnosis was 10 months. Twenty-eight AIDS patients without MAC infection (22 males and six females, 36 ± 9 years) were also followed longitudinally during the same period. The two AIDS groups were matched for CD4+ T cell counts (<50 × 106/l), age, anti-retroviral treatment (i.e. zidovudine and lamivudine) and Pneumocystis jiroveci prophylaxis. In the non-MAC group, 13 patients had opportunistic infection or cancer at the time of the second blood sampling, corresponding to samples taken at the time of MAC diagnosis in the MAC group [P. jiroveci pneumonia (n = 5), cytomegalovirus (CMV) disease (n = 6) and lymphoma (n = 2)]. For comparison, serum samples were also obtained from 20 sex- and age-matched asymptomatic HIV-infected patients with CD4 T cell counts >500 × 106/l and 20 sex- and age-matched healthy controls. In the separate substudies, PBMCs were obtained from additional six healthy controls and nine HIV-infected patients (none with MAC infection) before and 4 and 24 weeks after initiating HAART. Informed consent for blood sampling was obtained from all subjects. The study was approved by the local ethical committee.

Blood sampling protocol

Peripheral venous blood was drawn into sterile blood collection tubes without any additives (Becton Dickinson, San Jose, CA, USA), immersed immediately in melting ice and allowed to clot before centrifugation (1000 g for 10 min). Serum specimens were stored at −80°C and were thawed fewer than three times.

Isolation and stimulation of PBMC

PBMC, obtained from heparinized blood by density gradient centrifugation (Lymphoprep; Nycomed Pharma, Oslo, Norway), were incubated in flat-bottomed 96-well trays (2 × 106/ml; Costar, Cambridge, MA, USA) in medium alone [RPMI-1640 with 2 mm l-glutamine and 25 mm HEPES buffer (PAA Laboratories, Pasching, Austria) supplemented with 10% fetal calf serum (PAA Laboratories)] or stimulated with recombinant human CCL19 (100 ng/ml; R&D Systems, Minneapolis, MN, USA), recombinant human CCL21 (100 ng/ml; R&D Systems), purified-protein derivative from MAC (PPD-MAC; 2 mg/ml; gift of R. Bjørlo, National Veterinary Institute, Oslo, Norway), recombinant HIV-1 tat protein (HIV-tat; 1 µg/ml; ImmunoDiagnostics, Woburn, MA, USA), recombinant HIV-1 gp120 protein (HIV-gp120; 1 µg/ml; ImmunoDiagnostics), recombinant HIV-1 p24 protein (HIV-p24; 1 µg/ml; ImmunoDiagnostics) and lipopolysaccharide (LPS) from Escherichia coli 026:B6 (1 µg/ml; Sigma, St Louis, MO, USA). Cell-free supernatants were harvested and stored at −80°C after culturing for 20 h. The concentrations of the various stimulants that were used in the present study were based on dose – response experiments.

Enzyme immunoassays (EIA)

Concentrations of CCL19 and CCL21 were measured in duplicates by EIA obtained from R&D Systems.

Statistical analysis

For comparisons of two groups of individuals, the Mann – Whitney U-test was used. When more than two groups were compared, the Kruskal – Wallis test was used. If a significant difference was found, the Mann – Whitney U-test was used to determine the differences between each pair of groups. The Wilcoxon signed-rank test was used for comparison of paired data. Kaplan – Meier curves were generated and the log-rank test was used to compare mortality rates in relation to CCL19 and CCL21 levels (both as tertiles) independent of MAC infection. To test if any association between chemokine level and mortality was independent of MAC infection, a Cox regression was performed including CCL19, CCL21 (top tertile versus the two lower) and MAC status. In the text, the data are given as mean ± standard error of the mean (s.e.m.) if not stated otherwise. Probability is two-sided and considered to be significant when P < 0·05.

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Disclosure
  9. References

Serum levels of CCL19 and CCL21: influence of MAC infection

Several significant findings were revealed when examining serum levels of CCL19 and CCL21 in 38 AIDS patients with MAC infection and in 28 AIDS patients with comparable immunodeficiency without MAC infection. First, the two groups of AIDS patients had markedly raised serum levels of CCL19 compared with asymptomatic HIV-infected patients (n = 20) and healthy controls (n = 20), with particularly high levels in those with MAC infection (Fig. 1a). Secondly, all HIV-infected patients had raised serum levels of CCL21 compared with healthy controls, but in contrast to CCL19 there were no differences between AIDS patients with and without MAC infection, and the difference between AIDS patients and patients with asymptomatic HIV infection did not reach statistical significance (Fig. 1b). Thirdly, the two AIDS groups were followed longitudinally (see Methods, Fig. 2), and whereas patients with AIDS who developed MAC-related disease showed a marked increase in CCL19 before the time of MAC diagnosis, followed by a significant decrease, AIDS patients without MAC infection showed no significant changes in CCL19 levels during the observation period (Fig. 2a). In contrast, both subgroups showed a significant decrease in CCL21 from baseline during diagnosis and follow-up (Fig. 2b). Finally, 13 patients in the non-MAC group also had opportunistic infection or cancer at the time of blood sampling included in Fig. 1 and at time-point 2 in Fig. 2[P. jiroveci pneumonia (n = 5), CMV disease (n = 6) and lymphoma (n = 2)]. In contrast to the patients with systemic MAC infection, serum levels of CCL19 (and CCL21) in these patients did not differ from CCL19 (and CCL21) levels in the other patients in the non-MAC group (data not shown).

image

Figure 1. CCL19 and CCL21 in human immunodeficiency virus (HIV)-infected patients with or without infection with Mycobacterium avium complex (MAC). Serum levels of CCL19 and CCL21 were measured in HIV-infected patients with CD4+ T cell count <50 cells/µl with (n = 38) and without (n = 28) diagnosed MAC infection. For comparison, CCL19 and CCL21 were also measured in HIV-infected patients with CD4+ T cell count >500 cells/µl (n = 20) and healthy controls (n = 20). *P < 0·05 and ***P < 0·001 versus healthy controls. #P < 0·05 and ##P < 0·01 versus HIV-infected patients with CD4+ T cell count >500 cells/µl. ≠P < 0·05 versus HIV-infected patients with CD4+ T cell count below 50 cells/µl without MAC infection. Data are mean ± standard error of the mean.

Download figure to PowerPoint

image

Figure 2. Serum levels of CCL19 and CCL21 during longitudinal testing in human immunodeficiency virus (HIV)-infected patients with Mycobacterium avium complex (MAC) infection. The panels show serum levels of CCL19 (a) and CCL21 (b) in 66 acquired immune deficiency syndrome (AIDS) patients (all with CD4+ T cell counts <50 × 106/l) with (n = 38) and without (n = 28) MAC infection. Serum samples were collected at baseline (3–6 months prior to diagnosis of MAC infection), at the time of MAC diagnosis and at follow-up (3–6 months after diagnosis). Blood samples were taken at comparable time-points in the non-MAC group. *P < 0·05 and **P < 0·01 versus those without MAC infection. #P < 0·05 versus baseline; ≠P < 0·05 versus time of MAC diagnosis. Data are mean ± standard error of the mean.

Download figure to PowerPoint

Serum levels of CCL19 and CCL21 as prognostic markers in advanced HIV infection

During a mean follow-up of 13·4 months, 59 patients died [21 without MAC infection and 38 (all) with MAC infection]. Figure 3a,b shows the Kaplan – Meier curves according to CCL19 and CCL21 concentrations at baseline (3–6 months prior to MAC diagnosis in the MAC group), indicating a higher mortality rate in subjects with high CCL19 levels. Furthermore, the Cox regression analysis suggests that the association of CCL19 on a poorer survival was independent of MAC infection [MAC, hazard ratio (HR) = 2·22, 95% confidence interval (CI) 1·24–3·95; CCL19, HR = 1·96, 95% CI 1·10–3·49]. Thus, it seems that high CCL19 levels may predict mortality in AIDS patients with advanced immunodeficiency independent on ongoing MAC infection. In contrast to CCL19, baseline levels of CCL21 were not associated with mortality (Fig. 3b).

image

Figure 3. Kaplan – Meier curves according to serum concentrations of CCL19 and CCL21. During a mean follow-up of 13·4 months, 59 patients died [all Mycobacterium avium complex (MAC)-infected patients (n = 38) and 21 acquired immune deficiency syndrome (AIDS) patients without MAC infection]. (a,b) Kaplan – Meier curves demonstrating the cumulative incidence of death during the observation period across tertiles (T1–3) according to CCL19 (T1: 87–348 pg/ml; T2: 353–487 pg/ml; T3: 492–1129 pg/ml) (a) and CCL21 (T1: 577–1218 pg/ml; T2: 1281–1447 pg/ml; T3: 1458–1829 pg/ml) (b) concentrations at baseline (3–6 months prior to diagnosis of MAC infection). Similar results were found using dichotomized CCL19 levels, and a more pronounced trend was observed comparing the top tertile with the two lower, P = 0·008.

Download figure to PowerPoint

CCL19 levels in PBMC: effects of various inflammatory stimuli

To elucidate further the relation between CCL19 and MAC infection, we examined the release of CCL19 in PBMC from healthy controls (n = 6) that had been exposed to MAC-PPD, LPS or various HIV antigens. As can be seen in Fig. 4, MAC-PPD induced a marked release of CCL19 in PBMC after culturing for 20 h. LPS and HIV-tat protein, but not HIV-p24 and HIV-gp120 proteins, also induced a significant CCL19 release, but the responses were not so pronounced as for the MAC-PPD-induced CCL19 release (Fig. 4). There were no detectable levels of CCL21 in either unstimulated or stimulated cells (data not shown).

image

Figure 4. Effects of various inflammatory stimuli on CCL19 production in peripheral blood mononuclear cells (PBMC). The effect of lipopolysaccharide (LPS; 1 µg/ml), purified-protein derivative from Mycobacterium avium complex (MAC); final dilution 1 : 100), human immunodeficiency virus-1 (HIV-1) gp120 protein (gp120; 1 µg/ml), HIV-1 p24 protein (p24; 1 µg/ml) and HIV-1 tat protein (tat; 1 µg/ml) on the release of CCL19 in peripheral blood mononuclear cells (PBMCs) from six healthy controls after culturing for 20 h. *P < 0·05 and ***P < 0·001 versus unstimulated (unstim) cells. Data are given as mean ± standard error of the mean.

Download figure to PowerPoint

CCL19 levels in PBMC: effects of HAART

In nine HIV-infected patients not included in the serum study, PBMCs were sampled before and 4 and 24 weeks after initiating HAART. With regard to serum levels, the spontaneous release of CCL19 in PBMC was raised significantly in HIV-infected patients compared to healthy controls (n = 6), with a significant decrease during HAART (Fig. 5a). Notably, an opposite pattern was seen for the MAC-PPD stimulated release of CCL19, with significantly decreased responses in HIV-infected patients compared to healthy controls that, at least partly, were reversed during HAART (Fig. 5b). CCL21 was not detected in PBMC from either patients or controls, independent of ongoing HAART (data not shown).

image

Figure 5. The release of CCL19 in peripheral blood mononuclear cells (PBMC) from HIV-infected patients during highly active anti-retroviral therapy (HAART). The panels show spontaneous (a) and Mycobacterium avium complex (MAC)-purified protein derivative (PPD)-stimulated (final dilution 1 : 100) (b) release of CCL19 in PBMC supernatants after culturing for 20 h in six healthy controls (contr) and in nine HIV-infected patients before (basel.) and at different time-points (w, weeks) after initiating HAART.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Disclosure
  9. References

Herein we show that HIV-infected patients with advanced immunodeficiency were characterized by markedly increased serum levels of CCL19 and CCL21, with particularly high levels of CCL19 in AIDS patients with accompanying systemic MAC infection. We also present data suggesting that high levels of CCL19 may prognosticate an unfavourable outcome in AIDS patients with advanced immunodeficiency, independently of ongoing MAC infection.

The ability of CCL19 to predict unfavourable outcome in advanced HIV infection could reflect the pathogenic activity of several upstream inflammatory stimuli. Thus, in the present study we showed that MAC-PPD and HIV-tat protein, as well as LPS, also known to be raised in circulation in advanced HIV infection [12], promoted release of CCL19 in PBMCs. Previously, CCL19 and CCL21 have been shown to enhance HIV-1 replication in activated T cells [13]. Moreover, HIV-1 strongly upregulates the expression of CCR7 in plasmacytoid dendritic cells [14]. It is tempting to hypothesize that the enhanced CCL19 levels in relation to disease severity in AIDS patients with advanced immunodeficiency may be a marker of inappropriate and sustained immune activation associated with immune exhaustion in these patients.

In the present study we found higher CCL19 levels in AIDS patients with advanced immunodeficiency and accompanying MAC infection. Moreover, in contrast to the patients with systemic MAC infection, serum levels of CCL19 in patients with other opportunistic infections or cancer did not differ from CCL19 levels in the other patients in the non-MAC group. Although these data may suggest a link between CCL19 and MAC infection as opposed to other AIDS-related opportunistic infections, the lack of data from HIV-seronegative individuals with MAC infection and the lack of prospective data showing that high CCL19 may predict MAC infection underscore that these data should be interpreted with caution. In fact, our Kaplan – Maier curves suggest that high CCL19 levels predict mortality independent on ongoing MAC infection.

At a very late stage of MAC-infection, serum levels of CCL19 and CCL21 were found to decrease. One possibility is that this decrease is a response to MAC therapy. Conversely, studies in animal models have suggested a decline in CCL19 during advanced immunodeficiency as a result of immune exhaustion [8]. Moreover, we have previously shown declining CCL19 expression in HIV-infected patients with a rapid progression of disease without MAC infection [9]. Furthermore, while untreated HIV-infected patients with advanced immunodeficiency had enhanced spontaneous release of CCL19 in PBMC, cells from these patients had an attenuated CCL19 release upon MAC-PPD activation and, notably, this pattern was reversed by HAART. We have previously reported a similar pattern in phytohaemagglutinin-exposed mononuclear cells from HIV-infected patients [15,16]. While a persistently raised CCL19 level may be harmful in relation to HIV-related immunodeficiency, it is possible that the ability of mononuclear cells to mount an enhanced and balanced CCL19 response against MAC and other microbes could be protective. This may suggest the involvement of CCL19/CCL21 in both adaptive and maladaptive responses during HIV-related immunodeficiency and MAC infection, dependent at least partly on the response pattern (e.g. enhanced and balanced versus low-grade and persistent).

In the present study we found an association between serum levels of CCL19 and HIV progression, but not with serum levels of CCL21. The reason for this discrepancy is not clear at present, but could reflect different cellular sources of these two homeostatic chemokines. Thus, while CCL19 is produced by several types of cells such as T cells, monocytes and macrophages, CCL21 seems to be produced primarily by stromal cells, and in relation to advanced AIDS the former cells may be of particular importance. Moreover, although both chemokines act on a common receptor (CCR7) they could still mediate different effects, at least to some degree.

The present study has some limitations, such as the inclusion of relatively few patients, and in particular those with other opportunistic infections than MAC. The lack of the exact cause of death in the non-MAC group is another limitation. None the less, our finding may suggest the involvement of CCL19 in AIDS patients with advanced immunodeficiency, potentially mediating both adaptive and maladaptive responses.

Acknowledgements

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Disclosure
  9. References

This work was supported by grants from the Research Council of Norway and Oslo University Hospital Rikshospitalet.

Disclosure

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Disclosure
  9. References

The authors have no disclosures in relation to the article.

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Disclosure
  9. References