Tuberculosis (TB) remains the leading cause of morbidity and mortality due to any one infectious agent worldwide [1]. It has also been estimated that approximately 1·2 billion people harbour at least one species of intestinal parasite worldwide. In fact, 50–100 million people are infected with Strongyloides stercoralis, with a high prevalence in tropical regions of Africa, Asia and South America, particularly in Brazil [2–4]. It has been suggested that helminthic infections in humans are associated with a significant hyperactive humoral immunity and a depressed cellular immune response. Ultimately, this immunological dysregulation may facilitate concomitant infection or increase pathogenicity of other microbes; however, this is not fully understood. Co-infection with Mycobacterium tuberculosis (MTB) is common in the developing world; however, whether helminthic infection worsens immune responses to TB is unclear, as evidenced by recent studies [5–8].

Active TB is characterized by a profound and prolonged suppression of MTB-specific T cell responses, as evidenced by decreased production of the T helper 1 (Th1) cytokines, interleukin (IL)-2 and interferon (IFN)-γ [9–13]. Overproduction of immunosuppressive cytokines [IL-10 and transforming growth factor (TGF)-β] by mononuclear phagocytes has been implicated in decreased T cell function during TB [8,14–16]. Intestinal helminth infections in both human and experimental animal models are, generally, associated with a strong Th2-like immune response and significant suppression of Th1-type responses [2,6,7,17–20]. Considering that effective anti-MTB immunity is dependent upon an intact Th1-type immune response, it is possible that pre-existing infection with intestinal helminths may down-regulate the required Th1-type immune response via up-regulation of Th2-type cytokine production, thus ultimately facilitating progressive mycobacterial infection [5–7,21,22].

Active TB has been associated with the expansion of CD4⁺ CD25^high+ subset of CD4 T cells (regulatory T cells, T_reg) which may also participate, at least in part, in the suppression of MTB-specific T cell responses during active TB [23,24]. Interestingly, it has been demonstrated that T_reg frequencies are augmented in nematode infections [25,26] and that antigen-specific T_reg cells contribute to the immunosuppression associated with chronic onchocerciasis [25]. However, whether T_reg cells in dually infected TB + Helm patients are further expanded is still not known.

The present study was conducted to evaluate cellular phenotype and cytokine profiles in patients with newly diagnosed active pulmonary tuberculosis with or without concomitant intestinal helminth infection and to evaluate the effect of helminth infection on the clinical course of TB.

Table 1 Gender and age of tuberculosis (TB) patients, patients with both TB and intestinal helminth infection and healthy control subjects.

Concomitant intestinal helminth infection (TB + Helm patients) was found in 27·5% (11/40) of patients with TB. Parasitological findings showed that S. stercoralis was the most prevalent (72·7%) intestinal helminth among TB + Helm patients (eight of 11, data not shown).

Table 2 Differential white cell counts (cells/mm3) in blood from tuberculosis (TB) patients, patients with both TB and intestinal helminth infection and healthy control subjects.

To investigate whether the observed lymphopenia was a result of a proportional reduction in all lymphocyte subpopulations, we next enumerated absolute counts of B, T and natural killer (NK) T cells. CD19⁺ B cell counts were significantly lower in both TB and TB + Helm patients as opposed to controls (P < 0·001 for both; Fig. 1a), but comparable between both groups of TB patients. Numbers of CD3⁺ T cells also were lower in blood from both TB patient groups compared to controls (P < 0·004; Fig. 1b). However, in contrast to findings for B cells, absolute frequencies of CD3⁺ T cells from TB + Helm patients were significantly decreased when compared to TB patients (P < 0·001). The distribution of CD4⁺, CD8⁺ and NK⁺ T cell subsets followed the same pattern observed for CD3⁺ T cells (Fig. 1c–e). However, when NK cells were evaluated no significant differences were observed between the three groups (Fig. 1f). CD4⁺ T cells are considered pivotal for the establishment of an effective anti-TB immune response. Considering that their absolute numbers were reduced significantly among TB + Helm patients (Fig. 1c), we next evaluated their frequency at both time of diagnosis and completion of TB treatment (Fig. 2a) in both TB and TB + Helm patients. In contrast to findings for absolute numbers, both CD4⁺ and CD8⁺ T cell frequencies at baseline were comparable between TB patient groups. However, at completion of anti-MTB therapy (week 24 of treatment) CD4⁺ T cell frequencies were significantly lower in blood from TB + Helm compared to TB patients (P = 0·019), whereas CD8⁺ T cell frequencies were similar (data not shown). These data suggest that TB + Helm patients undergo a defect in their T cell compartment that remains sustained by completion of TB treatment.

Fig. 1 Absolute frequencies (cells/mm3) of T, B and natural killer (NK) cells and T cell subpopulations in blood from healthy control subjects, tuberculosis (TB) and TB and intestinal helminth infection (TB + Helm) patients. Results are expressed as mean average (more ...)

Fig. 2 (a) CD4+ T cell frequencies in blood from tuberculosis (TB) () and TB and intestinal helminth infection (TB + Helm) (□) patients at the beginning and at the completion of TB therapy. Cytokine profile (b) interferon (IFN)-γ and (c) interleukin (more ...)

As stated, IL-10 is induced strongly in response to intestinal helminth infections and also plays a role in immune suppression associated with active TB. In order to establish further the basis for depressed IFN-γ levels, we evaluated IL-10 immunoreactivity in culture supernatants in parallel. Interestingly, IL-10 levels were not significantly different in culture supernatants of TB and TB + Helm patients at baseline (Fig. 2c). However, at completion of treatment, IL-10 levels decreased significantly in culture supernatants from TB patients (P = 0·05) but remained elevated in supernatants from TB + Helm patients. These findings suggest that at least at the end of TB treatment, persistently low IFN-γ production may be a result of a sustained production of IL-10. Interestingly, IL-5 was not detectable in culture supernatants from either TB patients or TB + Helm patients (data not shown). Neither IL-4 nor TGF-β were evaluated in the current study. Preliminary data demonstrated that soluble IL-4 was not detectable in supernatants from WB cultures stimulated with MTB culture filtrate, and that TGF-β was not considered, because high background levels of TGF-β in WB cultures (released by platelets during degranulation) preclude meaningful analysis of antigen-induced changes in production of the cytokine.

Findings described above do not explain fully the basis for depressed IFN-γ production observed in supernatants of TB + Helm patients at the time of TB diagnosis. However, recently, another mode of immune suppression involving CD4 CD25^high T cells (i.e. T_reg) was described in patients with active pulmonary TB [24]. To determine whether T_reg were involved in regulation of IFN-γ production in WB cultures, we enumerated absolute counts of CD4 CD25^high T cells (Fig. 3). Surprisingly, although significantly higher when compared to healthy individuals (data not shown), numbers of T_reg were lower in TB + Helm compared to TB patients (P = 0·051), suggesting that T_reg suppressor function was probably not the only mechanism responsible for the observed low IFN-γ levels in MTB-CF-stimulated WB culture supernatants.

Fig. 3 CD4+ CD25high T cell absolute counts in blood from tuberculosis (TB) () and TB and intestinal helminth infection (TB + Helm) (□) patients at the beginning of TB therapy. Results are expressed as mean average ± s.d. Significant differences (more ...)

Research over the last decade has identified a multitude of mechanisms for depressed MTB-antigen-stimulated production of IFN-γ at the time of diagnosis of active TB [9,10,14–16,33]. However, mechanisms responsible for the prolonged delay in recovery of MTB-specific T cell IFN-γ production by peripheral blood mononuclear cells (PBMC) of TB patients [9] remain unclear. The impact of intestinal helminth infections on susceptibility to M. tuberculosis is still controversial. Recently, Brown et al. [34] reported that nematode infection and progression of active tuberculosis among HIV-1 infected patients were not correlated. However, a significant association between intestinal helminthic infections and mycobacterial diseases, such as pulmonary tuberculosis and multi-bacillary leprosy, has been demonstrated by several authors [8,17,35,36]. In the present study, 27·5% (11 of 40) of all patients with TB, were positive for at least one intestinal helminth, confirming previous findings [35,36]. Interestingly, infection with S. stercoralis was found in 72·7% (eight of 11) of TB + Helm patients. Infection with S. stercoralis has been associated with a strong Th2-type immune response [4].

Previous studies indicate that active pulmonary TB is associated with a reduction in absolute numbers of both circulating CD4 and CD8 T cells [23]. Further, lymphopenia has also been described in patients with chronic intestinal helminth infection [37]. Consistent with these findings, we found that T cell lymphopenia was even more pronounced among TB + Helm patients compared to TB alone. Therefore, it is possible that the significant decline in lymphocyte counts in TB + Helm patients may be the result of additive effects of both TB and intestinal helminthiasis on the host immune system. Surprisingly, and despite the fact that eosinophilia has been associated mainly with helminth infections, absolute counts of eosinophils were elevated and comparable in both TB and TB + Helm patients. One possible explanation for this finding could be that TB patients harboured occult helminth infection, despite the fact that all control subjects and patients underwent extensive examination for helminths in at least three consecutive stool samples investigated by the Lutz [29], Kato-Katz [30] and Baerman-Moraes [31,32] methods. However, in a previous study we have demonstrated that up to 25% of TB patients without intestinal helminths had eosinophil counts higher than 600 cells/mm³ [35]. Here, IL-5, an important cytokine in helminth-associated eosinophilia was undetectable in WB culture supernatants from both TB and TB + Helm patients. Recently, data from a longitudinal evaluation of immune responses during an experimental infection of a human volunteer with Necator americanus demonstrated that elevated eosinophilia following the primary infection (up to 815 days post-infection) occurred in the absence of detectable IL-5 production in WB cultures stimulated with worm-soluble antigens [38]. These authors suggested that focal presentation of adult worm antigens in the intestinal mucosa-associated lymphoid tissue may be able to maintain local IL-5 production by antigen-specific T cells, but these are not recruited to the systemic circulation.

TB immune suppression has been associated with both low IFN-γ production and depression of (MTB) antigen-driven T cell proliferative responses [9–14]. Interestingly, IFN-γ and lymphoproliferative responses to mycobacterial antigens appear to be suppressed in a correlated manner [14]. Thus, T cell responses could be assessed by antigen-specific IFN-γ production in WB culture.

Here, we used MTB-CF-stimulated WB cultures to extend these findings to patients with dual TB and intestinal helminthic infections. In fact, IFN-γ levels in CF-stimulated WB cultures from TB + Helm patients were fivefold lower compared to TB patients at baseline and remained depressed at the end of MTB treatment. While this reduction of cytokine production may be solely the result of lower absolute numbers of CD4⁺ T cells (Fig. 1c) in TB + Helm patients, it is also possible that this profound and persistent suppression of IFN-γ production may be due to the expansion of MTB-specific memory T cells primed to produce Th2-type cytokines [39]. Previous studies from our group support the concept that IL-10 and TGF-β mediate antigen hyporesponsiveness, as evidenced by suppression of antigen-driven blastogenesis and IFN-γ production, which are characteristics of tuberculosis [14,40]. In the current study, we found persistently elevated IL-10 levels in CF-stimulated WB culture supernatants from TB + Helm patients compared to TB patients at the end of TB treatment. As noted above, preliminary results from the current study showed the absence of detectable of IL-4 levels in WB culture supernatants form both TB and TB + Helm patients. This finding concurs with previous data from our group, showing that IL-4 was undetectable in PBMC culture from both TB patients and matched household contacts [14]. Whether this indicates that IL-4 does not play a role in suppression of IFN-γ production in our culture system or was not detected due to limitation in the immunoassay used is still unclear. Future studies are needed to resolve this question.

Another mechanism for depressed T cell function in TB, described recently, involves expansion of CD25^high CD4⁺ T cells with a regulatory phenotype (T_reg) [23,24]. We have demonstrated that the immunosupression observed normally during tuberculosis was associated with a threefold increase in the frequency of circulating T_reg cell in the blood of patients with TB when compared to healthy controls [23]. In the same study we also demonstrated that depletion of CD4 CD25^high cells from PBMC cultures led to an increase in IFN-γ levels in culture supernatants. Data from Guyot-Revol et al. [24] confirmed our findings and demonstrated that CD4 CD25^high expressed higher levels of forkheard box P3 (FoxP3) mRNA, suggesting that they were indeed T_reg cells [24].

Although monocytes–macrophages, T cells and B cells express IL-10 following activation in both human and experimental tuberculosis, its production has been associated mainly with monocytes [14,23,40]. However, it is possible that in the current study IL-10 production in WB cultures may be sustained by both T cells and monocytes. Specifically, it has been described that CD4 CD25^high (T_reg) cells produce substantial amounts of IL-10 in human tuberculosis [23,24] and in chronic helminth infection [25,26]. In a recent report, CD4 CD25^high T cells in mice infected with Schistosoma mansoni produced large amounts of IL-10 [26], a finding that may explain the cytokine profile in CF-stimulated WB culture supernatants in the current study described above. To investigate this possibility further, we enumerated absolute numbers of CD4 CD25^high T cells in blood from TB and TB + Helm patients. Interestingly, numbers of T_reg were lower in TB + Helm as opposed to TB patients. One possible explanation for decreased T_reg counts in peripheral blood of TB + Helm patients may be that these cells are sequestered at sites of both active helminth infection in the gut and the lung during active pulmonary TB. In addition, it is also possible that T_reg frequencies in these patients are lower due to the extension of pulmonary disease observed during TB + Helm infection, which would sequester more cells in the lung. Support for this hypothesis stems from our own recent study demonstrating compartmentalization of T_reg at sites of active TB in the lung [23]. In addition, Campanelli et al. [41] showed that functional T_reg cells accumulate in skin lesions from patients with cutaneous leishmaniasis. Thus, taken together, it is possible that in TB + Helm patients T_reg numbers are reduced due to sequestration of CD4 CD25^high T cells at more sites of active helminth and mycobacterial infections, such as gut and lung, respectively.

Although not statistically significant, we also observed a tendency of more advanced disease among patients with dual infections when compared to TB patients (data not shown). As an efficient clinical response to MTB is dependent upon an effective Th1 response, the more protracted clinical course in TB + Helm patients in association with increased IL-10 levels may indicate that intestinal helminth infection in TB patients may skew their cytokine profile towards a Th2 response. Data from other authors, demonstrating that chronic immune activation associated with intestinal helminth infections results in a predominant Th2-type cytokine profile, impaired signal transduction and anergy [7,19,20,42] and that a nematode-derived homologue of the human macrophage migration inhibitory factor (MIF) could interfere with host immunity systemically, shifting it towards a Th2-type immune response [43], support our hypothesis.

In summary, data presented here make a strong case for the negative impact of intestinal helminth infection on both anti-MTB immunity and clinical response to TB therapy. Timely diagnosis and treatment of intestinal helminths may be important for a successful response to anti-TB treatment. This may be critical to the success of the global effort to control TB.

We would like to thank the physicians (Drs Valderio Dettoni, Ricardo Tristão-Sá and David J. Hadad) and nursing staff at the NDI clinical research centre (CPC) at UFES, Vitoria, Brazil for their involvement in patient recruitment, follow-up and specimen collection. In addition, we appreciate the assistance by the personnel at the NDI microbiology laboratory with sputum processing and culture. Most importantly, however, we would like to acknowledge all TB patients and healthy volunteers in Vitória enrolled in the present study. Without their participation the project would not have been possible. This study was supported by funding from NIAID NIH (RFP AI-45244/AI-95383, Tuberculosis Research Unit).