Development of TB in HIV infected patients is based on a predisposition to reactivation of latent M. tuberculosis infection and to susceptibility to primary progressive M. tuberculosis infection . However, the relationship of host immune responses to the development of TB during different stages of HIV disease is not clear. The opportunistic behavior of M. tuberculosis during human HIV infection can be explained by suppression of type-1 responses at the level of antigen-presenting cells, CD4 T cells and effector macrophages.
In vitro studies have shown that lowering of intracellular GSH levels decreases cell survival, alters T cell functions and increases HIV replication, NF-kB activation, and sensitivity to TNF-α induced cell death [10, 11, 19]. A role has also been proposed for GSH as a carrier molecule for NO. Nitric oxide also reacts with GSH to form GSNO, an NO donor with greater stability [34, 35].
We first reported that GSH facilitates the control of intracellular M. bovis BCG in NO-deficient macrophages derived from iNOS knock out mice, and in HMDM . These studies indicated that GSH has direct antimycobacterial activity distinct from its role as an NO carrier. Furthermore, in our recent studies we demonstrated that GSH is vital for growth control of intracellular H37Rv in J744.1 macrophages .
It has been reported that production of IFN-γ is crucial to the control of M. tuberculosis infection . Impaired production of IFN-γ correlates with progression of immunodeficiency and is likely related to abnormalities in the IL-12-IFN-γ axis [8, 31]. We therefore tested the growth of H37Rv in HMDM from healthy subjects that are unstimulated or stimulated in vitro with IFN-γ, LPS. We observed a significant, four-fold increase in growth of H37Rv inside unstimulated HMDM, between 1 h and 7 days (Fig 1a). Stimulation of H37Rv-infected HMDM cells with IFN-γ, LPS also resulted in a three-fold increase in growth of intracellular H37Rv (Fig 1a). Since our earlier studies suggested a role for GSH in innate immunity against M. tuberculosis, we tested whether NAC treatment would induce HMDM to inhibit the growth of H37Rv. We observed that NAC at 10 mM concentration induced growth inhibition of H37Rv in three out of six healthy individuals tested (Fig 1b). Although normal levels of GSH are present in cells derived from healthy subjects, those levels might decrease during oxidative and nitrosative stress generated during TB infection. Therefore, addition of NAC to HMDM caused growth inhibition of M. tuberculosis by augmenting intracellular GSH levels. These results suggest that growth inhibition of H37Rv in NAC treated HMDM is due to the direct antimycobacterial effects of GSH. Furthermore, the inability of HMDM from some healthy individuals to inhibit M. tuberculosis growth is probably due to the inability of macrophages to maintain adequate GSH levels, despite NAC treatment.
As described before, innate and adaptive immunity are essential for successful elimination of M. tuberculosis. Macrophages interact with other immune cells in vivo, for successful growth retardation of M. tuberculosis. The whole blood model of infection resembles an in vivo system in promoting cellular interactions. This model differs from other intracellular infection models in that all blood elements are represented. Infection of blood cultures from healthy volunteers with H37Rv resulted in an almost two-fold increase in H37Rv growth (Fig 1c). The increase in H37Rv growth was statistically significant. In contrast to HMDM, treatment of blood cultures with NAC (10 mM) caused growth inhibition of H37Rv, in all seven individuals tested (Fig 1c). Our results suggest that growth inhibition of H37Rv in NAC treated blood cultures is due to direct antimycobacterial effects of GSH and due to activation of blood cells induced by GSH.
We have confirmed the work of others that GSH levels are decreased in patients with HIV-1 infection [5, 11, 14, 23], and then hypothesized that this decrease would be associated with reduced capacity of monocytes to kill intracellular M. tuberculosis. We further proposed that NAC treatment would improve the killing of M. tuberculosis. We tested our hypothesis by determining GSH levels in healthy and HIV positive subjects. We observed a significant and more than 50% decrease in GSH levels in PBMC and RBC from HIV patients compared to healthy subjects (Fig 2a, 2b). Since GSH enhances innate and adaptive immune functions, GSH deficiency in PBMC may contribute to the progressive immune dysfunction of HIV infection. Macrophages play a central role in HIV and TB infection because they are among the first cells to be infected . Moreover, macrophages serve as an important reservoir for both HIV and M. tuberculosis. The major obstacle to eradication of HIV is latent virus in these reservoirs which has prompted the search for new drugs and strategies to protect this cell compartment. Erythrocytes have been used as a carrier system to deliver antiretroviral molecules to macrophages selectively. Fraternale et al  have reported that treatment of mice with AZT+DD1+GSH-loaded RBC significantly reduces the proviral DNA content, compared to mice treated with AZT+DD1. This result is consistent with our hypothesis and suggests that low levels of GSH in RBC, as observed in this and other studies, will affect the GSH carrier functions of RBC, compromising GSH delivery to macrophages.
In order to determine the effects of NAC treatment on PBMC and RBC in reducing the growth of intracellular H37Rv, whole blood cultures from HIV patients were treated in vitro with NAC and infected with H37Rv. We observed significant growth of H37Rv in unstimulated blood cultures from HIV patients (Fig 3a). In vitro NAC treatment to blood cultures derived from HIV subjects caused inhibition in growth of intracellular H37Rv (Fig 3b). Furthermore, BSO treatment abrogated the inhibitory effect brought about by NAC treatment (Fig 3c). This suggests that restoration of GSH levels in HIV subjects caused enhancement in immune cell functions to contain M. tuberculosis growth.
The decreased GSH content in immune cells of HIV-positive individuals was atleast in part attributed to the decreased in plasma cysteine and increased plasma glutamate (an inhibitor of cysteine permeation via the Xc- transport system), as observed during early infection. The decreased intracellular GSH and plasma cysteine observed in HIV patients is due to chronic oxidative stress, which may lead to the progression of the disease. The decreased availability of cysteine can be overcome to some extent by the cysteine precursor NAC . A recent report of a carefully conducted clinical trial indicates that NAC treatment improves the clinical situation and delays the HIV disease progression . This study showed that long-term administration of NAC to AIDS patients improves their hematological profile, GSH content and life expectancy .
We measured cytokine levels in whole blood culture supernatants from healthy and HIV infected subjects. No clear trend in cytokine profile was observed in healthy subjects. Interestingly, we observed that in vitro infection with H37Rv induced the whole blood cultures from HIV patients to synthesize increased levels of cytokines such as IL-1, TNF-α, IL-6 and IL-10 (Fig 4, 5). IL-1, TNF-α, IL-6 are the early pro-inflammatory cytokines produced by monocytes after various bacterial infections and share a wide array of biological activities [4, 5]. In vitro studies have shown that mycobacterial preparations, including lipoarabinomannan, can cause the release of TNF-α and IL-1 from human PBMC [25, 42, 44].
The release of pro-inflammatory cytokines after mycobacterial infection is a host immune response that may be propitious or deleterious to the host. Newman et al. reported that increased survival of M. avium intracellulare (MAI) in isolated macrophages is correlated with the efficiency with which TNF-α and IL-6 are produced in response to MAI infection . Nevertheless, increased levels of these pro-inflammatory cytokines may be disadvantageous to the host because they not only cause acute-phase events, such as fever, but also mediate cachexia, hemorrhagic necrosis and lethal shock [29, 30, 37]. TNF-α by classical cascade is known to up-regulate the levels of IL-1 and IL-6.
Elevated levels of IL-6 are present in plasma of patients with TB . Studies by Van Heyningen et al  indicate that macrophages infected with M. bovis BCG released copious amounts of IL-6 which in turn inhibited the macrophage capacity to induce proliferation of CD4 T cell hybridoma. Nagabhushanam et al.  reported a novel function of IL-6 in inhibiting cellular immune response to eradicate M. tuberculosis infection. Their studies show that IL-6 produced by M. tuberculosis-infected macrophages selectively inhibited macrophage responses to IFN-γ. In other words, secretion of IL-6 by M. tuberculosis-infected macrophages may contribute to the inability of IFN-γ to eradicate M. tuberculosis infection .
The high levels of IL-6 released by infected macrophages have implications for co-infection with HIV . Mycobacterial infections are one of the most common AIDS-defining illnesses and may even accelerate progression to AIDS . The two infections seem to synergize, causing a shift of the host-pathogen balance in favor of the pathogen, which cannot be reversed by treatment with antimycobacterial agents .
TNF-α and IL-6, as well as IL-1, can increase HIV replication [3, 21]. Thus, decreasing the pro-inflammatory cytokine production in vivo may enhance the control of viral replication. Elevated levels of IL-6, TNF-α and IL-10 have been described previously in cases of advanced HIV disease [1, 20, 22]. Therefore, increases in the levels of pro-inflammatory cytokines will cause a positive feedback loop in which the two infections complement one another, leading to accelerated progression of both diseases.
In our studies, we observed that NAC treatment caused down-regulation of the synthesis of IL-1, IL-6, and TNF-α (Fig 4a, 4b, 4c), and up-regulation of the synthesis of IFN-γ (Fig 4d). These results suggest that GSH might have a crucial role in vivo in reducing the levels of pro-inflammatory cytokines thereby protecting the host against disease progression.
Active TB is associated with suppression of T cell responses  and enhanced production and activity of immunosuppressive such as IL-10. IL-10 has been shown to be produced by macrophages infected with mycobacteria. IL-10 and TGF-β overlap with each other in many of their biological effects including, inhibition of T cell proliferation and IFN-γ production . Elevated levels of IL-10 in serum during advanced HIV infection may enhance immune suppression, allowing opportunistic infections . In our studies, we observed that NAC treatment decreased the levels of IL-10 favoring immune activation (Fig 5b).
We demonstrate growth inhibition of intracellular H37Rv in our in vitro studies using NAC-treated blood cultures from HIV patients. Furthermore, treatment of blood cultures with NAC modulated the production of cytokines in favor of the host. As described in the model (Fig 6a), our results strongly indicate that the immune cell enhancing and antimycobacterial functions of GSH are important for growth control of H37Rv in blood cultures from healthy and HIV-infected subjects (Fig 6a). Additionally, NAC treatment down-regulated the synthesis of IL-10 and pro-inflammatory cytokines in blood cultures from HIV-infected subjects favoring immune activation (Fig 6b). Current interventions to prevent tuberculosis in areas where TB and HIV are endemic, such as sub-Saharan Africa, have serious limitations. ART is limited by its cost and by its requirement for a sophisticated health care delivery system. Isoniazid chemoprophylaxis has limited efficacy in regions of high TB transmission, particularly in highly susceptible individuals with advanced HIV infection. In addition, isoniazid is ineffective against INH-resistant TB strains, which may account for 10–20% of all cases in some areas. NAC is inexpensive and non-toxic (it is considered a food supplement in the US, and is available without prescription in health food stores). The findings from this study may lead to long-term research that will be of potential importance for control of TB worldwide.