key: cord-032504-tmohg1mj
authors: Auld, Sara C.; Staitieh, Bashar S.
title: HIV and the tuberculosis “set point”: how HIV impairs alveolar macrophage responses to tuberculosis and sets the stage for progressive disease
date: 2020-09-23
journal: Retrovirology
DOI: 10.1186/s12977-020-00540-2
sha: 
doc_id: 32504
cord_uid: tmohg1mj

As HIV has fueled a global resurgence of tuberculosis over the last several decades, there is a growing awareness that HIV-mediated impairments in both innate and adaptive immunity contribute to the heightened risk of tuberculosis in people with HIV. Since early immune responses to Mycobacterium tuberculosis (Mtb) set the stage for subsequent control or progression to active tuberculosis disease, early host–pathogen interactions following Mtb infection can be thought of as establishing a mycobacterial “set point,” which we define as the mycobacterial burden at the point of adaptive immune activation. This early immune response is impaired in the context of HIV coinfection, allowing for a higher mycobacterial set point and greater likelihood of progression to active disease with greater bacterial burden. Alveolar macrophages, as the first cells to encounter Mtb in the lungs, play a critical role in containing Mtb growth and establishing the mycobacterial set point. However, a number of key macrophage functions, ranging from pathogen recognition and uptake to phagocytosis and microbial killing, are blunted in HIV coinfection. To date, research evaluating the effects of HIV on the alveolar macrophage response to Mtb has been relatively limited, particularly with regard to the critical early events that help to dictate the mycobacterial set point. A greater understanding of alveolar macrophage functions impacted by HIV coinfection will improve our understanding of protective immunity to Mtb and may reveal novel pathways amenable to intervention to improve both early immune control of Mtb and clinical outcomes for the millions of people worldwide infected with HIV.

Although tuberculosis incidence declined worldwide over the course of the 20 th century [1] [2] [3] [4] , these gains were quickly reversed with the onset of the HIV epidemic in the 1980s [5] . Tuberculosis is now the leading global killer among infectious diseases and the leading cause of death among people with HIV [6] . The profound T cell depletion induced by HIV has long been recognized as a significant driver of the tuberculosis epidemic in people with HIV [7, 8] . However, there is growing awareness that HIV-mediated impairments in innate immunity also contribute to the increased risk of tuberculosis in this vulnerable population. These HIV-mediated impairments in innate immunity are particularly relevant early in Mycobacterium tuberculosis (Mtb) infection, prior to the onset of adaptive immune responses [9] and, by impairing early control of Mtb, set the stage for accelerated progression to active tuberculosis disease.

The pattern of early immune control setting the stage for longer term control-and clinically meaningful endpoints-parallels our understanding of the HIV set point.

It was first recognized in the late 1990s that the HIV viral load soon after seroconversion, i.e., the HIV set point, is predictive of disease progression [10, 11] . Following infection, HIV replicates unchecked until an initial, albeit ineffective, host immune response develops and the set point is established [12] [13] [14] . When viewed through this same lens, early host-pathogen interactions following Mtb infection can be thought of as establishing a "mycobacterial set point" (Fig. 1) . Unlike the set point in HIV, which occurs after an ineffective adaptive immune response, the mycobacterial set point is established before the onset of adaptive immunity.

Although we currently lack the tools to measure the mycobaterial set point clinically, this theoretical construct provides a framework for understanding early immune responses to Mtb, particularly those of the alveolar macrophage. While a multitude of cell types, including recruited interstitial macrophages, dendritic cells, and neutrophils, are infected prior to the onset of adaptive immunity [15] , work by Cohen and others has confirmed the importance of the alveolar macrophage to the early Mtb response [16] . As we will emphasize in this review, a focus on this critical early period and the mycobacterial set point are useful in understanding the relative impacts, both on a cellular and clinical level, of various host and pathogen impairments that occur in the context of coinfection with Mtb and HIV. The concept of a mycobacterial set point also provides us with an overall index of immune health in the context of HIV/Mtb coinfection and thereby offers a useful gauge of efficacy when specific pathways are manipulated for therapeutic benefit. Host immunity to Mtb remains poorly understood, but a focus on the role of alveolar macrophages in establishing a mycobacterial set point and on alterations in macrophage control in the context of HIV coinfection has the potential to yield great insight into some of the earliest immune events in Mtb infection.

Although Mtb-and HIV-mediated impairments in pulmonary innate immunity and alveolar macrophage function have been reviewed elsewhere [17, 18] , the impact of these impairments in the context of coinfection has not been well described. In this review, we will explore how HIV coinfection makes alveolar macrophages, the first cells to encounter Mtb, more permissive of early mycobacterial growth and is therefore likely to increase the mycobacterial burden at the point at which adapative immunity is activated and the mycobacterial set point is established [16] . By curtailing early immune control and delaying the onset of adaptive immunity, HIV coinfection leads to a higher mycobacterial set point and, consequently, a greater bacterial burden (Fig. 1) .

In order to demonstrate the utility of the mycobacterial set point framework, we will first review epidemiologic data supporting the importance of innate immunity in people with tuberculosis and HIV coinfection. We will then discuss some of the model systems that are used to study macrophage responses to Mtb and HIV and animal model findings in support of a mycobacterial set point. We will then highlight existing data on alveolar macrophage functions-from macrophage phenotype, key cytokines, and oxidative stress, to pathogen recognition and phagocytosis, cell death pathways, and activation Fig. 1 Graphical depiction of the mycobacterial set point as a theoretical construct, represented as the Mycobacterium tuberculosis (Mtb) bacterial burden after infection, either alone or in the context of HIV coinfection. After infection, Mtb bacterial burden increases until adaptive immunity is activated, thereby establishing the mycobacterial set point. Alveolar macrophages, the first cells to encounter Mtb, are an important component of the early innate immune response to Mtb. As reviewed in the main text, the mycobacterial burden at the point of adaptive immune activation plays a important role in the subsequent control of infection and progression to active tuberculosis disease. In the case of HIV coinfection, impairments in alveolar macrophage functions lead to poor early immune control of Mtb which, in turn, leads to a higher mycobacterial set point and greater Mtb bacterial burden at the time when adaptive immunity is activated of adaptive immunity-that are impacted by either Mtb infection, HIV infection, or coinfection. The examples we provide, which come from a range of cell types and organ systems, including primary alveolar macrophages, monocyte-derived macrophages, and cell lines, are not meant to be exhaustive or comprehensive. Rather, our goal is to provoke further inquiry and investigation into this critical area by offering a framework by which to contextualize studies in the field. We will close the review by highlighting several promising avenues for studying the mycobacterial set point in humans, in order to underscore the potential benefit of this concept as a framework for clinically relevant research.

From the earliest days of the HIV epidemic it has been clear that tuberculosis outcomes were markedly worse in people with HIV. In a series of nosocomial tuberculosis outbreaks among people with HIV in the early 1990s, the overwhelming majority of patients died, often within a month or two of their diagnosis [19] [20] [21] . These individuals had very advanced HIV, which undoubtedly contributed to their high mortality rates [21] . However, their rapid progression from exposure to active disease, in a time frame prior to the expected onset of adaptive immunity, suggests that impaired innate immunity was likely a contributing factor. Similarly, in a study of South African gold miners newly infected with HIV, the risk of tuberculosis disease was found to increase very early in the course of HIV infection, when CD4 counts would still be in a normal range [22] . Another study from South Africa found that people with HIV on antiretroviral therapy (ART) with normal CD4 counts had four times the risk of tuberculosis than people without HIV [23] . Likewise, in Italy, a setting with a low burden of both HIV and tuberculosis, nearly a third of patients diagnosed with HIV-associated tuberculosis in one study had been receiving ART for a median of 27 months, at which point CD4 counts would typically have reached a normal range [24] . Taken together, these data offer epidemiologic evidence that defects in both innate and adaptive immunity are likely to contribute to the greater risk of tuberculosis in people with HIV. While these clinical observations cannot establish mechanism or causality, they do provide a rationale for pursuing further basic science and translational studies to better understand the role of innate immunity in controlling Mtb infection.

Over the last century, scientists from successive eras have made remarkable discoveries by bringing the newest available techniques to bear on their investigations of Mtb, with various model systems employed over time.

While each system has distinct advantages and disadvantages in addressing particular questions, the sheer volume of data derived from the many available systems can make it hard to develop a coherent narrative regarding Mtb infection in humans. Although a full review of these features of the tuberculosis literature is beyond the scope of this review, we will offer a brief overview of the main characteristics of some of the major models to provide context for the reader [25, 26] .

Despite their inability to model sophisticated cellular interactions, in vitro systems have long been used as tools for studying Mtb [27] . A variety of myeloid cell lines have been employed, and are often helpful in quantifying cytokine and other signaling responses to Mtb. In addition, due to the ease with which in vitro systems can be manipulated, studies with primary macrophages from animal models have enabled investigators to dissect the roles of individual proteins and pathways in the Mtb response [28, 29] . Finally, human primary cells have been used in a variety of ways, including induced pluripotent stem cells, human macrophages, granuloma models, and many more. Each system has proven useful in addressing different aspects of the Mtb response, but all are hampered by their inability to coordinate the full spectrum of in vivo responses. An additional challenge is the conflation of different types of macrophages, whether from cell-lines, animal models, or humans, as a homogenous population. Rather, there is a growing appreciation of the distinctions between tissue-derived and bone-marrow derived macrophages, from their ontogeny to functional capacity and self-maintenance [30] [31] [32] . These distinctions are highly relevant to this review, as bone-marrow derived macrophages do not become polarized in the same manner as alveolar macrophages following Mtb exposure, and thus cannot be expected to behave the same [33] . However, given challenges in obtaining primary alveolar macrophages, particularly from humans [34] , many studies are conducted with derived cells.

With regards to animal systems, ease of use in handling the zebrafish-Mycobacterium marinum model has made it a powerful tool for the study of Mtb [35] . Several key components of the human response to Mtb are present in zebrafish, including the leukotriene A4 hydrolase locus, and it has been used to study everything from treatment strategies for drug-resistant Mtb to vaccine development. The mouse model has the virtue of capturing vertebrate, mammalian immune responses in a reproducible, easily scalable system. However, different mouse strains have different levels of susceptibility to Mtb and mice do not typically form granulomas or cavities [36] . As a result, the mouse model tends to be suited for studies of cell-mediated immunity and drug development, but less useful for studies of human clinical outcomes [37] . Guinea pigs provide another small animal model, with the benefit of recapitulating lung necrosis and other human features, but challenges in measuring their immune responses has historically limited their widespread use [25] . Rabbits have also provided an effective model system for Mtb research, largely due to the fact that the structural similarities between their lungs and human lungs leads to similar pathology in response to Mtb. That said, their larger size and the greater resources required to utilize them has limited their broad applicability [36, 38] . Finally, non-human primates, particularly rhesus and cynomolgus macaques, have marked genetic and immunologic similarities to humans and, as such, have yielded significant advances in our understanding of host-pathogen interactions and the potential for vaccineinduced protection in recent years [39, 40] . However, given the vast resources needed to sustain these animals, there are a limited number of research centers capable of conducting non-human primate studies.

While in-depth analysis of early immunologic events in human Mtb infection remains challenging for the reasons discussed above, Poulsen and others established over 50 years ago that tuberculin skin test (TST) conversion in humans, an indicator of the onset of the adaptive immune response, generally occurred by 6 weeks after exposure [41] [42] [43] . Poulsen also determined that this period prior to TST conversion is not a time of immunologic quiescence; rather, many individuals had evidence of a robust inflammatory response with fever and elevated inflammatory markers.

Since those landmark studies, much of our understanding of these early events has relied on data from animal models. Experiments in rabbits and guinea pigs by Lurie, Smith, and Harding, established that macrophages provide a niche within which Mtb multiplies logarithmically during early infection [44, 45] . Others have demonstrated that the onset of adaptive immunity in response to Mtb is indeed delayed, arising in part from delayed activation of CD4+ T cells within local lymph nodes [46, 47] . This delay of several weeks, as compared to several days with organisms such as Salmonella typhi and Listeria monocytogenes, translates into relatively unchecked bacterial growth during the early phase of infection [48, 49] . After this period of logarithmic growth, which has been demonstrated in rabbits, mice, and guinea pigs, the bacteria enter a relatively stationary phase [50] . During this critical time, Mtb subverts alveolar macrophage defenses, thereby ensuring prolonged intracellular survival and mycobacterial growth within this pulmonary niche.

Whether this failure of immunity represents an intrinsic deficiency of macrophages or an insufficient activation of macrophages by T cells remains uncertain [51] .

As discussed above, there has been a proliferation of work with non-human primate models in recent years that shed further light on the establishment of the set point, particularly with regards to granuloma formation [52] . Studies of macaques with aerosol exposure to Mtb have found that early mycobacterial dissemination, as reflected by the development of new granulomas in the first 3-6 weeks following infection (i.e., the period during which the set point is established), is associated with subsequent progression to active disease [53, 54] . Conversely, the absence of new granuloma formation, which could be construed as a "low" set point, is associated with immune control and long-term maintenance of latent infection. Granuloma size can also stratify the risk of dissemination, with larger granulomas associated with a greater potential for future mycobacterial spread [55] . By linking early immunologic events to clinically meaningful outcomes, these findings underscore the critical importance of early immune control prior to the onset of adaptive immunity. They also provide evidence to support the longstanding belief that progression to tuberculosis disease is determined at the level of the individual granuloma.

As the first cell type to encounter Mtb, the alveolar macrophage is a key player in the early stages of infection and Mtb is known to both cause and benefit from certain alterations in macrophage functionality [56] . Several excellent reviews have recently examined the heterogeneity and plasticity of macrophages, in addition to proposing a uniform approach to macrophage nomenclature [31, 32] . Alveolar macrophages play myriad roles in the lungs, from housekeeper to innate immune effector, and their role depends heavily on the microenvironment within the alveolar space. For example, in response to inflammatory stimuli such as lipopolysaccharide (LPS), alveolar macrophages develop an inflammatory phenotype with increased production of cytokines such as TNF-α and IL-6. Conversely, in the setting of IL-4 exposure, alveolar macrophages shift to an immunosuppressive phenotype and increase production of regulatory cytokines such as TGF-β [31] . Although debate continues as to how best to characterize macrophages and how dynamically tissue resident subtypes respond to their local cytokine milieu, it is well known that macrophages respond to different cytokine programs with a variety of functional changes. Most experts agree that these changes exist on a spectrum between the classically activated (or M1, or inflammatory) macrophage and an alternatively activated (or M2, or regulatory) macrophage.

In the context of Mtb infection, the "macrophage polarization ratio, " an indicator of the balance between proinflammatory and anti-inflammatory genes, has been employed to explore how macrophage functionality and subsequent granuloma formation affects the probabilities of Mtb control and dissemination (Fig. 2) [57] . With this approach, NF-kB signaling was found to be an essential component of the classical activation profile, which, when triggered early, leads to improved clinical tuberculosis outcomes in a nonhuman primate model [57] .

Alveolar macrophages are also severely impacted by the presence of HIV infection and have been shown to be a reservoir for the virus within the lungs [58] . Importantly, coinfection shifts macrophage polarization in a fashion favorable for pathogen survival [59] . Both HIV and Mtb stimulate a cytokine milieu in the alveolar space that pushes macrophages into an inflammatory phenotype optimized for the uptake of foreign microbes. Although these inflammatory cells would ordinarily excel at pathogen killing, the intracellular effects of HIV and Mtb render the macrophages less effective and, as a result, permit ongoing pathogen growth [60] . At the same time, there is ample evidence that Mtb infection of alveolar macrophages creates a permissive environment for HIV infection and replication, through a variety of mechanisms including increased surface expression of the HIV receptors CCR5 and CXCR4, increased HIV transcription and replication, and induction of nanotubes that promote cell-to-cell HIV viral spread between macrophages [61] [62] [63] [64] . Taken together, this synergy between Mtb-and HIV-mediated impacts on macrophages enables intracellular growth and survival of both pathogens and, more broadly, fuels the global syndemic of coinfection.

Macrophage function and polarization is also influenced by immunometabolism, the balance of cellular energy utilization. This represents another growing area of inquiry in which Mtb and HIV are likely to have complementary, yet deleterious, effects. Mtb infection of alveolar macrophages has been shown to shift cellular metabolism from oxidative phosphorylation to aerobic glycolysis, a shift associated with a more inflammatory, M1-like phenotype [65] . This shift triggers an increase in cellular IL-1β production along with a decrease in IL-10 production, which, in concert, should increase bacillary killing in vitro. Yet, other investigations suggest that persistence of fatty acid oxidation in alveolar macrophages may permit ongoing intracellular Mtb growth [66] . At the same time, HIV infection also alters cellular energy balance, oxidative stress levels, and mitochondrial bioenergetics, including for infected macrophages [67] [68] [69] [70] . Further study of the combined, net effects of Mtb and HIV on alveolar macrophage immunometabolism is needed to expand our understanding of how coinfection impairs early mycobacterial control. 

Macrophage cytokine expression profiles may also vary according to the stage of Mtb infection, whereby the secreted protein ESAT-6 initially skews macrophages towards a M1 phenotype, potentially to facilitate the establishment of a granuloma, and then later to a more permissive M2 phenotype [71] . IFN-γ, a paradigmatic pro-inflammatory cytokine, drives macrophages towards an inflammatory phenotype and, under normal circumstances, is critical for Mtb control. However, peripheral blood mononuclear cells isolated from people with HIV have depressed IFN-γ production in response to purified protein derivative (PPD) [72, 73] . IL-4, another key cytokine for macrophage polarization, is induced by HIV glycoprotein (gp) 120 [74] . This induction shifts macrophages to an anti-inflammatory phenotype, potentially further reducing early Mtb control. In people without HIV, elevated IL-4 levels have been associated with progression to tuberculosis disease and the development of pulmonary cavities, thus underscoring the clinical impact of these shifts [75] [76] [77] . Interestingly, the relative balance of these cytokines may vary according to the stage of HIV infection, which could translate into differential susceptibility to and control of Mtb infection [60] .

Several key cytokines important for both macrophage polarization and the response to Mtb are also altered in the presence of HIV coinfection. For example, GM-CSF plays an essential role in cellular control of Mtb [78, 79] and its activity is impaired in monocyte-derived macrophages infected with HIV [80] . The GM-CSF:M-CSF balance is also altered by the presence of Mtb [81] and M-CSF-driven macrophages are more permissive to Mtb replication and dissemination [82] . Similarly, interleukin-1 receptor-associated kinase M (IRAK-M), which is downstream of the GM-CSF-dependent transcription factor PU.1, correlates directly with mycobacterial load in human lung tissue and impacts Mtb survival [83] . Thus, the combined impact of Mtb and HIV on the balance of cytokines within the lungs leads to an environment that permits the growth of both pathogens.

Another avenue ripe for further exploration is the effects of HIV and Mtb on redox systems, which can also modulate macrophage function [84] [85] [86] . HIV is known to impair antioxidant defenses in the alveolar macrophage and raise levels of oxidative stress in the lung more generally [18, 87] . The effects in the lung are likely due, at least in part, to glutathione depletion in the airways of people with HIV [88, 89] . Mtb, on the other hand, has successfully adapted to hostile intracellular conditions, including high oxidative stress [90, 91] . Mtb may also directly contribute to oxidative stress by inducing heme oxygenase (HO)-1, the levels of which directly correlate with treatment outcomes in people with active tuberculosis disease [92] . Several studies have demonstrated improved Mtb control with glutathione supplementation, including one study using macrophages isolated from people with HIV in which supplementation led to enhanced inflammatory cytokine production and restored macrophage capacity to control Mtb growth [89, [93] [94] [95] [96] [97] . Of note, the oxidative stress induced by Mtb-infected macrophages is also likely to promote HIV reactivation, further complicating the relationship between these two pathogens [98] .

An important function of macrophages in the alveolar space is the differentiation of friend versus foe via pathogen recognition and phagocytosis (Fig. 3 ) [99] [100] [101] . Particularly germane to the recognition of Mtb is the mannose receptor, which is downregulated in people living with HIV [102] . HIV also leads to reduced expression and activity of toll-like receptors (TLR), another key receptor for macrophage pathogen recognition [103] [104] [105] . When combined with shifts in TLR-mediated regulation of nitric oxide production induced by Mtb [104] , these impairments have the potential to reduce Mtb uptake by alveolar macrophages and facilitate prolonged extracellular growth.

Once a pathogen like Mtb has been recognized, it then undergoes phagocytosis [99, 100] . HIV creates favorable conditions for bacteria by impairing phagocytosis of pathogens including Streptococcus [106] . The HIV protein Nef has also been shown to interfere with phagosome formation by impeding endosomal recruitment and remodeling [107] . Of note, impaired phagocytosis has been observed even in alveolar macrophages not directly infected by HIV [58, 108] . While phagocytosis of Mtb in the setting of HIV has not specifically been studied, such impairments would enable further extracellular growth [18, 109] .

HIV has also been found in macrophage endosomes and is known to impair phagolysosomal fusion in the macrophage [110, 111] . At least some of these impairments may be mediated through the effects of the HIVrelated protein transactivator of transcription (Tat) [112] . Another HIV protein, Vpr, has also been shown to perturb phagolysosomal maturation by altering microtubule-dependent trafficking within macrophages [113] . In addition, HIV has been shown to co-opt the Rag GTPases, resulting in further disruption of endosomal trafficking [114] . This disruption may facilitate HIV replication and simultaneously assist Mtb evasion of lysosomal fusion and killing. IL-10 has also been shown to dysregulate the response to Mtb in the setting of HIV, at least in part by impairing phagosome maturation [115, 116] . However, in one study of alveolar macrophages from humans with active tuberculosis disease, phagolysosomal acidification was equally disrupted among those with and without HIV, highlighting existing uncertainties about the impacts of Mtb and HIV on alveolar macrophage phagocytosis [117] . Thus, although further research is needed to define the relevant mechanisms by which HIV impairs the phagocytic response to Mtb, the bulk of the existing data suggest that endosomes fail to fuse with acidified lysosomes, thereby promoting Mtb survival [118] .

Alveolar macrophage cell death pathways range from apoptosis to autophagy and necrosis. Mtb is known to shift the balance toward necrosis by actively inhibiting apoptosis, thereby allowing the extracellular release of viable mycobacteria from dying alveolar macrophages [119, 120] . Similarly, HIV, via the viral proteins Tat and glycoprotein-120 (gp-120), stimulates the triggering receptor expressed on myeloid cells-1 (TREM-1), thereby inhibiting apoptosis in a p65-dependent manner [121] . HIV may also directly induce autophagy through the Nef protein, in order to modulate viral replication and survival, albeit in a manner that may be synergistic with Mtb-mediated inhibition of apoptosis [122, 123] . Gp-120 also impairs apoptosis-associated killing of pneumococcus within alveolar macrophages, indicating that this impairment may not be specific to Mtb [124] . Impaired apoptosis of Mtb has been tied by other investigators to IL-10, which is increased in the bronchoalveolar lavage fluid of people with HIV [125] . Additional research is needed to further delineate the impact of these shifts in cell death pathways and to identify whether there are therapeutic opportunities to improve Mtb control in patients with and without HIV. One promising recent report found that all-trans retinoic acid was able to promote autophagy of human alveolar macrophages in the setting of Mtb infection [126] .

The final step required for establishment of the mycobacterial set point is activation of the adaptive arm of the immune system (Fig. 3) . Mtb has been shown to delay the activation of adaptive immunity in mice by slowing bacterial transport to local lymph nodes [46] . This delay allows for a prolonged period of unchecked bacterial growth within the lungs, with a 10,000-100,000-fold expansion of the Mtb pulmonary population during this time. At least one driver of this delay is Mtb-mediated inhibition of MHC class II antigen presentation by dendritic [127] . Mtb also manipulates antigen presentation by diverting bacterial proteins through a vesicular export pathway, away from MHC class II presentation. In the context of HIV, the viral protein Nef has been shown to promote the expression of immature, functionally incompetent MHC class II complexes [128, 129] . Recent data have more directly linked these two by demonstrating that HIV-infected dendritic cells have reduced ability to upregulate key costimulatory molecules critical for antigen presentation (e.g., CD40, CD80, and CD86) in the presence of Mtb infection [130] . All of this would combine to delay the onset of adapative immunity and further increase the mycobacterial set point.

As should be clear from the preceding sections, studies detailing the effects of Mtb and HIV on the immune responses within the alveolar space have found an abundance of affected cytokines, pathways, and response elements. Although much of the available data suggests ways in which Mtb and HIV each make it more likely for the other to gain ground within an individual's lungs, the myriad as-yet-unstudied pathways leave open the distinct possibility that some of their particular effects may cumulatively cancel out those of the other pathogen. It is in this key domain that we believe the concept of a mycobacterial set point most clearly demonstrates its worth. Because the set point relates to the mycobacterial burden at the moment in which the adaptive immune system activates, it offers a clinically relevant marker that synthesizes the various effects of coinfections, comorbidities, and therapeutics. As a research tool, then, it will help us understand whether the net effect of a given intervention actually reduces the overall mycobacterial burden, as opposed to fruitfully manipulating a single pathway (with the knowledge that doing so may have deleterious effects on others).

Fortunately, several novel approaches provide hope for expanding our understanding of these early immunologic events in humans. These tools have the potential to vastly enhance our current knowledge by quantifying the functional impact of coinfection on bacterial burden and early human immune responses, while also offering insight into mechanisms underlying early immune control or escape.

The first such tool is 2-deoxy-2-[18F]fluoro-d-glucose ([18F]FDG) positron emission tomography combined with computed tomography (PET-CT), which identifies areas of increased cellular metabolic activity as would occur with an immune response to infection. Several macaque studies have leveraged PET-CT imaging to gain insight into early infection by following the progression of individual granulomas after aerosol infection [53, 54, 131, 132] . PET-CT has also been employed to study tuberculosis in humans, including a cohort of 35 people with HIV and latent tuberculosis infection, 10 of whom had PET-avid lesions suspicious for subclinical disease [133] . Those with PET activity were significantly more likely to develop active tuberculosis disease during the following 6 months, suggesting metabolic activity on PET imaging may be able to identify patients with ineffective early immune responses to Mtb infection.

Blood-based immunologic signatures represent another potential avenue for studying early infection. Recent studies in macaques and humans have identified RNA signatures predictive of progression to active tuberculosis disease [134] [135] [136] . Notably, the macaque signature was predictive just 3-6 weeks after aerosol infection, providing compelling evidence for the value of these signatures even very early in infection. This signature also identified a novel gene in tuberculosis pathogenesis, PRDX2, a member of the reactive oxygen species scavenger system. Further studies are warranted to explore the role of PRDX2 in early infection, and how that response may be skewed by HIV and other known tuberculosis risk factors. Similarly, characterization of Mtb-specific T cell responses may provide another means for understanding adaptive immune responses important in early infection. In a recent report of a whole blood assay of T cell responses to 60 Mtb antigens, immunogenic Mtb antigens were successfully identified across a diverse range of populations and infectious states [137] . Longitudinal studies are underway to determine if these T cell signatures correlate with early immune responses-both those that clear infection and those that permit ongoing Mtb growth and dissemination.

Yet another tool for identifying and studying early infection is plasma metabolomics, in which small molecules are identified using high-resolution mass spectrometry. This approach has been successfully employed to identify direct metabolites of Mtb, and may be able to identify low levels even early in infection. Metabolomics has also been used to identify distinct signatures of the host immune response, whereby investigators were able to distinguish between risk-associated metabolites associated with a higher likelihood of progression to tuberculosis disease and disease-associated metabolites that increased in a time-dependent manner among those with progression to active tuberculosis disease [138] [139] [140] . Looking ahead, it will be important to determine whether these metabolic signatures are similarly predictive and valid in people with HIV.

Across the scientific community, there is overwhelming consensus over the need to understand protective immunity to Mtb [55] . As a framework for considering early immunologic responses to Mtb, the mycobacterial set point provides a useful construct for understanding the impact of early immune control and clinical outcomes of tuberculosis. Given the disproportionate burden of tuberculosis disease and deaths among people with HIV despite widespread ART [23] , it is imperative to study the unique impacts of HIV on Mtb infection and susceptibility. To date, research evaluating the effects of HIV on the alveolar macrophage response to Mtb has been relatively limited, particularly with regard to the critical early events that help to dictate the mycobacterial set point. While we believe that a higher mycobacterial set point is a key contributor to the increased morbidity among people with HIV, additional research is needed to better understand the mechanisms and biological pathways that may be involved in facilitating mycobacterial immune evasion. A greater understanding of the impact of HIV on these earliest responses to Mtb may lead to the development of novel interventions and hostdirected therapies to reduce the persistent global burden of tuberculosis in this vulnerable population.

Treatment of tuberculosis. A historical perspective

Directly observed therapy for tuberculosis: history of an idea

Principles of IUATLD collaborative tuberculosis progammes

Tuberculosis therapy: past, present and future

Pneumocystis pneumonia-Los Angeles

HIV and tuberculosis: a deadly human syndemic

The importance of first impressions: early events in Mycobacterium tuberculosis infection influence outcome

The risk of disease progression is determined during the first year of human immunodeficiency virus type 1 infection

Initial plasma HIV-1 RNA levels and progression to AIDS in women and men

A whole-genome association study of major determinants for host control of HIV-1

Host genetic and viral determinants of HIV-1 RNA set point among HIV-1 seroconverters from Sub-Saharan Africa

Estimating the respective contributions of human and viral genetic variation to HIV control

Beyond macrophages: the diversity of mononuclear cells in tuberculosis

Alveolar macrophages provide an early Mycobacterium tuberculosis niche and initiate dissemination

Mechanisms of M. tuberculosis immune evasion as challenges to TB vaccine design

Pulmonary innate immune dysfunction in human immunodeficiency virus

Centers for Disease Control and Prevention. Multidrug-resistant tuberculosis outbreak on an HIV

Nosocomial transmission of multidrug-resistant tuberculosis among HIV-infected persons-Florida

An outbreak of multidrug-resistant tuberculosis involving HIV-infected patients of two hospitals in Milan, Italy. Italian Multidrug-Resistant Tuberculosis Outbreak Study Group

How soon after infection with HIV does the risk of tuberculosis start to increase? A retrospective cohort study in South African gold miners

Tuberculosis incidence rates during 8 years of follow-up of an antiretroviral treatment cohort in South Africa: comparison with rates in the community

Tuberculosis in HIV-infected persons in the context of wide availability of highly active antiretroviral therapy

Animal models of tuberculosis: an overview

Animal models of cavitation in pulmonary tuberculosis

Killing Mycobacterium tuberculosis in vitro: what model systems can teach us

Nitric oxide synthase in innate and adaptive immunity: an update

Large scale comparison of innate responses to viral and bacterial pathogens in mouse and macaque

Tissue-resident macrophage ontogeny and homeostasis

Macrophage activation and polarization: nomenclature and experimental guidelines

Macrophage heterogeneity in tissues: phenotypic diversity and functions

Differential polarization of alveolar macrophages and bone marrow-derived monocytes following chemically and pathogeninduced chronic lung inflammation

Human lung immunity against Mycobacterium tuberculosis: insights into pathogenesis and protection

The zebrafish model of tuberculosis-no lungs needed

Mouse model of tuberculosis

Animal models of tuberculosis: lesson learnt

Anti-vascular endothelial growth factor treatment normalizes tuberculosis granuloma vasculature and improves small molecule delivery

Prevention of tuberculosis in macaques after intravenous BCG immunization

Monkey models of tuberculosis: lessons learned

Some clinical features of tuberculosis. 1. Incubation period

Tuberculous infection in the light of tuberculin matriculation

The time-table of tuberculosis

Resistance to tuberculosis: experimental studies in native and acquired defensive mechanism

Animal model: experimental airborne tuberculosis in the guinea pig

Initiation of the adaptive immune response to Mycobacterium tuberculosis depends on antigen production in the local lymph node, not the lungs

Delayed-type hypersensitivity and cell-mediated immunity in the pathogenesis of tuberculosis

Tracking salmonella-specific CD4 T cells in vivo reveals a local mucosal response to a disseminated infection

Organ-specific CD4+ T cell response during Listeria monocytogenes infection

Progressive pulmonary tuberculosis is not due to increasing numbers of viable bacilli in rabbits, mice and guinea pigs, but is due to a continuous host response to mycobacterial products

Immunity to tuberculosis

Experimental Mycobacterium tuberculosis infection of cynomolgus macaques closely resembles the various manifestations of human M. tuberculosis infection

Early Changes by (18)Fluorodeoxyglucose positron emission tomography coregistered with computed tomography predict outcome after Mycobacterium tuberculosis infection in cynomolgus macaques

Digitally barcoding Mycobacterium tuberculosis reveals in vivo infection dynamics in the macaque model of tuberculosis

Protective immunity against tuberculosis: what does it look like and how do we find it?

Intranasal Poly-IC treatment exacerbates tuberculosis in mice through the pulmonary recruitment of a pathogen-permissive monocyte/macrophage population

Macrophage polarization drives granuloma outcome during Mycobacterium tuberculosis infection

Healthy HIV-1-infected individuals on highly active antiretroviral therapy harbor HIV-1 in their alveolar macrophages

Macrophage polarization: convergence point targeted by mycobacterium tuberculosis and HIV

The macrophage in HIV-1 infection: from activation to deactivation?

Tuberculosis exacerbates HIV-1 infection through IL-10/STAT3-dependent tunneling nanotube formation in macrophages

High turnover of tissue macrophages contributes to tuberculosis reactivation in simian immunodeficiency virus-infected rhesus macaques

Mycobacterium tuberculosis and HIV coinfection brings fire and fury to macrophages

The Mtb-HIV syndemic interaction: why treating M. tuberculosis infection may be crucial for HIV-1 eradication

Cutting edge: Mycobacterium tuberculosis induces aerobic glycolysis in human alveolar macrophages that is required for control of intracellular bacillary replication

Growth of Mycobacterium tuberculosis in vivo segregates with host macrophage metabolism and ontogeny

Emerging role and characterization of immunometabolism: relevance to HIV pathogenesis

Intracellular expression of Tat alters mitochondrial functions in T cells: a potential mechanism to understand mitochondrial damage during HIV-1 replication

Abnormal immunometabolism and gene accessibility in alveolar macrophages in HIV

HIV infection and latency induce a unique metabolic signature in human macrophages

Mycobacterium tuberculosis virulent factor ESAT-6 drives macrophage differentiation toward the pro-inflammatory M1 phenotype and subsequently switches it to the anti-inflammatory M2 phenotype

Depressed T-cell interferon-gamma responses in pulmonary tuberculosis: analysis of underlying mechanisms and modulation with therapy

The significance of interferon-γ in HIV-1 pathogenesis, therapy, and prophylaxis

HIV-1 gp120 Induces IL-4 and IL-13 release from human FcεRI + cells through interaction with the V H 3 region of IgE

Endogenously activated interleukin-4 differentiates disease progressors and non-progressors in tuberculosis susceptible families: a 2-year biomarkers follow-up study

Increased Interleukin-4 production by CD8 and gammadelta T cells in health-care workers is associated with the subsequent development of active tuberculosis

Increased production of interleukin 4 by CD4+ and CD8+ T cells from patients with tuberculosis is related to the presence of pulmonary cavities

Role of granulocyte-macrophage colony-stimulating factor production by T cells during Mycobacterium tuberculosis infection

Understanding tuberculosis-analyzing the origin of Mycobacterium tuberculosis pathogenicity

Human immunodeficiency virus type 1 infection inhibits granulocyte-macrophage colony-stimulating factor-induced activation of STAT5A in human monocyte-derived macrophages

Relative levels of M-CSF and GM-CSF influence the specific generation of macrophage populations during infection with Mycobacterium tuberculosis

Mycobacterium tuberculosis replicates within necrotic human macrophages

IRAK-M alters the polarity of macrophages to facilitate the survival of Mycobacterium tuberculosis

Glutathione and infection

Mechanisms of control of Mycobacterium tuberculosis by NK cells: role of glutathione

Glutathione levels in antigen-presenting cells modulate Th1 versus Th2 response patterns

Oxidative stress during HIV infection: mechanisms and consequences

Anti-retroviral therapy is associated with decreased alveolar glutathione levels even in healthy HIV-infected individuals

Glutathione supplementation improves macrophage functions in HIV

Mycobacterium tuberculosis WhiB3 maintains redox homeostasis by regulating virulence lipid anabolism to modulate macrophage response

Redox biology of tuberculosis pathogenesis

Mycobacterium tuberculosis induction of heme oxygenase-1 expression is dependent on oxidative stress and reflects treatment outcomes

Glutathione-redox balance regulates c-rel-driven IL-12 production in macrophages: possible implications in antituberculosis immunotherapy

Evidence for oxidative stress and defective antioxidant response in guinea pigs with tuberculosis

Glutathione levels and immune responses in tuberculosis patients

Liposomal glutathione supplementation restores TH1 cytokine response to Mycobacterium tuberculosis infection in HIV-infected individuals

Glutathione and adaptive immune responses against Mycobacterium tuberculosis infection in healthy and HIV infected individuals

Mycobacterium tuberculosis reactivates HIV-1 via exosome-mediated resetting of cellular redox potential and bioenergetics. mBio

The delivery of endocytosed cargo to lysosomes

Endosome-lysosome fusion

Insights into macrophage autophagy in latent tuberculosis infection: role of heat shock protein 16.3

Reduced binding and phagocytosis of Pneumocystis carinii by alveolar macrophages from persons infected with HIV-1 correlates with mannose receptor downregulation

Toll-like receptor 2-dependent extracellular signalregulated kinase signaling in Mycobacterium tuberculosis-infected macrophages drives anti-inflammatory responses and inhibits Th1 polarization of responding T cells

Toll-like receptor-induced arginase 1 in macrophages thwarts effective immunity against intracellular pathogens

Human immunodeficiency virus infection alters tumor necrosis factor alpha production via Toll-like receptor-dependent pathways in alveolar macrophages and U1 cells

The alveolar microenvironment of patients infected with human immunodeficiency virus does not modify alveolar macrophage interactions with Streptococcus pneumoniae

Inhibition of phagocytosis in HIV-1-infected macrophages relies on Nef-dependent alteration of focal delivery of recycling compartments

Small alveolar macrophages are infected preferentially by HIV and exhibit impaired phagocytic function

Defective phagocytosis by human monocyte/macrophages following HIV-1 infection: underlying mechanisms and modulation by adjunctive cytokine therapy

HIV-1 buds and accumulates in "nonacidic" endosomes of macrophages

Impairment of phagosome-lysosome fusion in HIV-1-infected macrophages

HIV-1 Tat alters neuronal autophagy by modulating autophagosome fusion to the lysosome: implications for HIV-associated neurocognitive disorders

The HIV-1 protein Vpr impairs phagosome maturation by controlling microtubule-dependent trafficking

HIV-1 enhances mTORC1 activity and repositions lysosomes to the periphery by co-opting Rag GTPases

HIV-1 infection of macrophages dysregulates innate immune responses to Mycobacterium tuberculosis by inhibition of interleukin-10

IL-10 blocks phagosome maturation in Mycobacterium tuberculosis-infected human macrophages

Mycobacterium tuberculosis resides in nonacidified vacuoles in endocytically competent alveolar macrophages from patients with tuberculosis and HIV infection

Endosomal membrane traffic: convergence point targeted by Mycobacterium tuberculosis and HIV

Apoptosis is an innate defense function of macrophages against Mycobacterium tuberculosis

Cell death and autophagy in tuberculosis

HIV-related proteins prolong macrophage survival through induction of Triggering receptor expressed on myeloid cells-1

Human immunodeficiency virus type 1 Nef inhibits autophagy through transcription factor EB sequestration

Mycobacterium tuberculosis and human immunodeficiency virus type 1 cooperatively modulate macrophage apoptosis via toll like receptor 2 and calcium homeostasis

HIV gp120 in the lungs of antiretroviral therapy-treated individuals impairs alveolar macrophage responses to pneumococci

Impaired M. tuberculosismediated apoptosis in alveolar macrophages from HIV + persons: potential role of IL-10 and BCL-3

All-trans retinoic acid augments autophagy during intracellular bacterial infection

Mycobacterium tuberculosis infects dendritic cells with high frequency and impairs their function in vivo

HIV-1 Nef impairs MHC class II antigen presentation and surface expression

Nef alleles from children with non-progressive HIV-1 infection modulate MHC-II expression more efficiently than those from rapid progressors

HIV interferes with Mycobacterium tuberculosis antigen presentation in human dendritic cells

Analysis of 18FDG PET/CT imaging as a tool for studying Mycobacterium tuberculosis infection and treatment in non-human primates

Sterilization of granulomas is common in active and latent tuberculosis despite within-host variability in bacterial killing

Characterization of progressive HIV-associated tuberculosis using 2-deoxy-2-[18 F] fluoro-d-glucose positron emission and computed tomography

Prospective discrimination of controllers from progressors early after low-dose Mycobacterium tuberculosis infection of cynomolgus macaques using blood RNA signatures

Four-gene pan-African Blood signature predicts progression to tuberculosis

A blood RNA signature for tuberculosis disease risk: a prospective cohort study

A high throughput whole blood assay for analysis of multiple antigenspecific T Cell responses in human Mycobacterium tuberculosis infection

Plasma metabolomics in human pulmonary tuberculosis disease: a pilot study

High-resolution plasma metabolomics analysis to detect Mycobacterium tuberculosis-associated metabolites that distinguish active pulmonary tuberculosis in humans

Metabolite changes in blood predict the onset of tuberculosis

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations

We would like to thank our colleague Cheryl L. Day, PhD, for review, thoughtful discussion, and feedback on this manuscript.

SCA and BSS conceived of the review, drafted and revised the manuscript. Both authors read and approved the final manuscript.

This work was supported by the following grants: NIH/NIAID K23 AI134182