Introduction

One third of the world’s population is estimated to be latently infected with tuberculosis (TB) and progression to TB disease causes 1.4 million deaths each year [1]. However, in the great majority of immunocompetent people latently infected with TB, antimycobacterial immunity prevents progression to symptomatic disease [1]. Furthermore, in the pre-antibiotic era many episodes of symptomatic TB disease were controlled by the host immune system and apparently resolved spontaneously [2]. Even with effective antibiotic treatment, the likelihood of treatment achieving a lasting cure is also influenced by host immunity [3,4]. Thus, TB infection represents a balance between progression of the mycobacterial infection and containment by antimycobacterial immunity. Antibiotic-resistant TB has increasing prevalence but limited diagnosis and treatment facilities and the emergence of TB strains resistant to almost all known antibiotics [5] is renewing interest in approaches to strengthen patients’ antimycobacterial immunity [6].

Altitude therapy and TB care in mountain sanatoria have been used historically for TB care [7] and epidemiological studies suggest that high altitude is associated with reduced incidence of TB infection, disease and mortality [8-17]. However, the mechanisms underlying this apparent altitude-mediated effect are poorly understood, and are likely to include both mycobacterial transmission and host antimycobacterial immunity [18].

Antimycobacterial immunity has been studied in vitro by quantifying the restriction of mycobacterial growth by immune cells [19,20]. This has been facilitated by the development of genetically modified luminescent mycobacteria that allow rapid and accurate mycobacterial quantification, overcoming their slow growth on solid media and inaccuracies caused by mycobacteria clumping [21,22]. A whole-blood model of the extent to which blood cells support or restrict the growth of luminescent mycobacteria has been used effectively for investigating aspects of human antimycobacterial cellular immunity [23]. These include correlation with cytokine production and tuberculin skin test status [24], evaluation of vaccine candidates [14,25] and of nutritional augmentation of antimycobacterial immunity [26].

To enable studies in a high altitude setting, we simplified previously published whole-blood assays of antimycobacterial immunity so that a technician in a low-technology setting with only portable equipment and without a biosafety cabinet, centrifuge or freezer could concurrently perform multiple assays. Since the antimycobacterial immunity assays were repeated at different times and locations, variations in temperature, oxygen tension, mycobacterial stocks, atmospheric pressure and other factors inevitably caused inter-assay differences in mycobacterial growth. To reduce the effect of these inter-assay variations in mycobacterial growth, antimycobacterial immunity was assessed as the extent to which participants’ whole blood supported or restricted mycobacterial growth relative to concurrent growth in positive-control culture broth. Humoral (acellular) immunity is relatively inactive against mycobacteria and plasma is generally tuberculostatic (neither supporting mycobacterial growth nor killing mycobacteria). Thus antimycobacterial immunity was also inferred from the extent to which whole blood supported mycobacterial growth relative to negative-control plasma.

We used this assay to assess the effects of high altitude on peoples’ antimycobacterial immunity.

Methods

Subjects

The effect of altitude on antimycobacterial immunity was studied by comparing samples collected under three conditions (Figure 1). Firstly, low-altitude residents living in Lima, Peru, less than 100 meters above sea level, provided an initial blood sample at low-altitude. Secondly, they then provided another sample after ascent to high altitude in Cusco, Peru, 3400 meters above sea level. Thirdly, samples were also obtained from permanent residents at the same high altitude location. Participants were identified by inviting university staff, affiliates and students to participate. Inclusion criteria were healthy adults who gave informed written consent for this research that had internationally accredited ethical committee approval. Exclusion criteria were travel to high altitude during the previous month or having a tuberculin skin test within the previous six months, in case these affected antimycobacterial immunity. A questionnaire was used to characterize demographic data, previous TB exposure and vaccination with bacille Calmette-Guérin (BCG). The nutritional status of the individuals was assessed by measuring mid-upper arm circumference, height, weight and calculating body mass index.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Figure 1. Study design.

Fifteen low altitude residents had blood drawn at low altitude, which was used to obtain whole blood and plasma samples that, together with culture broth samples, were then inoculated with luminescent mycobacteria. In half of these samples, baseline luminescence readings were measured straight away. The other half of these samples were incubated at 37°C for 96 hours and luminescence readings were repeated. The low altitude residents ascended to high altitude and together with a group of 47 high altitude residents, blood samples were taken and the experiment repeated.

https://doi.org/10.1371/journal.pone.0074220.g001

Results

(b): Antimycobacterial immunity

The restriction of mycobacterial growth by whole blood relative to positive-control culture broth is shown in Table 2b and in Figure 2. At low altitude, mycobacteria grew similarly abundantly in blood as in the positive-control culture broth (p=0.4). Specifically, mycobacterial growth in blood was only 0.17 logs (i.e. 1.5-times) less than growth in positive-control culture broth. In contrast, ascent to high altitude was associated with restriction (p=0.002) of mycobacterial growth in the same participants’ blood to be 0.62 logs (i.e. 4-times) less abundant than the positive-control culture broth. Blood from long-term high-altitude residents also restricted mycobacterial growth to be 0.65 logs (i.e. 4-times) less abundant than the positive-control culture broth (p=0.001 versus low altitude residents at low altitude).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Figure 2. Restriction of mycobacterial growth by blood relative to culture broth.

https://doi.org/10.1371/journal.pone.0074220.g002

Antimycobacterial immunity was assessed in 15 low altitude residents as they ascended to high altitude (square symbols joined by lines) and in 47 high altitude residents at high altitude (circular symbols). The vertical axis shows luminescence recorded as the log₁₀ of measured relative light units (arbitrary units) produced by genetically modified luminescent mycobacteria. The increase in luminescence over 96 hours incubation is shown positive-control culture broth minus whole blood. For low altitude residents at low altitude, mycobacteria grew similarly in culture broth and blood. Ascent to high altitude was associated with restriction of mycobacterial growth in whole blood relative to culture broth, similar to the values for high altitude residents at high altitude.

The extent to which whole blood supported mycobacterial growth relative to negative-control plasma is also shown in Table 2b. At low altitude, mycobacteria grew 1.4 logs (i.e. 25-times) more abundantly in participants’ whole blood than in their negative-control plasma. In contrast, ascent to high altitude was associated with restriction (p=0.01) of mycobacterial growth in the same participants’ blood to be only 0.76 logs (i.e. 6-times) more abundant than their negative-control plasma.

There was no evidence of differences in antimycobacterial immunity at high altitude between people who were usually low-altitude residents versus long-term high-altitude residents (Table 2b). This was true whether considering growth in whole blood relative to positive-control culture broth (p=0.4) or considering growth in whole blood relative to negative-control plasma (p=1.0).

Discussion

In this study we simplified an assay of human antimycobacterial immunity to study altitude effects. This demonstrated the feasibility of this assay for assessing antimycobacterial immunity in low-resource settings and revealed opposing effects of high altitude on the pathogen versus the host. Specifically, ascent to high altitude both facilitated the growth of the mycobacterial pathogen and appeared to augment healthy participants’ immune resistance to this pathogen. This antimycobacterial immunity was assessed as the restriction of mycobacterial growth in whole blood relative to positive-control culture broth and was also inferred from the mycobacterial growth in whole blood relative to negative-control acellular plasma.

High altitude settings have historically been associated with apparent protection from TB [7] and people diagnosed with TB still sometimes travel to high altitude because they believe that this will cure their disease [28]. There is epidemiological evidence from several countries that increasing altitude is associated with decreased rates of TB latent infection [8,9] and morbidity [12,14,15,29]. Even after adjusting these epidemiological analyses for potentially confounding variables, high altitude was associated with decreased rates of TB disease [10,11,13,14] and death [16,17]. Several of these studies were located in Peru, where the current research also took place [8,9,15]. In the pre-antibiotic era these and other factors led to patients with TB disease being treated with ascent to and supportive care provided at high altitude, so-called “altitude therapy” for TB [7]. Rigorous evaluation of this approach appears to be lacking and it has largely fallen into disuse since the advent of effective antimycobacterial antibiotics.

The association between increasing altitude and reduced TB might imply that high altitude directly impairs mycobacterial growth. We found the opposite: that ascent to high altitude was in fact associated with increased mycobacterial growth in both culture broth and human plasma. Human antimycobacterial immunity is largely cellular and consequently plasma would not be expected to mount effective dynamic antimycobacterial immune responses and plasma has long been known to be tuberculostatic, neither supporting TB growth nor killing TB [30]. The effect of altitude on extracellular mycobacterial growth in plasma does not appear to have been defined previously, and this part of our findings is consistent with studies of aeration and oxygenation that, depending upon experimental conditions, may stimulate or impair extracellular mycobacterial growth [31-33]. Thus, high altitude stimulated extracellular mycobacterial growth in human plasma under the experimental conditions that we used.

In the current research, altitude-mediated augmentation of extracellular mycobacterial growth in plasma was counteracted by findings consistent with altitude-mediated augmentation of cellular antimycobacterial immunity. This finding was apparent for mycobacterial growth in whole blood compared with both the positive-control culture-broth in each assay and also the negative-control plasma from each individual. We have not been able to identify previous human studies of altitude and antimycobacterial immunity. However, our novel finding is supported by experiments in: animals in which altitude augmented antimycobacterial immunity [34]; in cells in which hypoxia stimulated the human vitamin D-dependent antimycobacterial pathway [32]; and in humans whose non-specific cellular immunity was augmented by altitude [35,36]. These findings demonstrate increased immune resistance to mycobacteria at altitude, which offers an explanation for the apparent success of altitude therapy for TB. Other non-immunological factors that may also contribute to reduced TB transmission (and thus disease) at high altitude include germicidal effects of altitude-associated ultraviolet light [37], reduced humidity [38], and hypoxia [33].

Our objective to study the effect of altitude on human antimycobacterial immunity was challenged by factors that we addressed by developing a simplified assay of antimycobacterial immunity. Firstly, the complexity of published antimycobacterial immunity assays limited their feasibility in high-altitude settings [19-22,24-26], so we simplified an assay to require only portable equipment: hand-held incubator, pipettes; and a briefcase sized luminometer. Secondly, mycobacterial quantification is often inaccurate, technically demanding and usually requires many weeks [19,20], so we used genetically-modified luminescent mycobacteria [21,22,24-26] and simplified their processing to quantify them reliably and instantly, without centrifugation. Thirdly, large-volume blood samples discourage participation; so we scaled the assay to require less than 1.5 ml of blood. Fourthly, we found in preliminary studies that day-to-day and place-to-place variations in mycobacterial growth confounded assessments of how well blood restricted mycobacterial growth, an issue that was likely to worsen in the extreme environments involved in our research. Consequently, we interpreted assays as relative mycobacterial growth in blood compared with positive-control culture-broth included in each assay and also relative to concurrent negative-control plasma from each individual. We used 1.5 ml of blood per participant but found high replicability between quadruplicate assays, so recommend that in future only single (not quadruplicate) assays be performed, which would require only 0.3 ml (6 drops) of blood. This simplified, low blood volume assay that uses only portable equipment should facilitate assessment of human antimycobacterial immunity in the resource-poor settings where most TB occurs [6,25,26,39].

Limitations to this study include the lack of adjustment for differences between the low and high altitude residents, such as their age. However, we found no evidence of an association between participant age and the results of the antimycobacterial immunity assay. Furthermore, these methodological limitations would not affect our experiment in which antimycobacterial immunity was assessed in the same individuals prior to and after ascent to high altitude. The logistical challenges of transferring research participants and the team and equipment for assessing antimycobacterial immunity from low to high altitude restricted the number of participants in this part of the research. This small sample size limits the certainty of our research findings that should be replicated in a larger study. A potentially affordable strategy for addressing this limitation in future research may be to re-assay antimycobacterial immunity again in each participant after returning to low altitude. It is also noteworthy that there is no gold standard, validated in vitro correlate of antimycobacterial immunity because no test has been shown to predict peoples’ resistance or susceptibility to TB infection or disease [25]. Thus our finding that ascent to high altitude increased cellular restriction of mycobacterial growth is consistent with but cannot provide absolute confirmation for the long-reported altitude-associated resistance to TB infection, disease and mortality.

The field use of this assay in a low-resource high-altitude setting limited our capacity for: (1) mechanistic studies of cell counts, cell types and immunological mediators and processes, so future studies should store frozen aliquots of the assays to characterize mediators of high altitude-augmentation of antimycobacterial immunity and to determine whether altitude directly augmented cellular anti-mycobacterial immunity or whether indirect mechanisms such as altitude-mediated changes in the mycobacterial and/or cellular nutrients or cell counts in blood may have contributed [36]; (2) confirming that luminescence predicted mycobacterial growth at high altitude as it did in validation experiments at low altitude by ourselves and by others [24] and that the altitude-associated effects on mycobacterial luminescence indicated mycobacterial killing not reduced metabolism or senescence; (3) using virulent M. tuberculosis instead of low-virulence BCG mycobacteria that is a relatively safe but imperfect surrogate for Mycobacterium tuberculosis; and (4) studying the time course of altitude-mediated augmentation of antimycobacterial immunity. It was noteworthy that we found similar antimycobacterial immunity less than one day after ascent to altitude as after long-term residence there. Larger studies may differentiate the potentially contributory mechanisms of altitude per se [36] versus associated hypoxia [18], possible increased exposure to sunlight increasing vitamin D levels [39] and nutritional changes [26].

These findings do not imply that patients with difficult-to-treat TB should migrate to high altitude, where we found that not only the disease-resisting human immunity but also the disease-causing pathogens were potentiated. However, our findings provide some evidence for an altitude-mediated augmentation of human antimycobacterial immunity that may explain the anecdotal observation that TB patients often improve and may even be cured when they move to high altitude and the well-proven rarity of TB at high altitude. A so-called “antibiotic apocalypse” of emerging resistance of Mycobacterium tuberculosis to almost all known antibiotics [40,41] has been reported. Thus, our findings justify rigorous evaluation of how altitude affects antimycobacterial immunity as this may be exploited to develop new treatments [6,32]. Clinical research is also needed to determine whether, as was common practice in the pre-antibiotic era, the increasing numbers of patients with TB resistant to all antibiotics would benefit from their care being moved to high-altitude [40,41].

Supporting Information

Acknowledgments

The authors are grateful for the contributions of the research team, not all of whom meet the criteria to be co-authors and to Beate Kampmann for assistance optimising the assay used; Douglas Young for the gift of the mycobacteria used; Rosa Pacheco for providing the support of her laboratory; Luz Caviedes for mycobacteriology expertise; Carlos Santillan for clinical research assistance; Silvia Carrera, Paula Maguina and Maribel Rivero for expert administrative support; and to the participants who agreed to donate blood samples.