NAD(H) homeostasis is essential for host protection mediated by glycolytic myeloid cells in tuberculosis

Background

Tuberculosis (TB) is an infectious disease caused by the bacterial pathogen Mycobacterium tuberculosis (Mtb) that affects mainly the lungs and can be fatal, being one of the world’s deadliest infectious diseases¹. This pathogen can cause an infection due to its capacity to modulate the immune response in its favor. However, the molecular mechanisms by which Mtb evades host phagocytes it’s still incomplete.

Myeloid cells are the main Mtb reservoir and represent a lineage of phagocytes with important roles in pathogens control. Over the past decade, there has been growing evidence suggesting that the function of myeloid cells in the immune response and their metabolism are intrinsically linked; For example, upon inflammatory stimuli, macrophages switch their metabolism to a glycolytic profile, with a disruption in the TCA cycle and a high lactate production². It has already been shown that the downregulation of the glycolytic flux in myeloid cells impairs their function, interfering with the production of inflammatory mediators and phagocytosis³. Mtb is able to modulate the glycolytic flux in myeloid cells through an unclear mechanism, and the impact of this event on the immune response is unknown⁴.

During glycolysis, NAD+ is converted to NADH, which needs to be oxidized back to NAD+ for the regeneration of cellular energy. In immune cells, upon inflammatory signals, this process happens mainly through the activity of lactate dehydrogenase (LDH), culminating in lactate production. LDH is a tetramer composed of two subunits (LDHA and LDHB), where the presence of LDHA subunits favors the NADH re-oxidation while LDHB subunits favor the opposite reaction⁵. Despite its redox state, NAD+ levels can also be modulated during inflammatory diseases once the immune response promotes the expression of several NAD+ consuming enzymes, such as PARPs and CD38.

NAD+ synthesis pathways are dysregulated upon inflammatory signals, such as NAMPT, an enzyme that has a crucial role synthesizing NAD from its precursor nicotinamide (NAM)⁶. Therefore, the authors hypothesized that Mtb could affect myeloid cells NAD(H) homeostasis, impairing host cell glycolysis and allowing the infection. The study described in this preprint aimed to understand the mechanisms by which Mtb disrupts glycolysis and how it could impact the outcome of the infection.

 

Key findings

LDHA expression correlates with inflammatory and necrotic sites in human TB lung sections

Histoanalysis revealed consistent LDHA staining in early necrotic sites of lung tissues resected from TB patients, showing a reduced necrotic progression which was limited to the granulation layers. This suggests that the staining in the early lesions could be due to a release of LDHA by necrotic cells. The authors also detected LDHA expression in lymphoid aggregates, giant cells, macrophages, and lymphocytes. As these cells promote inflammation and no LDHA staining was observed in the vicinity of these cells, the authors raised the possibility that LDHA could be linked to immune function. Indeed, given that LDHA is the predominant LDH subunit expressed in myeloid cells, the staining of these cells in the inflammatory sites should be expected.

LDHA expression in myeloid cells is essential to protect mice against Mtb infection

The authors could show that the depletion of LDHA in myeloid cells increased the mortality of Mtb-infected mice and increased lung burden 10 weeks post-infection (wpi). While the lesions observed by histoanalysis in the lungs of wild-type (WT) mice were higher at early time points of the disease (4 wpi) and resolved by 30 wpi, mice lacking LDHA in myeloid cells (LDHA^LysM-/-) displayed worse lesions at 30 wpi.

LDHA genetic ablation in myeloid cells affects immune cell trafficking to the lungs in Mtb-infected mice

By using multiparameter flow cytometry, the authors analyzed the populations of immune cells present in the lungs of Mtb infected mice. Corresponding to the inflammatory profile observed in the histoanalysis, LDHA^LysM-/- mice exhibited a lower abundance of leukocytes at early time points (4 wpi). Interestingly the difference in the number of leukocytes was not restricted to the myeloid populations, but extended to lymphoid populations which were also decreased in LDHA^LysM-/- mice at 4 wpi. LDHAL^ysM-/- mice also presented higher levels of the inflammatory cytokines MCP-1, RANTES, MIP-1a, and MIP-1b.

LDHA is required for macrophage’s metabolic response to IFN-γ

Although LDHA^LysM-/- mice were more susceptible to Mtb infection, transcriptome analysis found higher levels of IFN-γ signaling transcripts, which was not reflected in the protein levels. Therefore, the authors decided to investigate the capacity of LDHA^LysM-/- mice bone marrow-derived macrophages (LDHA BMDM) to respond metabolically to IFN-γ treatment. Based on ECAR analysis, the authors suggested that upon IFN-γ treatment WT BMDM glycolytic activity was induced to a greater extent in comparison to the LDHA^-/- BMDM. Pyruvate addition rescued the deficit in the glycolytic response of LDHA^-/- BMDM to IFN-γ probably by enhancing LDHB activity and promoting the oxidation of NADH, as LDH inhibition with GSK 2837808A attenuated the rescue. Although the NAD/NADH ratio was lower in the LDHA^-/- BMDM compared to WT BMDM, glucose uptake was not affected. Metabolomic experiments revealed an accumulation of glycolysis and pentose phosphate pathway intermediates in the LDHA^-/- BMDM, which was no longer present after pyruvate supplementation, suggesting that the glycolytic dysfunction was due to a lack of NAD⁺.

Mtb infection disrupts glycolysis by decreasing NAD(H) levels

The authors next hypothesized that the disruption of the glycolytic flux by the virulent Mtb infection – in comparison to inactive or attenuated Mtb – could be blunting macrophage’s response to IFN-y and inflammatory stimuli. Therefore, this event could increase macrophage’s susceptibility to Mtb infection. Although Mtb infection of macrophages enhanced the glycolytic flux in comparison to no stimuli, the glycolytic flux in the infected cells upon IFN-y treatment was lower than in uninfected cells. The authors also found a decrease in the accumulation of glycolysis and PPP intermediates and an increase in pyruvate and isocitrate levels upon Mtb infection. The authors suggest that the pyruvate accumulation and reduction in the glycolytic flux could be due to a lack of NADH to fuel LDH activity, as they also observed lower NAD(H) levels upon Mtb infection. Supplementation of BMDMs with the NAD precursor NAM partially restored NAD levels and rescued the glycolytic capacity of these cells.

Nicotinamide supplementation has therapeutic effects on tuberculosis by enhancing glycolysis

NAM supplementation has already been shown as an effective treatment for TB through an uncharacterized mechanism. The authors could show that NAM supplementation controls Mtb replication in BMDM macrophages in a dose-dependent manner. NAM treatment of bacterial cultures of Mtb reduced bacterial growth. NAM treatment of Mtb-infected macrophages was shown to reduce the bacterial burden, and this effect was reversed when combining the treatment with the glycolytic inhibitor 2-DG or with the NAMPT inhibitor FK866, showing that the therapeutical effects of NAM rely on its conversion to NAD and on the glycolytic activity. Experiments using the LDHA-/- BMDM supplemented with NAM should be considered to reinforce the suggestion that NAM supplementation and the glycolytic activity do not control bacterial burden through independent processes.

Figure 1: Mtb disrupts glycolysis in myeloid cells by depleting NAD(H) levels. Mtb infection leads to a depletion in NAD(H) levels, impairing LDH function and promoting the accumulation of pyruvate, which is metabolized by the Krebs cycle and promotes an increase in (iso)citrate levels, that can allosterically inhibit phosphofructokinase 1, disrupting the glycolytic flux and impairing immune response.

 

Why I think this preprint is important:

As bacteria are able to acquire resistance to antibiotics, new therapies must be studied and developed to treat bacterial infections. This preprint investigates the mechanisms by which Mycobacterium tuberculosis disrupts the glycolytic flux in myeloid cells, and the role of this process in the onset of the disease caused by this pathogen. It highlights the importance of NAD(H) homeostasis and suggests a potential therapeutic target for the treatment of the disease.

 

Questions and suggestions for the authors 

1- Regarding the conclusion based on this first result, it would be more precise if the authors change the word “implicate” to “associate” as the authors did not show any direct role of LDH on the immune response and on the onset of human TB lesions. The authors could consider the possibility of assessing LDHB staining in TB lung sections to conclude that LDHA is the predominant isoform in areas of granulomatous inflammation.

2- LDHA genetic ablation in myeloid cells could affect cell viability as the depletion of this component of LDH has already been shown to induce apoptosis and decrease cell proliferation in cancer cells that also have a metabolism that relies on aerobic glycolysis⁷. Although we agree that the data regarding immune cells populations present in the lungs of LDHA^LysM-/- mice during Mtb infection indicates that cell viability was not affected, have you directly assessed cell viability and cell number of myeloid population in LDHA^LysM-/- mice? Alterations in myeloid cell number and viability could be the cause of the delay in the inflammatory process and cellular infiltration in LDHA^LysM-/- mice lungs. Therefore, perhaps you could consider a cell counting experiment and viability test of myeloid populations present in the blood and bone marrow of LDHA^LysM-/- mice.

3- Regarding the glycolytic capacity of LDHALysM-/- cells, the authors should carefully interpret the ECAR data as these might not directly reflect the glycolytic flux. ECAR could also be affected by CO2 produced by the TCA cycle⁸. A suggestion to the authors would be to use oxamate (a lactate dehydrogenase inhibitor) and fluorocitrate which disrupt the TCA cycle and to perform the ECAR experiments evaluating the contribution of glycolysis to the extracellular acidification. Lactate quantification experiments could also reinforce the glycolytic flux measurements.

4- Mtb infection in macrophages has already been shown to dysregulate the splicing machinery to promote non-productive RNAs or the translation of truncated proteins⁹. As you observed an increase in RNA levels of the IFN-y signaling pathway which did not translate to the proteins in Mtb-infected LDHA^LysM-/- mice, you might find it interesting to investigate RNA processing in future studies to better explain this unusual result.

5- the mouse sub-strain C57BL/6J is a mutant for nicotinamide nucleotide transhydrogenase, an enzyme that catalyzes the reduction of NADP at the expense of NADH oxidation in the mitochondria^10,11. This mutation leads to abnormalities in the energy metabolism and in the redox balance of NAD derivatives¹⁰. It would be worth addressing how this mutation was considered in the overall conclusions of the study.

6- An alternative cytosolic pathway to regenerate NAD+ from NADH is the glycerol-3-phosphate (G3P) shuttle, which culminates in G3P production. You found an increase in G3P levels in LDHA^-/- BMDM, however, you did not discuss the possible cause of it. Did you consider the possibility of the G3P shuttle compensating for the lack of LDHA in BMDM as you found an increase in G3P levels in this condition? Investigating this possibility could provide further insights into the metabolic changes.

7- Regarding the statistical analysis, it’s not clear if the samples follow a gaussian distribution. If so, to enhance the transparency of the statistical analysis, you could consider specifying which normality test was used in the materials and methods section. Alternatively, explaining the criteria for selecting specific statistical tests, even without assuming normal data distribution, would be beneficial.

 

References

1. Bussi C, Gutierrez MG. Mycobacterium tuberculosis infection of host cells in space and time. FEMS Microbiol Rev 2019; 43: 341–361.
2. Viola A, Munari F, Sánchez-Rodríguez R, Scolaro T, Castegna A. The Metabolic Signature of Macrophage Responses. Front Immunol 2019; 10: 1462.
3. Soto-Heredero G, Gómez de las Heras MM, Gabandé-Rodríguez E, Oller J, Mittelbrunn M. Glycolysis – a key player in the inflammatory response. FEBS J 2020; 287: 3350–3369.
4. Park J-H, Shim D, Kim KES, Lee W, Shin SJ. Understanding Metabolic Regulation Between Host and Pathogens: New Opportunities for the Development of Improved Therapeutic Strategies Against Mycobacterium tuberculosis Infection. Front Cell Infect Microbiol; 11, https://www.frontiersin.org/articles/10.3389/fcimb.2021.635335 (2021, accessed 20 June 2023).
5. Osis G, Traylor AM, Black LM, Spangler D, George JF, Zarjou A, Verlander JW, Agarwal A. Expression of lactate dehydrogenase A and B isoforms in the mouse kidney. Am J Physiol Renal Physiol 2021; 320: F706–F718.
6. Fang J, Chen W, Hou P, Liu Z, Zuo M, Liu S, Feng C, Han Y, Li P, Shi Y, Shao C. NAD+ metabolism-based immunoregulation and therapeutic potential. Cell Biosci 2023; 13: 81
7. Zhang W, Wang C, Hu X, Lian Y, Ding C, Ming L. Inhibition of LDHA suppresses cell proliferation and increases mitochondrial apoptosis via the JNK signaling pathway in cervical cancer cells. Oncol Rep 2022; 47: 77.
8. Mookerjee SA, Goncalves RLS, Gerencser AA, Nicholls DG, Brand MD. The contributions of respiration and glycolysis to extracellular acid production. Biochim Biophys Acta BBA – Bioenerg 2015; 1847: 171–181.
9. Kalam H, Fontana MF, Kumar D. Alternate splicing of transcripts shape macrophage response to Mycobacterium tuberculosis infection. PLoS Pathog 2017; 13: e1006236.
10. Ronchi JA, Figueira TR, Ravagnani FG, Oliveira HCF, Vercesi AE, Castilho RF. A spontaneous mutation in the nicotinamide nucleotide transhydrogenase gene of C57BL/6J mice results in mitochondrial redox abnormalities. Free Radic Biol Med 2013; 63: 446–456.
11. Enríquez JA. Mind your mouse strain. Nat Metab 2019; 1: 5–7.

Tags: immunometabolism, mycobacterium tuberculosis, nad metabolism

doi: https://doi.org/10.1242/prelights.35744

Read preprint (No Ratings Yet)