Intervening along the spectrum of tuberculosis: meeting report from the World TB Day nanosymposium in the Institute of Infectious Disease and Molecular Medicine at the University of Cape Town

Tuberculosis (TB), caused by the highly infectious Mycobacterium , remains a leading cause of death worldwide, with an tuberculosis estimated 1.6 million associated deaths reported in 2017. In South Africa, an estimated 322,000 (range 230,000-428,000) people were infected with TB in 2017, and a quarter of them lost their lives due to the disease. Bacille Calmette-Guérin (BCG) remains the only effective vaccine against disseminated TB, but its inability to confer complete protection against pulmonary TB in adolescents and adults calls for an urgent need to develop new and better vaccines. There is also a need to identify markers of disease protection and develop novel drugs. It is within this backdrop that we convened a nanosymposium at the Institute of Infectious Disease and Molecular Medicine at the University of Cape Town to commemorate World 1 2 1,3,4 5


Disclaimer
The views expressed in this article are those of the author(s). Publication in Gates Open Research does not imply endorsement by the Gates Foundation.
Tuberculosis is a debilitating communicable disease affecting millions of people worldwide, with the highest incidence and mortality rates in Southern Africa. In 2017 an estimated 1.6 million people died of TB worldwide, making it the leading cause of death due to a single infectious agent 1 . TB is transmitted through airborne inhalation of Mycobacterium tuberculosis (Mtb) that seeds in alveolar spaces in the lungs. Realizing the limitations of the TST and IGRA, it has been estimated that almost one-third of the world population that is exposed to Mtb infection is able to control the infection and remain asymptomatic, otherwise known as latently infected 2 . However, 5-10% of these latently infected individuals progress to active TB and manifest symptoms of the disease 3 . More recently, the situation has been made more complex by the emergence of drugresistant bacteria and co-infections with the human immunodeficiency virus (HIV). In 2017, the WHO estimated that 558 000 (range 483,000-639,000) people developed resistance to rifampicin (RR-TB), and 82% of these people had multidrug resistant TB (MDR-TB) 4 . This underpins an urgent need to develop better and efficacious vaccines than the currently available BCG and to develop novel drugs for TB.

Identifying new drug targets
Mtb has a sophisticated metabolic repertoire: it is able generate its own nutrients, but it can also scavenge for some nutrients from the host 5 . Studies have shown that Mtb is able to synthesise essential amino acids such as L-arginine and tryptophan, where deletion of these key metabolites restricts Mtb growth in culture and renders it more susceptible to host immune pressure reducing its survival 6,7 . Dr Melissa Chengalroyen, a research officer in the Molecular Mycobacteriology Research Unit, under the directorship/supervision of Prof Valerie Mizrahi, opened the morning session and spoke about the key elements required for Mtb survival and how Mtb sequesters these elements to evade host-recognition by non-conventional T cells. She then described a complex de novo riboflavin metabolic pathway and different tools to create Mtb mutants lacking key enzymes involved in this pathway to facilitate understanding of each step in the pathway. Her study showed that not all the enzymes involved in the riboflavin pathway are essential-some play redundant roles, while others are absolutely necessary for Mtb survival, and these may hold promise as new candidate drug targets for TB or play an essential role in alerting non-restricted T cell immune arm for faster clearance of the bacteria.
The second speaker in the morning session was Dr Kehilwe Nakedi, a postdoctoral research fellow in the laboratory of Prof Jonathan Blackburn. Her research project was aimed at identifying novel substrates for mycobacterial protein kinase G (PknG) using a mass spectrometry-based phosphoproteomics approach to elucidate mechanisms by which mycobacteria interfere with the host signalling during LTBI. She identified 3164 phosphopeptides with high confidence using label-free data analysis 8 . Moreover, she identified 63 host phosphopeptides that were phosphorylated in macrophages infected with M. bovis BCG only and not those infected with the mutant lacking PknG. Further analysis of the data revealed that these substrates phosphorylated in the presence of PknG play a key role in regulating actin polymerisation and cytoskeleton integrity 8 . This work suggest that pathogenic mycobacteria survives inside the host macrophages during early TB infection through interfering with the host's cytoskeletal dynamics mediated by PknG.

Development of host-directed drug therapies
Although the currently available TB treatment regimens are effective at killing the bacteria, the emergence of drug resistance and the long duration of treatment threaten their long-term efficacy. This underpins an urgent need to develop new anti-TB drugs and exploration of other treatment strategies to control the disease. One such strategy is to develop host-directed drug therapies (HDTs) with the aim of boosting the host's innate ability to fight the infection and also limit the deleterious tissue pathology 9 . Although this field is still in its infancy, it holds huge potential as adjunctive therapy for TB in clinical settings with a high disease burden.
Associate Prof Reto Guler discussed in detail their published pre-clinical data on the use of statins as a potential host-directed therapy for TB in mice 10,11 . He then spoke about the translation of this work in a proof-of-concept phase IIB, double-blind, randomized, placebo-controlled trial launched in Khayelitsha township and funded by the European & Developing Countries Clinical Trials Partnership ((EDCTP), RIA2017T-2004. The coordinator of this consortium is Reto Guler (University of Cape Town, Division of Immunology). Chief principal investigator of the clinical trial is Friedrich Thienemann (University of Zürich), local PI in Cape Town is Sandra Mukasa (University of Cape Town). Other project partners include Robert J. Wilkinson (Imperial College London), Claudia Schacht (LINQ Management GmbH, Germany), Gunar Günther and Emmanuel Nepolo (University of Namibia). The aim of the clinical trial is to investigate the use of statins to prevent chronic lung inflammation and

Amendments from Version 2
In the new version, we have changed the abstract and added additional information suggested by reviewer 2. Changes are in Page 2, lines 36, 41-46. We have also made changes in the Main body section as suggested by reviewer 2, changes appear in Page 2, lines 57, 64, 66-67. Under Transmission of Mtb in Page 4 and 5, we made changes in lines 166-168 and line 186. Under TB/HIV co-infections, we made changes in Page 6, line 213-214. We also made changes in line 224-228 and added additional information on biomarkers in TB-IRIS patients who died or survived. Under Engineering Tools for TB in page 6, we changed wording and included additional references on geometric sketching as suggested by reviewer. These changes appear on Page 6, line 236-238. We all refer the reviewer to the file with point by point reply to some of the changes not included in the final version of this manuscript. We thank the reviewer for all the insights in improving this report.
Any further responses from the reviewers can be found at the end of the article REVISED potentially TB relapse in patients post completion of a standard TB treatment regimen. He also talked about other potential candidate targets for HDTs such as the transcriptional factor BATF2 12 and microRNA-143 and microRNA-365 13 .
Another speaker on this topic was Dr Suraj Parihar, a Senior Research Officer and contributing investigator at CIDRI-Africa in the Institute of Infectious Diseases and Molecular Medicine (IDM). His talk focused mainly on preclinical studies that investigated the efficacy of repurposed drug (barberine) generally used to reduce blood glucose and cholesterol as HDTs for TB. He found that this drug was able to reduce lung inflammation in pre-clinical in vitro and in vivo models of TB. He then went on to speak about how growth factors can also be repurposed to enhance the killing ability of human and mouse macrophages. Although some of these studies are still at the pre-clinical stage, they hold a great promise and may pave way for alternative therapies and new clinical trials supported by strong pre-clinical data.

Transmission of Mtb
Mtb is transmitted through the inhalation of Mtb-containing aerosols from person to person. Transmission of the bacteria is high in places such as schools, churches, mines, hospitals and heavily congested neighbourhoods of Cape Town such as informal settlements and townships 14,15 . In places with high prevalence of HIV, such as Khayelitsha, the rate of new Mtb infections accounts for more than half of TB cases. In 2006, it was reported that the incidence was as high as 1500/100,000 in some townships, exceeding that of national average 16 . There are many challenges when it comes to measuring Mtb transmission via aerosol such as the low numbers of bacilli that can be captured, contamination by other bacterial or fungal particles in patients and other airborne particulate matter 17 . New aerosol methods for capturing Mtb such as the respiratory aerosol sampling chamber (RASC) hold promise in quantifying rate of transmission especially in high endemic areas. The capacity to measure the rate of transmission and the type of bacilli strain circulating becomes even more critical in the era of high antimicrobial resistance 18 .
Dr Anastasia Koch, a Carnegie Developing the Next African Leaders (DEAL) early career fellow, mentored by Prof Helen Cox and Prof Digby Warner, started off her talk by mentioning the potential of WGS technology in identifying Mtb genotypes resistant to the first line TB drugs 19 . She discussed the major differences in genetic diversity observed in broth cultured Mtb populations and those derived directly from sputum, and moreover how simple culturing could result in loss of some of key genotypes. Anastasia then moved on to discussing the importance of getting a sample as close to that being transmitted by an infected person as possible in order to accurately study the strains that are being transmitted and driving disease in a community. She gave an example of how colleagues from the MMRU and the Desmond Tutu HIV Centre, have been able to capture and isolate Mtb strains from RASC bio-aerosols. She was able to culture these samples and compare whole genome of those strains with sputum induced strains. The ability to isolate Mtb from bio-aerosols and combining that with whole genome sequencing could greatly inform our understanding of TB transmission and treatment, particularly of emerging drug or multi-drug resistant strains.

TB/HIV co-infection
More than 36.7 million people live with HIV/AIDS globally and most of these people live in sub-Saharan Africa. In 2017, it was estimated that more than 350 000 people died due to HIV/TB co-infection, making the TB a highest contributor to death in people living with HIV 20 . HIV targets and depletes CD4 T cells at later stages of disease, including protective TB specific CD4 T cells 21,22 . Early antiretroviral (ARV) drug treatment is associated with improved outcomes and helps restore CD4 T cell count, including protective TB-specific CD4 T cells. However, there are complications associated with early ARV treatment in people who also are starting TB treatment. Some of these people develop TB-associated immune reconstitution syndrome (TB-IRIS), which can be fatal, unless controlled by host-directed immune suppressants such as corticosteroids 23 .
To set the scene, Mohau Makatsa, a PhD candidate in the laboratory of Prof Wendy Burgers in the Division of Medical Virology, talked about a particular subset of CD4 T helper cells expressing IL-22, or "Th22 cells", which are targeted by HIV and are implicated as a key player in natural resistance to TB 24 . He showed how these cells can be stimulated ex-vivo by Mtb antigens and express different surface molecules compared to Th1 cells. There is a similar magnitude of these cells compared to Th1 cells in people with latent TB, but they are depleted in the peripheral blood in active TB disease and HIV co-infection. IL-22 has been shown to be important for the control of Mtb in mice 25 . It is unclear how these cells play a role in the pathogenesis of TB in humans.
Dr Muki Shey, a Senior Research Officer & Wellcome Intermediate Fellow at CIDRI-Africa in the IDM followed and talked about identification of predictive immunological biomarkers associated with mortality in people who died of severe HIV-associated TB (HIV-TB). The study identified molecules such as interleukin-1 receptor agonist (IL-1Ra), IL-6, IL-8, macrophage inflammatory protein-1 beta (MIP-1B), and interferon gamma induced-protein 10 (IP-10) to be immune mediators that segregated patients who died from those who survived 26 . The identification of these predictive markers could lead to better management of patients in the clinics, and also potentially lead to development of host-directed therapies.

Engineering tools for TB
The international guest speaker at the Nanosymposium was Assistant Prof Bryan Bryson from the Massachusetts Institute of Technology. Bryan spoke about using cutting-edge single cell RNA sequencing technology to dissect activation states of macrophages infected with Mtb. He identified key regulatory proteins that are differentially expressed in granulocyte-macrophage colony-stimulating factor (GM-CSF) versus M-CSF differentiated macrophages and how these heterogenous macrophages play a role in the control of Mtb infection 27 . He also described a new method they have developed to summarize transcriptomic heterogeneity within a dataset using geometric sketching 28 . This method allows for greater understanding of macrophage heterogeneity and spatiotemporal localisation within granulomas. He also talked about phagosomics, a new way of measuring phagosome maturation and identifying new genes/ proteins associated with phagosome formation during Mtb infection. He also talked about how to build a phagosome de novo, which enables for large scale testing of Mtb host stresses such as drugs. This uses tagged Mtb strains in combination with gene expression data to allow for better understanding of phagosome transcriptional changes in the presence of multiple Mtb stresses.

Way forward
Prof Valerie Mizrahi, the director of the IDM gave the closing speech at the end of the symposium. In her concluding remarks she said, "It's with incredible passion that people are progressing TB research, and that is because we're living with it. It's incumbent on all of us to think about why we're doing what we're doing and to remember that at the end of the day, it's about the TB patients. Ultimately, one of the things we want to do is put ourselves out of business."

Conclusion
It is evident from the talks given by the various speakers that a lot of research is being done to combat TB at the IDM in the Faculty of Health Sciences at the University of Cape Town. The research ranges from the identification of new candidate drug targets, investigating the utility of repurposed drugs as host-directed drug therapies, understanding the transmission of the bacteria in high burden communities, investigating the immunology of TB/HIV co-infection and identification of biomarkers for TB disease progression. An important highlight of this year's World TB Day Nanosymposium is that a bulk of this research is being undertaken and led by early career research; thus, demonstrating the depth and breadth of talented TB researchers at the IDM.

Data availability
No data are associated with this article.
The meeting took place in 2019. The WHO Global Report on Tuberculosis 2018 was published 6 months before the meeting and reflects data gained in 2017. The reference should therefore be corrected.
The statement in the abstract (which would be better in the preamble) that "In South Africa, 322,000 people were infected with TB in 2017" is actually the incidence of disease rather than infection and the wide confidence intervals (230,000-428,000) should be given.
Reference 2 is not a primary source estimate for either the prevalence of latent tuberculosis infection (see Behr , 2018 ) nor the prevalence of an immune response to tuberculin or in an et al. interferon-gamma release assay (see Houben and Dodd, 2016 ).
Although the WHO 2018 report records that there were 160,684 cases of MDR/RR-TB detected and notified in 2017, the same paragraph gives an estimated incidence of 558,000 cases.
The statement that "This underpins an urgent need to develop better and efficacious vaccines than the currently available BCG and to develop novel drugs for TB" misses several steps in logical argument from the previous sentences regarding epidemiology to this conclusion.
When describing the speakers' contributions, it is helpful to present a brief statement indicating where their research fits into the TB research agenda. For instance, reference 3 indicates how tryptophan protects Mtb from macrophages stimulated by CD4+ T cells and reference 4 describes how arginine deprivation makes Mtb susceptible to killing via reactive oxygen intermediates. However, it is then unclear whether these essential amino acids were the "elements" sequestered by Mtb and were, in a different way, responsible for evading host recognition (not mentioned in either of the two references). How was riboflavin related to these references? My reading of the literature is that riboflavin metabolites might be recognized by MAIT cells. The section on Dr Nakedi's work is more understandable, although its positioning within New TB drugs rather than TB pathogenesis is difficult to fathom.
Under "Host-directed drug therapies", statins should be mentioned in the first sentence and an explanation given as to how they might affect a successful response to respiratory pathogens, especially Mtb. The final sentence does not give the reasons why BATF2 and miR-143 and 365 might be relevant to the topic of host directed drug therapies.
The section on Dr Purihar's work on barbarine is understandable.
The section on "Transmission of Mtb" appeared to deal with whole genome sequencing rather than transmission. It was not clear how these might be related. The need to obtain early samples would seem to relate more to the transcriptome (and proteome) of such bacilli and perhaps to epigenetic changes. The paragraph on the talk by Dr Shey would benefit from a listing of "predictive immunological biomarkers" which distinguish severe illness from likely death from tuberculosis.
The paragraph "Engineering tools for TB" covers a range of topics and bullet points for the different parts of the talk would be helpful. The difference in the macrophage populations needs an introductory sentence. The term "parametric stitching" is a neologism -presumably sequences are aligned and differences noted?
This sounds an interesting symposium, but needs context and discussion to realise its value.

Gates Open Research
Are all factual statements correct, and are statements and arguments made adequately supported by citations? Partly

Is the Open Letter written in accessible language? Partly
Where applicable, are recommendations and next steps explained clearly for others to follow? No No competing interests were disclosed.

Competing Interests:
Reviewer Expertise: Tuberculosis I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard, however I have significant reservations, as outlined above. The abstract does not well describe the meeting. The material up to "The theme of the nanosymposium was…" could be deleted.
: We note the comment from the reviewer. We have modified the abstract to Authors response address this comment by the reviewer.

·
A symposium consists of both presentations and discussion. The paper does not describe the discussion.
: We note this comment by the reviewer. However, due to time constraints, we Authors response could not set aside time dedicated to discussion. However, there was an opportunity for engagement between speakers and guests during the poster session. Also, people were given time to ask question to speakers during the question and answer session. However, these questions and responses were not captured in this meeting report. We will look to implement this in future meeting reports.

·
The meeting took place in 2019. The WHO Global Report on Tuberculosis 2018 was published 6 months before the meeting and reflects data gained in 2017. The reference should therefore be corrected.
: We have corrected the reference. Authors response · The statement in the abstract (which would be better in the preamble) that "In South Africa, 322,000 people were infected with TB in 2017" is actually the incidence of disease rather than infection and the wide confidence intervals (230,000-428,000) should be given.
: We have given the wide confidence interval as suggested by the reviewer. Authors response · Reference 2 is not a primary source estimate for either the prevalence of latent tuberculosis infection (see Behr , 2018 ) nor the prevalence of an immune response to tuberculin or in an et al. interferon-gamma release assay (see Houben and Dodd, 2016 ).
: We thank the reviewer for bringing this to our attention. We have cited the Authors response review Behr et al., 2018 study by Houben andDodd, 2019. · Although the WHO 2018 report records that there were 160,684 cases of MDR/RR-TB detected and notified in 2017, the same paragraph gives an estimated incidence of 558,000 cases.
: We have amended the sentence to reflect the estimated range by the WHO. Authors response The sentence now reads "In 2017, the WHO estimated that 558 000 (range 483,000-639,000) people developed resistance to rifampicin (RR-TB), and 82% of these people had multi-drug resistant TB (MDR-TB) ".

·
The statement that "This underpins an urgent need to develop better and efficacious vaccines than the currently available BCG and to develop novel drugs for TB" misses several steps in logical argument from the previous sentences regarding epidemiology to this conclusion.
: The sentence is drawn from all the presented epidemiology presented in the Authors respone paragraph. The aim of it is to chart a way forward in the fight to combat the scourge of TB.

·
When describing the speakers' contributions, it is helpful to present a brief statement indicating where their research fits into the TB research agenda. For instance, reference 3 indicates how tryptophan protects Mtb from macrophages stimulated by CD4+ T cells and reference 4 describes how arginine deprivation makes Mtb susceptible to killing via reactive oxygen intermediates. However, it is then unclear whether these essential amino acids were the "elements" sequestered by Mtb and were, in a different way, responsible for evading host recognition (not mentioned in either of the two references). How was riboflavin related to these references? My reading of the literature is that riboflavin metabolites might be recognized by MAIT cells.
: The two references show that Mtb requires some essential amino acids and if Authors response we disable its ability to sequester these amino acids, it becomes susceptible to immune mediated killing. Dr Chengalroyen described her research on nutrients required for Mtb survival and focus specifically on the metabolism of riboflavin and explored the draggability of this pathway.

·
The section on Dr Nakedi's work is more understandable, although its positioning within New TB drugs rather than TB pathogenesis is difficult to fathom.
: We thank the reviewer for this comment. The candidate proteins identified by Authors response mass spectrometry can be investigated as potential candidate drug targets, thus, leading to the development of new drugs. We do however take the point of the reviewer but we think that the talk is still well placed in the topic of New TB drugs.

·
Under "Host-directed drug therapies", statins should be mentioned in the first sentence and an explanation given as to how they might affect a successful response to respiratory pathogens, especially Mtb. The final sentence does not give the reasons why BATF2 and miR-143 and 365 might be relevant to the topic of host directed drug therapies.
: The statins are mentioned in the first sentence of the second paragraph and we Authors response provide a citation of the work published by Prof Guler and colleagues. The citation is provided for the readers who would like to know more about the role of statins in response to Mtb infection. The last sentence is stating the other molecular targets Prof Guler is investigating as HDT for TB. 1 2 1 last sentence is stating the other molecular targets Prof Guler is investigating as HDT for TB. Citations are also provided for the readers that would want to know more about these targets and why they are important for TB.

·
The section on Dr Purihar's work on barbarine is understandable. : Noted. Authors response · The section on "Transmission of Mtb" appeared to deal with whole genome sequencing rather than transmission. It was not clear how these might be related. The need to obtain early samples would seem to relate more to the transcriptome (and proteome) of such bacilli and perhaps to epigenetic changes.
: Whole genome sequencing has been suggested as a tool to investigate and Authors response track transmission of Mtb (Hatherill et al., 2016. BMC Medicine), however, little work has been done in high burden settings to understand how genetically related Mtb strains change during transmission. The need to obtain early samples is not related to the transcriptome or proteome of such bacteria but related to capturing the strains that might be driving transmission in communities .

·
The section on IL-22 could usefully have referred to Ronacher (2018 ) as a summary of et al. the position at the time of the symposium. New data (Ardain , 2019 ) make a more interesting et al. role regarding group 3 innate lymphoid cells in mice in combination with IL-17.
: We thank the reviewer for bringing these references to our attention. We have Authors response cited Ronacher et al., 2018.

·
The paragraph on the talk by Dr Shey would benefit from a listing of "predictive immunological biomarkers" which distinguish severe illness from likely death from tuberculosis.
: We thank the reviewer for this comment. We have provided a list of predictive Authors Response biomarkers and cited the recently published manuscript for readers with keen interest on the topic.

·
The paragraph "Engineering tools for TB" covers a range of topics and bullet points for the different parts of the talk would be helpful. The difference in the macrophage populations needs an introductory sentence. The term "parametric stitching" is a neologism -presumably sequences are aligned and differences noted?
: -Page 4, first paragraph: "capturing Mtb such as respiratory…." Would add "the" after as. changed as suggested.

Response:
-Page 4, third paragraph: "TB a highest contributor…" Would change "a" to "the". changed as suggested Response: -Page 4, 5 paragraph: "…after presenting to hospital…" Would add "the" after to. changed as suggested.

Response:
-Page 4, 6 paragraph: "allows" appears twice and should be followed by "for". changed as suggested.

Response:
No competing interests were disclosed. Competing Interests: th th