Mycobacterium tuberculosis (Mtb), the etiologic agent of tuberculosis (TB), is a successful pathogen that primarily resides within human host cells. These host cells provide very different environments for the intracellular bacilli. Indeed, inside host cells Mtb localises within various cellular compartments also referred as “intracellular niches”. These biochemically diverse compartments might be composed by one or more intact host membranes surrounding the pathogen. These membranes can be actively damaged by Mtb thus allowing the direct exposure to the cytosol (Bussi and Gutierrez, 2019). How the presence of host membranes and the features of distinct intracellular environments affect antibiotic efficacy is still poorly understood. This is important because the treatment of TB is usually recommended for at least six months with a cocktail of four antibiotics and we still do not really know why TB treatment takes so long.
Understanding the impact of Mtb intracellular lifestyle in the pathogenesis and treatment of TB is one of the fundamental aspects we investigate in our research laboratory at the Francis Crick Institute in London. Because Mtb is an intracellular pathogen, our group integrates concepts from both cell biology and microbiology to decipher how TB disease progresses and how to design better treatments.
When I joined Max’ group at the Crick for my postdoctoral stay, the lab was interested in studying how Mtb residence within host cells may impact antibiotic accumulation and efficacy. This was extremely challenging due to the lack of imaging technologies available enabling to track antimicrobials at high-resolution. Luckily, the laboratory has recently developed, in collaboration with the research group of Dr. Haibo Jiang based at the University of Hong Kong (formerly at the University of Western Australia) a high-resolution correlative imaging approach to visualise antibiotics at subcellular resolution (Greenwood et al., 2019). This correlative approach combines light (fluorescent), electron, and ion microscopy (CLEIM) in order to define “where” and to “which extent” antibiotic molecules accumulate within infected cells (Greenwood et al., 2019; Fearns et al., 2020).
As a trained microbiologist in a cell biology lab, I was really interested in investigating both cellular and molecular mechanisms responsible for the efficacy of anti-TB drugs within infected human macrophages. Before I joined the Crick, Daniel Greenwood a former PhD student in the lab (now at ETH, Zurich, Switzerland) that had developed CLEIM, showed that isotopically-labelled antibiotic pyrazinamide (PZA) was detectable by NanoSIMS.
There are currently many proposed modes of action for PZA, and most of them are based in in vitro studies, so I decided to combine the CLEIM approach with “in cellulo” studies in human macrophages to define how this critical antibiotic target and efficiently kill intracellular Mtb.
Discovered 70 years ago, PZA is a prodrug that needs to be converted into pyrazinoic acid (POA) to display efficacy against Mtb. Surprisingly, this highly potent drug in vivo is poorly active in vitro when using standard culture conditions; except when tested at pH 5.5 or below. This suggests that acidic pH faced by the bacteria in the context of infection might be critical for PZA anti-mycobacterial efficacy. We hypothesized that the intracellular lifestyle of Mtb might affect whether PZA reaches the pathogen and subsequently impact its efficacy.
By using the CLEIM pipeline, we observed that PZA does not distribute within specific macrophage organelles but accumulates within specific intracellular bacteria. We were very surprised to find that this accumulation pattern was highly heterogenous, where some neighbouring bacteria displayed very different levels of antibiotic. We also noticed that the distribution of PZA/POA molecules were highly heterogenous in between cells contained within the same biological sample. This cell-to-cell variation and the intra-cellular single-bacterial heterogeneity in drug accumulation was unexpected and highlighted that antibiotic targeting of intracellular pathogens is a puzzling process with multiple layers of complexity, even in very simple two-dimensional infection models.
Very excited with these results, we then postulated that the diversity of subcellular environments that Mtb encounters during the infection of human macrophages could be responsible for this heterogeneity in PZA/POA levels.
To test this hypothesis, we turn into mycobacterial genetics and compared the PZA/POA accumulation profile of the Mtb H37Rv WT strain with a Mtb ΔRD1 mutant strain. The latter lacks a functional type seven secretion system (ESX-1) that allows Mtb to access the pH-neutral nutrient-rich cytosolic environment. Analysis of the PZA/POA profile at the single-bacterial level showed increase accumulation in the Mtb ΔRD1 strain in comparison to Mtb WT virulent strain suggesting that restriction to intact phagosomes enhances PZA/POA levels.
Knowing the potential pH-dependent mode of action of PZA/POA and that Mtb phagosomal restriction is normally associated with increase acidification, we designed a second series of experiments in the presence of pharmacological inhibitors of phagosome acidification such as Bafilomycin A1 or Concanamycin A. The use of vacuolar-type H+ ATPase inhibitors impaired intrabacterial PZA/POA accumulation in both Mtb WT and Mtb ΔRD1 suggesting that intracellular acidic microenvironments are critical for this antibiotic to reach and accumulate within its bacterial target.
Thus, localisation of Mtb and host intracellular pH changed the intrabacterial accumulation profile of PZA/POA, but was this effect having an impact on antibiotic efficacy? Using a high content imaging approach to monitor bacterial replication in Biosafety level-3 laboratory, we found that both phagosomal pH neutralisation and phagosomal escape negatively impact PZA/POA efficacy confirming that intracellular localisation and host-cell microenvironments largely contribute to antibiotic accumulation and subsequently antimicrobial efficacy.
Altogether, our results raise some new concepts regarding intracellular pharmacokinetics and antibiotic efficacy in the context of infection with intracellular pathogens. Moreover, the development of CLEIM and its association with high-content imaging technologies has proven to be a powerful approach to dissect antibiotic mode of action in complex biological systems. Finally, we hope that our findings will open new avenues regarding the study of antibiotics mechanism of action and pave the way for the development and implementation of efficient, short and compliant-friendly therapeutic alternatives which are desperately needed worldwide.
Our study entitled “Intracellular localisation of Mycobacterium tuberculosis affects efficacy of the antibiotic pyrazinamide” is now available here in Nature Communications.
Subcellular visualization of pyrazinamide and bedaquiline antibiotics in Mycobacterium tuberculosis-infected human macrophages
Mycobacterium tuberculosis-infected human macrophages were treated with Pyrazinamide (PZA) and Bedaquiline (BDQ) antibiotics and further imaged by high-resolution Nano-Secondary Ion Mass Spectrometry Imaging (NanoSIMS). Intracellular bacteria are displayed in Blue (31P signal), PZA enrichment (15N/14N signal) is displayed in Red and Bedaquiline distribution (79Br) is displayed in Green. Images were processed using the open source software Fiji where a median filter was applied to the three channels.
Pierre Santucci: firstname.lastname@example.org
MSCA Postdoctoral Fellow, Host-Pathogen Interactions in Tuberculosis Laboratory, The Francis Crick Institute, London, United Kingdom