Evaluation of novel TB drug regimens by targeting resuscitation promoting factor-dependent persistent Mycobacterium tuberculosis in the Cornell model

Tuberculosis, although curable, still remains one of the biggest killers in the world. It kills nearly 2 million people worldwide every year. 98% of tuberculosis deaths are in the developing world affecting mostly young adults in their productive years. A quarter of a million tuberculosis deaths are also infected with HIV with weakened immune systems, most of these people are in Africa. Tuberculosis especially affects the most vulnerable populations including children, the poorest and malnourished. Tuberculosis is caused by the bacterium called Mycobacterium tuberculosis. One important characteristic of this disease is that the bacterium has an unusual ability to grow and survive for extended periods of time in human body. Therefore, it has been estimated that 2 billion people, equal to one-third of the world's total population, are infected with the bacterium in whom it causes unnoticeable latent infections that lead to a 5-10% lifetime risk of active disease. These persistent bacteria cannot be cultured using standard microbiological methods and are not killed by the current tuberculosis drugs. Therefore, tuberculosis treatment needs a long period of time with four drugs to cure the patients. This long-term treatment is extremely difficult to implement especially in developing countries because of lack of affordability and limited healthcare services and infrastructure.

The proposed research aims to identify and quantify those persistent bacteria in tuberculosis-infected mice before, during and after treatment to predict the outcome of human tuberculosis treatment. The persistent bacteria will be "woken" by the addition of resuscitation promoting factors (RPF) which are proteins produced by M. tuberculosis to restart growth. We have shown in our recent study in mice that high-dose rifampicin or bedaquiline which was added in the current treatment drug regimen was able to kill persistent RPF resuscitated bacteria. This meant that the treatment duration could be shortened with a reduced relapse rate. We also modify the Cornell model to better mimic human infection by enriching persistent bacteria and determining their clearance. In this proposal, we will apply the same principals and techniques which we have learned from studying mice using a set of novel drug regimens to predict the outcome of human tuberculosis treatment, especially disease relapse. We will establish a model system in mice as a testbed to evaluate the potencies of new drug regimens prior to their application in more expensive and time-consuming human testing.

Treatment of tuberculosis with novel regimens in a modified Cornell model. The novel drug regimens will be tested using a modified Cornell mouse model in which treatment starts at 6 weeks after a low dose M. tuberculosis H37Rv infection. Treatment will be given for 14 to 16 weeks by daily oral administration. Mouse organs will be harvested at a two-week interval to monitor CFU and broth counts with 7H9 and culture filtrates. After the termination of the chemotherapy, the remaining mice will be administered hydrocortisone for 8 weeks. CFU counts from lungs and spleens will be performed to determine disease relapse.
Standardizing resuscitation promoting factor production from M. tuberculosis and preparation of culture filtrates. M. tuberculosis is grown in 7H9 for 15 to 20 days until an optical density of 1 to 1.5 is reached. The cultures are harvested by centrifugation and filtered with 0.2 um filters twice. The culture filtrates will be used immediately and RPF stability testing will be performed before and after storage at -20 or -70-degree centigrade.





Resuscitation of M. tuberculosis in mouse organs. Broth counting is performed as serial 10-fold dilutions in triplicate in which 0.5 ml of organ homogenates are added to 4.5 ml of the culture filtrates or 7H9 medium. At 10-day intervals over a 2-month period of incubation, the broth cultures will be examined for visible turbidity and confirmed by colonial morphology on 7H11 agar. The most probable number of viable bacilli is then estimated from the patterns of positive and negative tubes.
Resuscitation of RPF-dependent persisters in Mycobacteria Growth Indicator Tube (MGIT). Mouse organ homogenates will be incubated in MGITs, then culture filtrate will be added into MGIT before and after 42-day incubation and continue to incubate for up to 2 months.

Tuberculosis patients will benefit from this research. Tuberculosis caused by Mycobacterium tuberculosis is the single most important infectious disease in the world and leads to approximately 2 million deaths annually. Tuberculosis infects human populations of all age groups, but mostly affects adults in their most productive years. One million children (0-14 years) fell ill with TB, and 140 000 children died from the disease in 2014. More than 95% of tuberculosis cases and deaths reside in the developing world. HIV co-infection contributes 20 to 30 times more risk for individuals to develop active tuberculosis. More than 20% of TB cases worldwide are attributable to smoking. Furthermore, about one-third of the world's population harbor dormant M. tuberculosis which lives in infected individuals for the rest of their lives, and this provides a huge pool of potential disease since 5-10% of those who are infected will suffer from the active disease in the future. Effective global disease control is severely hindered by the requirement for 6 months of antibiotic therapy which is needed to achieve a low disease relapse rate. The prolonged multi-drug therapy which is challenging to implement, especially in high-burden developing countries inevitably leads to poor patient compliance, high relapse rates and drug resistance.
Shortening TB treatment duration with a low relapse rate is a critically important objective to significantly improve global tuberculosis disease control. Currently, persistent M. tuberculosis is undetectable using conventional bacterial culture methods, rendering clinicians unable to detect, and thus successfully treat these elusive bacteria. These persistent bacteria fail to show after acid-fast staining, do not multiply on solid agar plates or in broth medium. Essentially, to successfully eliminate these persisters, one must first detect them. One of the most intuitive and promising methods is to "wake up" the persistent bacteria from their dormant state, and induce them to recommence multiplication. We previously showed that resuscitation-promoting factors (RPF), i.e. self-generated proteins by M. tuberculosis to induce multiplication, successfully generated measurable growth counts from dormant persisters. Using RPF to wake up persistent bacteria may enable quantitative detection of the persistent population in patients with tuberculosis and holds promise for becoming a powerful risk stratification tool for clinical treatments. We hypothesize that the rate of bacterial elimination and disease relapse in tuberculosis treatment is related to the level of RPF resuscitated persistent bacteria. If we reduce or remove these persistent bacteria, tuberculosis chemotherapy will be shortened and the relapse rate will be reduced. Accurate detection and quantification of persistent M. tuberculosis will make novel drug and drug regimen discovery feasible which may act as a persistent biomarker with predictive value to shortened tuberculosis treatment.