Transcriptomic signatures predict regulators of drug synergy and clinical regimen efficacy against Tuberculosis

Abstract

22 23 Classification: BIOLOGICAL SCIENCES: Microbiology; Biophysics and Computational Biology 24 25 ABSTRACT 34 35 The rapid spread of multi-drug resistant strains has created a pressing need for new drug 36 regimens to treat tuberculosis (TB), which kills 1.8 million people each year. Identifying new 37 regimens has been challenging due to the slow growth of the pathogen M. tuberculosis (MTB), 38 coupled with large number of possible drug combinations. Here we present a computational 39 model (INDIGO-MTB) that identified synergistic regimens featuring existing and emerging anti-40 TB drugs after screening in silico over 1 million potential drug combinations using MTB drug 41 transcriptomic profiles. INDIGO-MTB further predicted the gene Rv1353c as a key 42 transcriptional regulator of multiple drug interactions, and we confirmed experimentally that 43 Rv1353c up-regulation reduces the antagonism of the bedaquiline-streptomycin combination. 44 Retrospective analysis of 57 clinical trials of TB regimens using INDIGO-MTB revealed that 45 synergistic combinations were significantly more efficacious than antagonistic combinations (p-46 value = 1 x 10 -4 ) based on the percentage of patients with negative sputum cultures after 8 47 weeks of treatment. Our study establishes a framework for rapid assessment of TB drug 48 combinations and is also applicable to other bacterial pathogens. 49 50 IMPORTANCE 51 52 Multi-drug combination therapy is an important strategy for treating tuberculosis, the world's 53 deadliest bacterial infection. Long treatment durations and growing rates of drug resistance 54 have created an urgent need for new approaches to prioritize effective drug regimens. Hence, 55 we developed a computational model called INDIGO-MTB, which identifies synergistic drug 56 regimens from an immense set of possible drug combinations using pathogen response 57 transcriptome elicited by individual drugs. Although the underlying input data for INDIGO-MTB 58 was generated under in vitro broth culture conditions, the predictions from INDIGO-MTB 59 correlated significantly with in vivo drug regimen efficacy from clinical trials. INDIGO-MTB also 60identified the transcription factor Rv1353c as a regulator of multiple drug interaction outcomes, 61which could be targeted for rationally enhancing drug synergy. 62 63 64 65 66 67 68 69 70 71 131 gene associations that are correlated with synergy and antagonism. In the example above, MTB 132 upregulation of both gene 1 and gene 3 in response to the drugs measured in monotherapy is predictive 133 of antagonism when the drugs are combined. By perturbing individual genes and known targets of 134 Transcription Factors (TFs) in the model, we can infer the impact of individual gene and TF activity 135 respectively on drug interactions and subsequently engineer interaction outcomes.136 137 138

Carprofen elicits pleiotropic mechanisms of bactericidal action with the potential to reverse antimicrobial drug resistance in tuberculosis

“…These routes of tackling antimicrobial resistance offer great promise. 24 , 25 Verapamil is a Ca 2+ channel blocker used to treat cardiovascular disorders and has been shown to restore the susceptibility of MDR-TB strains to rifampicin, isoniazid and bedaquiline, among others. 26 A similar investigation on carprofen is required.…”

Transcriptome signature of cell viability predicts drug response and drug interaction for Tuberculosis

“…Using DRonA-predicted viability scores, MLSynergy accurately predicted synergy and antagony for 2- and 3-drug combinations. This performance compares to INDIGO-MTB 28 , an existing strategy that quantifies synergistic and antagonistic drug regimens using transcriptomes of Mtb treated with individual drugs, but only with drugs with known drug-drug interactions. INDIGO-MTB requires known drug-drug interactions to learn patterns and identify combinations most likely to be synergistic.…”

A multi-scale pipeline linking drug transcriptomics with pharmacokinetics predictsin vivointeractions of tuberculosis drugs

“…The large number of potential drug combinations greatly complicates TB treatment design 8 . Therapy involving drug combinations can lead to surprising non-linear effects; some drugs can enhance each other's action leading to higher potency (synergy), or drugs can interfere with their action leading to reduced potency (antagonism) 9,10 .…”

“…This study addresses these challenges by creating a multi-scale pipeline combining two cutting-edge computational approaches, operating at different biological scales, to evaluate combination therapies using drug transcriptomics and pharmacokinetics /pharmacodynamics (PK/PD) ( Figure 1). First, to rapidly predict drug-drug interactions (synergy/antagonism) among combinations of two or more drugs, we utilize the existing in silico tool -inferring drug interactions using chemogenomics and orthology (INDIGO) optimized for Mtb (INDIGO-MTB) 8,12 . INDIGO-MTB uses a training data of known drug interactions along with drug transcriptomics data as inputs.…”

“…We previously used INDIGO-MTB to identify synergistic drug regimens for treating TB from over a million possible drug combinations using the pathogen response transcriptome elicited by individual drugs. The INDIGO-MTB model contains 164 drugs with anti-TB activity and it accurately predicted novel interactions of two-drug and threedrug combinations in vitro 8 . Next, to predict in vivo interactions and efficacy, here we integrate INDIGO-MTB predicted drug interactions within an existing multi-scale model of pathogen-immune dynamics leading to granuloma formation, known as GranSim 22,[25][26][27][28] .…”