The Diversity Outbred Mouse Population Is an Improved Animal Model of Vaccination against Tuberculosis That Reflects Heterogeneity of Protection

Immune correlates of protection as a game changer in tuberculosis vaccine development

The absence of validated correlates of protection (CoPs) hampers the rational design and clinical development of new tuberculosis vaccines. In this review, we provide an overview of the potential CoPs in tuberculosis vaccine research. Major hindrances and potential opportunities are then discussed. Based on recent progress, it is reasonable to anticipate that success in the ongoing efforts to identify CoPs would be a game-changer in tuberculosis vaccine development.

Reappraising the Role of T Cell-Derived IFN-γ in Restriction of Mycobacterium tuberculosis in the Murine Lung

T cells producing IFN-γ have long been considered a stalwart for immune protection against Mycobacterium tuberculosis (Mtb), but their relative importance to pulmonary immunity has been challenged by murine studies that achieved protection by adoptively transferred Mtb-specific IFN-γ-/- T cells. Using IFN-γ-/- T cell chimeric mice and adoptive transfer of IFN-γ-/- T cells into TCRβ-/-δ-/- mice, we demonstrate that control of lung Mtb burden is in fact dependent on T cell-derived IFN-γ, and, furthermore, mice selectively deficient in T cell-derived IFN-γ develop exacerbated disease compared with T cell-deficient control animals, despite equivalent lung bacterial burdens. Deficiency in T cell-derived IFN-γ skews infected and bystander monocyte-derived macrophages to an alternative M2 phenotype and promotes neutrophil and eosinophil influx. Our studies support an important role for T cell-derived IFN-γ in pulmonary immunity against tuberculosis.

Immunological roads diverged: mapping tuberculosis outcomes in mice

The journey from phenotypic observation to causal genetic mechanism is a long and challenging road. For pathogens like Mycobacterium tuberculosis (Mtb), which causes tuberculosis (TB), host-pathogen coevolution has spanned millennia, costing millions of human lives. Mammalian models can systematically recapitulate host genetic variation, producing a spectrum of disease outcomes. Leveraging genome sequences and deep phenotyping data from infected mouse genetic reference populations (GRPs), quantitative trait locus (QTL) mapping approaches have successfully identified host genomic regions associated with TB phenotypes. Here, we review the ongoing optimization of QTL mapping study design alongside advances in mouse GRPs. These next-generation resources and approaches have enabled identification of novel host-pathogen interactions governing one of the most prevalent infectious diseases in the world today.

B cells in perivascular and peribronchiolar granuloma-associated lymphoid tissue and B-cell signatures identify asymptomatic Mycobacterium tuberculosis lung infection in Diversity Outbred mice

Because most humans resist Mycobacterium tuberculosis infection, there is a paucity of lung samples to study. To address this gap, we infected Diversity Outbred mice with M. tuberculosis and studied the lungs of mice in different disease states. After a low-dose aerosol infection, progressors succumbed to acute, inflammatory lung disease within 60 days, while controllers maintained asymptomatic infection for at least 60 days, and then developed chronic pulmonary tuberculosis (TB) lasting months to more than 1 year. Here, we identified features of asymptomatic M. tuberculosis infection by applying computational and statistical approaches to multimodal data sets. Cytokines and anti-M. tuberculosis cell wall antibodies discriminated progressors vs controllers with chronic pulmonary TB but could not classify mice with asymptomatic infection. However, a novel deep-learning neural network trained on lung granuloma images was able to accurately classify asymptomatically infected lungs vs acute pulmonary TB in progressors vs chronic pulmonary TB in controllers, and discrimination was based on perivascular and peribronchiolar lymphocytes. Because the discriminatory lesion was rich in lymphocytes and CD4 T cell-mediated immunity is required for resistance, we expected CD4 T-cell genes would be elevated in asymptomatic infection. However, the significantly different, highly expressed genes were from B-cell pathways (e.g., Bank1, Cd19, Cd79, Fcmr, Ms4a1, Pax5, and H2-Ob), and CD20+ B cells were enriched in the perivascular and peribronchiolar regions of mice with asymptomatic M. tuberculosis infection. Together, these results indicate that genetically controlled B-cell responses are important for establishing asymptomatic M. tuberculosis lung infection.

Preclinical murine models for the testing of antimicrobials against Mycobacterium abscessus pulmonary infections: Current practices and recommendations

Mycobacterium abscessus, a rapidly growing nontuberculous mycobacterium, is increasingly recognized as an important pathogen of the human lung, disproportionally affecting people with cystic fibrosis (CF) and other susceptible individuals with non-CF bronchiectasis and compromised immune functions. M. abscessus infections are extremely difficult to treat due to intrinsic resistance to many antibiotics, including most anti-tuberculous drugs. Current standard-of-care chemotherapy is long, includes multiple oral and parenteral repurposed drugs, and is associated with significant toxicity. The development of more effective oral antibiotics to treat M. abscessus infections has thus emerged as a high priority. While murine models have proven instrumental in predicting the efficacy of therapeutic treatments for M. tuberculosis infections, the preclinical evaluation of drugs against M. abscessus infections has proven more challenging due to the difficulty of establishing a progressive, sustained, pulmonary infection with this pathogen in mice. To address this issue, a series of three workshops were hosted in 2023 by the Cystic Fibrosis Foundation (CFF) and the National Institute of Allergy and Infectious Diseases (NIAID) to review the current murine models of M. abscessus infections, discuss current challenges and identify priorities toward establishing validated and globally harmonized preclinical models. This paper summarizes the key points from these workshops. The hope is that the recommendations that emerged from this exercise will facilitate the implementation of informative murine models of therapeutic efficacy testing across laboratories, improve reproducibility from lab-to-lab and accelerate preclinical-to-clinical translation.

Host and pathogen genetic diversity shape vaccine-mediated protection to Mycobacterium tuberculosis

To investigate how host and pathogen diversity govern immunity against Mycobacterium tuberculosis (Mtb), we performed a large-scale screen of vaccine-mediated protection against aerosol Mtb infection using three inbred mouse strains [C57BL/6 (B6), C3HeB/FeJ (C3H), Balb/c x 129/SvJ (C129F1)] and three Mtb strains (H37Rv, CDC1551, SA161) representing two lineages and distinct virulence properties. We compared three protective modalities, all of which involve inoculation with live mycobacteria: Bacillus Calmette-Guérin (BCG), the only approved TB vaccine, delivered either subcutaneously or intravenously, and concomitant Mtb infection (CoMtb), a model of pre-existing immunity in which a low-level Mtb infection is established in the cervical lymph node following intradermal inoculation. We examined lung bacterial burdens at early (Day 28) and late (Day 98) time points after aerosol Mtb challenge and histopathology at Day 98. We observed substantial heterogeneity in the reduction of bacterial load afforded by these modalities at Day 28 across the combinations and noted a strong positive correlation between bacterial burden in unvaccinated mice and the degree of protection afforded by vaccination. Although we observed variation in the degree of reduction in bacterial burdens across the nine mouse/bacterium strain combinations, virtually all protective modalities performed similarly for a given strain-strain combination. We also noted dramatic variation in histopathology changes driven by both host and bacterial genetic backgrounds. Vaccination improved pathology scores for all infections except CDC1551. However, the most dramatic impact of vaccination on lesion development occurred for the C3H-SA161 combination, where vaccination entirely abrogated the development of the large necrotic lesions that arise in unvaccinated mice. In conclusion, we find that substantial TB heterogeneity can be recapitulated by introducing variability in both host and bacterial genetics, resulting in changes in vaccine-mediated protection as measured both by bacterial burden as well as histopathology. These differences can be harnessed in future studies to identify immune correlates of vaccine efficacy.

HBB contributes to individualized aconitine-induced cardiotoxicity in mice via interfering with ABHD5/AMPK/HDAC4 axis

The root of Aconitum carmichaelii Debx. (Fuzi) is an herbal medicine used in China that exerts significant efficacy in rescuing patients from severe diseases. A key toxic compound in Fuzi, aconitine (AC), could trigger unpredictable cardiotoxicities with high-individualization, thus hinders safe application of Fuzi. In this study we investigated the individual differences of AC-induced cardiotoxicities, the biomarkers and underlying mechanisms. Diversity Outbred (DO) mice were used as a genetically heterogeneous model for mimicking individualization clinically. The mice were orally administered AC (0.3, 0.6, 0.9 mg· kg-1 ·d-1) for 7 d. We found that AC-triggered cardiotoxicities in DO mice shared similar characteristics to those observed in clinic patients. Most importantly, significant individual differences were found in DO mice (variation coefficients: 34.08%-53.17%). RNA-sequencing in AC-tolerant and AC-sensitive mice revealed that hemoglobin subunit beta (HBB), a toxic-responsive protein in blood with 89% homology to human, was specifically enriched in AC-sensitive mice. Moreover, we found that HBB overexpression could significantly exacerbate AC-induced cardiotoxicity while HBB knockdown markedly attenuated cell death of cardiomyocytes. We revealed that AC could trigger hemolysis, and specifically bind to HBB in cell-free hemoglobin (cf-Hb), which could excessively promote NO scavenge and decrease cardioprotective S-nitrosylation. Meanwhile, AC bound to HBB enhanced the binding of HBB to ABHD5 and AMPK, which correspondingly decreased HDAC-NT generation and led to cardiomyocytes death. This study not only demonstrates HBB achievement a novel target of AC in blood, but provides the first clue for HBB as a novel biomarker in determining the individual differences of Fuzi-triggered cardiotoxicity.

Monitoring Staphylococcus aureus nasal colonization murine model using a bioluminescent methicillin-resistant S. aureus (MRSA)

Staphylococcus aureus nasal carriage is considered a risk factor for infections, and the development of nasal decolonization strategies is highly relevant. Despite they are not naturally colonized by Staphylococcus, mice are a good model for S. aureus nasal colonization. Murine models are easy to manipulate, and inter-laboratory reproducibility makes them suitable for nasal colonization studies. Strategies using bioluminescent bacteria allow for the monitoring of infection over time without the need to sacrifice animals for bacterial quantification. In this study, we evaluated S. aureus nasal colonization in three mouse strains (BALB/c, C57BL/6, and Swiss Webster) using a bioluminescent strain (SAP231). In vitro, a visible Bioluminescent Signal Emission (BLSE) was observed until 106 bacteria and detected by IVIS® imaging system up to 104 cells. Animals were inoculated with one or two doses of approximately 109 colony-forming units (CFU) of SAP231. Swiss Webster mice showed the longest colonization time, with some animals presenting BLSE for up to 140 h. In addition, BLSE was higher in this strain. BALB/c and C57BL/6 strains showed consistent BLSE results for 48 h. BLSE intensity was higher in Swiss Webster inoculated with both doses. Three different positions for image capture were evaluated, with better results for the lateral and ventrodorsal positions. After the loss of BLSE, bacterial quantification was performed, and Swiss Webster mice presented more bacteria in the nasal cavity (approximately 105 CFU) than the other strains. Our results demonstrate that bioluminescent S. aureus allow monitoring of nasal colonization and estimation of the bacterial burden present in live animals until 48 h.

Re-appraising the role of T-cell derived interferon gamma in restriction of Mycobacterium tuberculosis in the murine lung: T-cell derived IFNγ is required to restrict pulmonary Mtb

T cells producing interferon gamma (IFNγ) have long been considered a stalwart for immune protection against Mycobacterium tuberculosis (Mtb), but their relative importance to pulmonary immunity has been challenged by murine studies which achieved protection by adoptively transferred Mtb-specific IFNγ-/- T cells. Using IFNγ-/- T cell chimeric mice and adoptive transfer of IFNγ-/- T cells into TCRβ-/-δ-/- mice, we demonstrate that control of lung Mtb burden is in fact dependent on T cell-derived IFNγ, and furthermore, mice selectively deficient in T cell-derived IFNγ develop exacerbated disease compared to T cell-deficient controls despite equivalent lung bacterial burdens. Deficiency in T cell-derived IFNγ skews infected and bystander monocyte-derived macrophages (MDMs) to an alternative M2 phenotype, and promotes neutrophil and eosinophil influx. Our studies support an important role for T cell-derived IFNγ in pulmonary immunity against TB.

Antigen identification strategies and preclinical evaluation models for advancing tuberculosis vaccine development

In its myriad devastating forms, Tuberculosis (TB) has existed for centuries, and humanity is still affected by it. Mycobacterium tuberculosis (M. tuberculosis), the causative agent of TB, was the foremost killer among infectious agents until the COVID-19 pandemic. One of the key healthcare strategies available to reduce the risk of TB is immunization with bacilli Calmette-Guerin (BCG). Although BCG has been widely used to protect against TB, reports show that BCG confers highly variable efficacy (0-80%) against adult pulmonary TB. Unwavering efforts have been made over the past 20 years to develop and evaluate new TB vaccine candidates. The failure of conventional preclinical animal models to fully recapitulate human response to TB, as also seen for the failure of MVA85A in clinical trials, signifies the need to develop better preclinical models for TB vaccine evaluation. In the present review article, we outline various approaches used to identify protective mycobacterial antigens and recent advancements in preclinical models for assessing the efficacy of candidate TB vaccines.

Multiple genetic loci influence vaccine-induced protection against Mycobacterium tuberculosis in genetically diverse mice

Mycobacterium tuberculosis (M.tb.) infection leads to over 1.5 million deaths annually, despite widespread vaccination with BCG at birth. Causes for the ongoing tuberculosis endemic are complex and include the failure of BCG to protect many against progressive pulmonary disease. Host genetics is one of the known factors implicated in susceptibility to primary tuberculosis, but less is known about the role that host genetics plays in controlling host responses to vaccination against M.tb. Here, we addressed this gap by utilizing Diversity Outbred (DO) mice as a small animal model to query genetic drivers of vaccine-induced protection against M.tb. DO mice are a highly genetically and phenotypically diverse outbred population that is well suited for fine genetic mapping. Similar to outcomes in people, our previous studies demonstrated that DO mice have a wide range of disease outcomes following BCG vaccination and M.tb. challenge. In the current study, we used a large population of BCG-vaccinated/M.tb.-challenged mice to perform quantitative trait loci mapping of complex infection traits; these included lung and spleen M.tb. burdens, as well as lung cytokines measured at necropsy. We found sixteen chromosomal loci associated with complex infection traits and cytokine production. QTL associated with bacterial burdens included a region encoding major histocompatibility antigens that are known to affect susceptibility to tuberculosis, supporting validity of the approach. Most of the other QTL represent novel associations with immune responses to M.tb. and novel pathways of cytokine regulation. Most importantly, we discovered that protection induced by BCG is a multigenic trait, in which genetic loci harboring functionally-distinct candidate genes influence different aspects of immune responses that are crucial collectively for successful protection. These data provide exciting new avenues to explore and exploit in developing new vaccines against M.tb.

Host Genetic Background Influences BCG-Induced Antibodies Cross-Reactive to SARS-CoV-2 Spike Protein

Mycobacterium bovis Bacillus Calmette-Guérin (BCG) protects against childhood tuberculosis; and unlike most vaccines, BCG broadly impacts immunity to other pathogens and even some cancers. Early in the COVID-19 pandemic, epidemiological studies identified a protective association between BCG vaccination and outcomes of SARS-CoV-2, but the associations in later studies were inconsistent. We sought possible reasons and noticed the study populations often lived in the same country. Since individuals from the same regions can share common ancestors, we hypothesized that genetic background could influence associations between BCG and SARS-CoV-2. To explore this hypothesis in a controlled environment, we performed a pilot study using Diversity Outbred mice. First, we identified amino acid sequences shared by BCG and SARS-CoV-2 spike protein. Next, we tested for IgG reactive to spike protein from BCG-vaccinated mice. Sera from some, but not all, BCG-vaccinated Diversity Outbred mice contained higher levels of IgG cross-reactive to SARS-CoV-2 spike protein than sera from BCG-vaccinated C57BL/6J inbred mice and unvaccinated mice. Although larger experimental studies are needed to obtain mechanistic insight, these findings suggest that genetic background may be an important variable contributing to different associations observed in human randomized clinical trials evaluating BCG vaccination on SARS-CoV-2 and COVID-19.

Host Genetic Variation Impacts SARS-CoV-2 Vaccination Response in the Diversity Outbred Mouse Population

The COVID-19 pandemic led to the rapid and worldwide development of highly effective vaccines against SARS-CoV-2. However, there is significant individual-to-individual variation in vaccine efficacy due to factors including viral variants, host age, immune status, environmental and host genetic factors. Understanding those determinants driving this variation may inform the development of more broadly protective vaccine strategies. While host genetic factors are known to impact vaccine efficacy for respiratory pathogens such as influenza and tuberculosis, the impact of host genetic variation on vaccine efficacy against COVID-19 is not well understood. To model the impact of host genetic variation on SARS-CoV-2 vaccine efficacy, while controlling for the impact of non-genetic factors, we used the Diversity Outbred (DO) mouse model. We found that DO mice immunized against SARS-CoV-2 exhibited high levels of variation in vaccine-induced neutralizing antibody responses. While the majority of the vaccinated mice were protected from virus-induced disease, similar to human populations, we observed vaccine breakthrough in a subset of mice. Importantly, we found that this variation in neutralizing antibody, virus-induced disease, and viral titer is heritable, indicating that the DO serves as a useful model system for studying the contribution of genetic variation of both vaccines and disease outcomes.

Systems genetics uncover new loci containing functional gene candidates in Mycobacterium tuberculosis-infected Diversity Outbred mice

Mycobacterium tuberculosis, the bacillus that causes tuberculosis (TB), infects 2 billion people across the globe, and results in 8-9 million new TB cases and 1-1.5 million deaths each year. Most patients have no known genetic basis that predisposes them to disease. We investigated the complex genetic basis of pulmonary TB by modelling human genetic diversity with the Diversity Outbred mouse population. When infected with M. tuberculosis, one-third develop early onset, rapidly progressive, necrotizing granulomas and succumb within 60 days. The remaining develop non-necrotizing granulomas and survive longer than 60 days. Genetic mapping using clinical indicators of disease, granuloma histopathological features, and immune response traits identified five new loci on mouse chromosomes 1, 2, 4, 16 and three previously identified loci on chromosomes 3 and 17. Quantitative trait loci (QTLs) on chromosomes 1, 16, and 17, associated with multiple correlated traits and had similar patterns of allele effects, suggesting these QTLs contain important genetic regulators of responses to M. tuberculosis. To narrow the list of candidate genes in QTLs, we used a machine learning strategy that integrated gene expression signatures from lungs of M. tuberculosis-infected Diversity Outbred mice with gene interaction networks, generating functional scores. The scores were then used to rank candidates for each mapped trait in each locus, resulting in 11 candidates: Ncf2, Fam20b, S100a8, S100a9, Itgb5, Fstl1, Zbtb20, Ddr1, Ier3, Vegfa, and Zfp318. Importantly, all 11 candidates have roles in infection, inflammation, cell migration, extracellular matrix remodeling, or intracellular signaling. Further, all candidates contain single nucleotide polymorphisms (SNPs), and some but not all SNPs were predicted to have deleterious consequences on protein functions. Multiple methods were used for validation including (i) a statistical method that showed Diversity Outbred mice carrying PWH/PhJ alleles on chromosome 17 QTL have shorter survival; (ii) quantification of S100A8 protein levels, confirming predicted allele effects; and (iii) infection of C57BL/6 mice deficient for the S100a8 gene. Overall, this work demonstrates that systems genetics using Diversity Outbred mice can identify new (and known) QTLs and new functionally relevant gene candidates that may be major regulators of granuloma necrosis and acute inflammation in pulmonary TB.

Key advances in vaccine development for tuberculosis-success and challenges

Breakthrough findings in the clinical and preclinical development of tuberculosis (TB) vaccines have galvanized the field and suggest, for the first time since the development of bacille Calmette-Guérin (BCG), that a novel and protective TB vaccine is on the horizon. Here we highlight the TB vaccines that are in the development pipeline and review the basis for optimism in both the clinical and preclinical space. We describe immune signatures that could act as immunological correlates of protection (CoP) to facilitate the development and comparison of vaccines. Finally, we discuss new animal models that are expected to more faithfully model the pathology and complex immune responses observed in human populations.

Animal models for COVID-19 and tuberculosis

Respiratory infections cause tremendous morbidity and mortality worldwide. Amongst these diseases, tuberculosis (TB), a bacterial illness caused by Mycobacterium tuberculosis which often affects the lung, and coronavirus disease 2019 (COVID-19) caused by the Severe Acute Respiratory Syndrome Coronavirus type 2 (SARS-CoV-2), stand out as major drivers of epidemics of global concern. Despite their unrelated etiology and distinct pathology, these infections affect the same vital organ and share immunopathogenesis traits and an imperative demand to model the diseases at their various progression stages and localizations. Due to the clinical spectrum and heterogeneity of both diseases experimental infections were pursued in a variety of animal models. We summarize mammalian models employed in TB and COVID-19 experimental investigations, highlighting the diversity of rodent models and species peculiarities for each infection. We discuss the utility of non-human primates for translational research and emphasize on the benefits of non-conventional experimental models such as livestock. We epitomize advances facilitated by animal models with regard to understanding disease pathophysiology and immune responses. Finally, we highlight research areas necessitating optimized models and advocate that research of pulmonary infectious diseases could benefit from cross-fertilization between studies of apparently unrelated diseases, such as TB and COVID-19.

Transcriptional signatures measured in whole blood correlate with protection against tuberculosis in inbred and outbred mice

Although BCG has been used for almost 100 years to immunize against Mycobacterium tuberculosis, TB remains a global public health threat. Numerous clinical trials are underway studying novel vaccine candidates and strategies to improve or replace BCG, but vaccine development still lacks a well-defined set of immune correlates to predict vaccine-induced protection against tuberculosis. This study aimed to address this gap by examining transcriptional responses to BCG vaccination in C57BL/6 inbred mice, coupled with protection studies using Diversity Outbred mice. We evaluated relative gene expression in blood obtained from vaccinated mice, because blood is easily accessible, and data can be translated to human studies. We first determined that the average peak time after vaccination is 14 days for gene expression of a small subset of immune-related genes in inbred mice. We then performed global transcriptomic analyses using whole blood samples obtained two weeks after mice were vaccinated with BCG. Using comparative bioinformatic analyses and qRT-PCR validation, we developed a working correlate panel of 18 genes that were highly correlated with administration of BCG but not heat-killed BCG. We then tested this gene panel using BCG-vaccinated Diversity Outbred mice and revealed associations between the expression of a subset of genes and disease outcomes after aerosol challenge with M. tuberculosis. These data therefore demonstrate that blood-based transcriptional immune correlates measured within a few weeks after vaccination can be derived to predict protection against M. tuberculosis, even in outbred populations.

Recombinant BCG expressing the LTAK63 adjuvant increased memory T cells and induced long-lasting protection against Mycobacterium tuberculosis challenge in mice

Vaccine-induced protection against Mycobacterium tuberculosis (Mtb) is usually ascribed to the induction of Th1, Th17, and CD8⁺ T cells. However, protective immune responses should also involve other immune cell subsets, such as memory T cells. We have previously shown improved protection against Mtb challenge using the rBCG-LTAK63 vaccine (a recombinant BCG strain expressing the LTAK63 adjuvant, a genetically detoxified derivative of the A subunit from E. coli heat-labile toxin). Here we show that mice immunized with rBCG-LTAK63 exhibit a long-term (at least until 6 months) polyfunctional Th1/Th17 response in the draining lymph nodes and in the lungs. This response was accompanied by the increased presence of a diverse set of memory T cells, including central memory, effector memory and tissue-resident memory T cells. After the challenge, the T cell phenotype in the lymph nodes and lungs were characterized by a decrease in central memory T cells, and an increase in effector memory T cells and effector T cells. More importantly, when challenged 6 months after the immunization, this group demonstrated increased protection in comparison to BCG. In conclusion, this work provides experimental evidence in mice that the rBCG-LTAK63 vaccine induces a persistent increase in memory and effector T cell numbers until at least 6 months after immunization, which correlates with increased protection against Mtb. This improved immune response may contribute to enhance the long-term protection.

Variation in histone configurations correlates with gene expression across nine inbred strains of mice

The Diversity Outbred (DO) mice and their inbred founders are widely used models of human disease. However, although the genetic diversity of these mice has been well documented, their epigenetic diversity has not. Epigenetic modifications, such as histone modifications and DNA methylation, are important regulators of gene expression and, as such, are a critical mechanistic link between genotype and phenotype. Therefore, creating a map of epigenetic modifications in the DO mice and their founders is an important step toward understanding mechanisms of gene regulation and the link to disease in this widely used resource. To this end, we performed a strain survey of epigenetic modifications in hepatocytes of the DO founders. We surveyed four histone modifications (H3K4me1, H3K4me3, H3K27me3, and H3K27ac), as well as DNA methylation. We used ChromHMM to identify 14 chromatin states, each of which represents a distinct combination of the four histone modifications. We found that the epigenetic landscape is highly variable across the DO founders and is associated with variation in gene expression across strains. We found that epigenetic state imputed into a population of DO mice recapitulated the association with gene expression seen in the founders, suggesting that both histone modifications and DNA methylation are highly heritable mechanisms of gene expression regulation. We illustrate how DO gene expression can be aligned with inbred epigenetic states to identify putative cis-regulatory regions. Finally, we provide a data resource that documents strain-specific variation in the chromatin state and DNA methylation in hepatocytes across nine widely used strains of laboratory mice.

Immune evasion and provocation by Mycobacterium tuberculosis

Mycobacterium tuberculosis, the causative agent of tuberculosis, has infected humans for millennia. M. tuberculosis is well adapted to establish infection, persist in the face of the host immune response and be transmitted to uninfected individuals. Its ability to complete this infection cycle depends on it both evading and taking advantage of host immune responses. The outcome of M. tuberculosis infection is often a state of equilibrium characterized by immunological control and bacterial persistence. Recent data have highlighted the diverse cell populations that respond to M. tuberculosis infection and the dynamic changes in the cellular and intracellular niches of M. tuberculosis during the course of infection. M. tuberculosis possesses an arsenal of protein and lipid effectors that influence macrophage functions and inflammatory responses; however, our understanding of the role that specific bacterial virulence factors play in the context of diverse cellular reservoirs and distinct infection stages is limited. In this Review, we discuss immune evasion and provocation by M. tuberculosis during its infection cycle and describe how a more detailed molecular understanding is crucial to enable the development of novel host-directed therapies, disease biomarkers and effective vaccines.

Recent Developments in Mycobacteria-Based Live Attenuated Vaccine Candidates for Tuberculosis

Vaccination is an excellent approach to stimulating the host immune response and reducing human morbidity and mortality against microbial infections, such as tuberculosis (TB). Bacillus Calmette-Guerin (BCG) is the most widely administered vaccine in the world and the only vaccine approved by the World Health Organization (WHO) to protect against TB. Although BCG confers "protective" immunity in children against the progression of Mycobacterium tuberculosis (Mtb) infection into active TB, this vaccine is ineffective in protecting adults with active TB manifestations, such as multiple-, extensive-, and total-drug-resistant (MDR/XDR/TDR) cases and the co-existence of TB with immune-compromising health conditions, such as HIV infection or diabetes. Moreover, BCG can cause disease in individuals with HIV infection or other immune compromises. Due to these limitations of BCG, novel strategies are urgently needed to improve global TB control measures. Since live vaccines elicit a broader immune response and do not require an adjuvant, developing recombinant BCG (rBCG) vaccine candidates have received significant attention as a potential replacement for the currently approved BCG vaccine for TB prevention. In this report, we aim to present the latest findings and outstanding questions that we consider worth investigating regarding novel mycobacteria-based live attenuated TB vaccine candidates. We also specifically discuss the important features of two key animal models, mice and rabbits, that are relevant to TB vaccine testing. Our review emphasizes that the development of vaccines that block the reactivation of latent Mtb infection (LTBI) into active TB would have a significant impact in reducing the spread and transmission of Mtb. The results and ideas discussed here are only based on reports from the last five years to keep the focus on recent developments.

Pre-clinical development of a vaccine for human lymphatic filariasis

This study was conducted to optimize a fusion protein vaccine for translational development as a vaccine against the human tropical parasitic infection, lymphatic filariasis (LF). The vaccine candidate, His-tagged rBmHAXT was developed previously in our laboratory and was tested in various animal models including mouse, gerbils and Rhesus macaque where it exhibited significant levels of vaccine-induced protection. However, for commercial manufacturing and for regulatory approval for human use, there was a need to modify the vaccine antigen and its production and analytical release methods. Therefore, the major focus of this study was to develop a process for manufacturing an affinity tag-free rBmHAXT and evaluate its immunogenicity, potency and protective efficacy in both inbred and outbred mouse models, as well as in outbred gerbil models. Our results demonstrate that the tag-free rBmHAXT vaccine produced with a process suitable for cGMP production had protective properties equivalent to the original His-tagged rBmHAXT.

An interaction of inorganic arsenic exposure with body weight and composition on type 2 diabetes indicators in Diversity Outbred mice

Type 2 diabetes (T2D) is a complex metabolic disorder with no cure and high morbidity. Exposure to inorganic arsenic (iAs), a ubiquitous environmental contaminant, is associated with increased T2D risk. Despite growing evidence linking iAs exposure to T2D, the factors underlying inter-individual differences in susceptibility remain unclear. This study examined the interaction between chronic iAs exposure and body composition in a cohort of 75 Diversity Outbred mice. The study design mimics that of an exposed human population where the genetic diversity of the mice provides the variation in response, in contrast to a design that includes untreated mice. Male mice were exposed to iAs in drinking water (100 ppb) for 26 weeks. Metabolic indicators used as diabetes surrogates included fasting blood glucose and plasma insulin (FBG, FPI), blood glucose and plasma insulin 15 min after glucose challenge (BG15, PI15), homeostatic model assessment for [Formula: see text]-cell function and insulin resistance (HOMA-B, HOMA-IR), and insulinogenic index. Body composition was determined using magnetic resonance imaging, and the concentrations of iAs and its methylated metabolites were measured in liver and urine. Associations between cumulative iAs consumption and FPI, PI15, HOMA-B, and HOMA-IR manifested as significant interactions between iAs and body weight/composition. Arsenic speciation analyses in liver and urine suggest little variation in the mice's ability to metabolize iAs. The observed interactions accord with current research aiming to disentangle the effects of multiple complex factors on T2D risk, highlighting the need for further research on iAs metabolism and its consequences in genetically diverse mouse strains.

A diversity outbred F1 mouse model identifies host-intrinsic genetic regulators of response to immune checkpoint inhibitors

Immune checkpoint inhibitors (ICI) have improved outcomes for a variety of malignancies; however, many patients fail to benefit. While tumor-intrinsic mechanisms are likely involved in therapy resistance, it is unclear to what extent host genetic background influences response. To investigate this, we utilized the Diversity Outbred (DO) and Collaborative Cross (CC) mouse models. DO mice are an outbred stock generated by crossbreeding eight inbred founder strains, and CC mice are recombinant inbred mice generated from the same eight founders. We generated 207 DOB6F1 mice representing 48 DO dams and demonstrated that these mice reliably accept the C57BL/6-syngeneic B16F0 tumor and that host genetic background influences response to ICI. Genetic linkage analysis from 142 mice identified multiple regions including one within chromosome 13 that associated with therapeutic response. We utilized 6 CC strains bearing the positive (NZO) or negative (C57BL/6) driver genotype in this locus. We found that 2/3 of predicted responder CCB6F1 crosses show reproducible ICI response. The chromosome 13 locus contains the murine prolactin family, which is a known immunomodulating cytokine associated with various autoimmune disorders. To directly test whether prolactin influences ICI response rates, we implanted inbred C57BL/6 mice with subcutaneous slow-release prolactin pellets to induce mild hyperprolactinemia. Prolactin augmented ICI response against B16F0, with increased CD8 infiltration and 5/8 mice exhibiting slowed tumor growth relative to controls. This study highlights the role of host genetics in ICI response and supports the use of F1 crosses in the DO and CC mouse populations as powerful cancer immunotherapy models.

Leveraging Antibody, B Cell and Fc Receptor Interactions to Understand Heterogeneous Immune Responses in Tuberculosis

Despite over a century of research, Mycobacterium tuberculosis (Mtb), the causative agent of tuberculosis (TB), continues to kill 1.5 million people annually. Though less than 10% of infected individuals develop active disease, the specific host immune responses that lead to Mtb transmission and death, as well as those that are protective, are not yet fully defined. Recent immune correlative studies demonstrate that the spectrum of infection and disease is more heterogenous than has been classically defined. Moreover, emerging translational and animal model data attribute a diverse immune repertoire to TB outcomes. Thus, protective and detrimental immune responses to Mtb likely encompass a framework that is broader than T helper type 1 (Th1) immunity. Antibodies, Fc receptor interactions and B cells are underexplored host responses to Mtb. Poised at the interface of initial bacterial host interactions and in granulomatous lesions, antibodies and Fc receptors expressed on macrophages, neutrophils, dendritic cells, natural killer cells, T and B cells have the potential to influence local and systemic adaptive immune responses. Broadening the paradigm of protective immunity will offer new paths to improve diagnostics and vaccines to reduce the morbidity and mortality of TB.

BCG and BCGΔBCG1419c transiently protect hypersusceptible I/St mice and induce different influx of macrophages and neutrophils during pulmonary tuberculosis

Background. Host genetic factors influence both susceptibility to Mycobacterium tuberculosis infection and immune responses generated by vaccination. Genetically susceptible mice help to study mechanisms of immune protection which may differ from those operating in more resistant models.Methods. In this work, we compared the efficacy of protection conferred by subcutaneous vaccination of hypersusceptible I/St mice with BCG and the first-generation, hygromycin resistant version of the vaccine candidate BCGΔBCG1419c, against tuberculosis (TB), measured as survival, weight loss and replication in lungs. We further characterized the relative presence of immune cells in lungs.Results. We found that in I/St mice, vaccination with BCG or BCGΔBCG1419c provided similar level of protection against TB-driven weight loss and M. tuberculosis replication in lungs, while prolonging median survival time compared with unvaccinated controls. Despite affording similar protection to parental BCG, BCGΔBCG1419c led to a reduced presence of macrophages in lungs during early TB and to an increased neutrophil recruitment to the lungs during chronic TB.Conclusions. BCGΔBCG1419c protects I/St mice in a different manner than wild-type BCG against pulmonary TB by promoting different influx of macrophages and neutrophils at distinct times post-infection. These findings prompt us to suggest that preclinical evaluation of novel TB vaccine candidates should include evaluation of efficacy not only in commonly used resistant inbred mice, but also in susceptible hosts, to further determine their potential application to populations varying in their genetic. This would likely impact their intended use depending on host resistance or susceptibility to TB.

Utilizing preclinical models of genetic diversity to improve translation of phytochemical activities from rodents to humans and inform personalized nutrition

Mouse models are an essential tool in different areas of research, including nutrition and phytochemical research. Traditional inbred mouse models have allowed the discovery of therapeutical targets and mechanisms of action and expanded our knowledge of health and disease. However, these models lack the genetic variability typically found in human populations, which hinders the translatability of the results found in mice to humans. The development of genetically diverse mouse models, such as the collaborative cross (CC) or the diversity outbred (DO) models, has been a useful tool to overcome this obstacle in many fields, such as cancer, immunology and toxicology. However, these tools have not yet been widely adopted in the field of phytochemical research. As demonstrated in other disciplines, use of CC and DO models has the potential to provide invaluable insights for translation of phytochemicals from rodents to humans, which are desperately needed given the challenges and numerous failed clinical trials in this field. These models may prove informative for personalized use of phytochemicals in humans, including: predicting interindividual variability in phytochemical bioavailability and efficacy, identifying genetic loci or genes governing response to phytochemicals, identifying phytochemical mechanisms of action and therapeutic targets, and understanding the impact of genetic variability on individual response to phytochemicals. Such insights would prove invaluable for personalized implementation of phytochemicals in humans. This review will focus on the current work performed with genetically diverse mouse populations, and the research opportunities and advantages that these models can offer to phytochemical research.

Advancing Adjuvants for Mycobacterium tuberculosis Therapeutics

Tuberculosis (TB) remains one of the leading causes of death worldwide due to a single infectious disease agent. BCG, the only licensed vaccine against TB, offers limited protection against pulmonary disease in children and adults. TB vaccine research has recently been reinvigorated by new data suggesting alternative administration of BCG induces protection and a subunit/adjuvant vaccine that provides close to 50% protection. These results demonstrate the need for generating adjuvants in order to develop the next generation of TB vaccines. However, development of TB-targeted adjuvants is lacking. To help meet this need, NIAID convened a workshop in 2020 titled “Advancing Vaccine Adjuvants for Mycobacterium tuberculosis Therapeutics”. In this review, we present the four areas identified in the workshop as necessary for advancing TB adjuvants: 1) correlates of protective immunity, 2) targeting specific immune cells, 3) immune evasion mechanisms, and 4) animal models. We will discuss each of these four areas in detail and summarize what is known and what we can advance on in order to help develop more efficacious TB vaccines.

Metabolomic Analysis of Diverse Mice Reveals Hepatic Arginase-1 as Source of Plasma Arginase in Plasmodium chabaudi Infection

Infections disrupt host metabolism, but the factors that dictate the nature and magnitude of metabolic change are incompletely characterized. To determine how host metabolism changes in relation to disease severity in murine malaria, we performed plasma metabolomics on eight Plasmodium chabaudi-infected mouse strains with diverse disease phenotypes. We identified plasma metabolic biomarkers for both the nature and severity of different malarial pathologies. A subset of metabolic changes, including plasma arginine depletion, match the plasma metabolomes of human malaria patients, suggesting new connections between pathology and metabolism in human malaria. In our malarial mice, liver damage, which releases hepatic arginase-1 (Arg1) into circulation, correlated with plasma arginine depletion. We confirmed that hepatic Arg1 was the primary source of increased plasma arginase activity in our model, which motivates further investigation of liver damage in human malaria patients. More broadly, our approach shows how leveraging phenotypic diversity can identify and validate relationships between metabolism and the pathophysiology of infectious disease. IMPORTANCE Malaria is a severe and sometimes fatal infectious disease endemic to tropical and subtropical regions. Effective vaccines against malaria-causing Plasmodium parasites remain elusive, and malaria treatments often fail to prevent severe disease. Small molecules that target host metabolism have recently emerged as candidates for therapeutics in malaria and other diseases. However, our limited understanding of how metabolites affect pathophysiology limits our ability to develop new metabolite therapies. By providing a rich data set of metabolite-pathology correlations and by validating one of those correlations, our work is an important step toward harnessing metabolism to mitigate disease. Specifically, we showed that liver damage in P. chabaudi-infected mice releases hepatic arginase-1 into circulation, where it may deplete plasma arginine, a candidate malaria therapeutic that mitigates vascular stress. Our data suggest that liver damage may confound efforts to increase levels of arginine in human malaria patients.

CXCL1: A new diagnostic biomarker for human tuberculosis discovered using Diversity Outbred mice

More humans have died of tuberculosis (TB) than any other infectious disease and millions still die each year. Experts advocate for blood-based, serum protein biomarkers to help diagnose TB, which afflicts millions of people in high-burden countries. However, the protein biomarker pipeline is small. Here, we used the Diversity Outbred (DO) mouse population to address this gap, identifying five protein biomarker candidates. One protein biomarker, serum CXCL1, met the World Health Organization’s Targeted Product Profile for a triage test to diagnose active TB from latent M.tb infection (LTBI), non-TB lung disease, and normal sera in HIV-negative, adults from South Africa and Vietnam. To find the biomarker candidates, we quantified seven immune cytokines and four inflammatory proteins corresponding to highly expressed genes unique to progressor DO mice. Next, we applied statistical and machine learning methods to the data, i.e., 11 proteins in lungs from 453 infected and 29 non-infected mice. After searching all combinations of five algorithms and 239 protein subsets, validating, and testing the findings on independent data, two combinations accurately diagnosed progressor DO mice: Logistic Regression using MMP8; and Gradient Tree Boosting using a panel of 4: CXCL1, CXCL2, TNF, IL-10. Of those five protein biomarker candidates, two (MMP8 and CXCL1) were crucial for classifying DO mice; were above the limit of detection in most human serum samples; and had not been widely assessed for diagnostic performance in humans before. In patient sera, CXCL1 exceeded the triage diagnostic test criteria (>90% sensitivity; >70% specificity), while MMP8 did not. Using Area Under the Curve analyses, CXCL1 averaged 94.5% sensitivity and 88.8% specificity for active pulmonary TB (ATB) vs LTBI; 90.9% sensitivity and 71.4% specificity for ATB vs non-TB; and 100.0% sensitivity and 98.4% specificity for ATB vs normal sera. Our findings overall show that the DO mouse population can discover diagnostic-quality, serum protein biomarkers of human TB.

More humans die of tuberculosis (TB) than any other infectious disease, yet diagnostic tools remain limited. Here, we used the Diversity Outbred mouse population to discover candidate protein biomarkers of human TB. By applying statistical methods and machine learning to multidimensional data, we identified CXCL1 and MMP8 as the two most promising protein biomarker candidates. When evaluated in samples from human patients, CXCL1 achieved the World Health Organization’s targeted profile for a triage diagnostic test, discriminating active TB from important clinical differential diagnoses: latent Mtb infection and non-TB lung disease in HIV-negative adults. Overall, our studies show how a translationally relevant animal population model can accelerate TB biomarker discovery, validation, and testing for humans.

Three Immunocompetent Small Animal Models That Do Not Support Zika Virus Infection

Zika virus (ZIKV) is a mosquito-borne flavivirus that is primarily transmitted to humans through the bite of an infected mosquito. ZIKV causes disease in infected humans with added complications of Guillain-Barré syndrome and birth defects in infants born to mothers infected during pregnancy. There are several large immunocompetent animal models for ZIKV including non-human primates (NHPs). NHP models closely reflect human infection; however, due to sample size restrictions, investigations into the effects of transmission route and the impacts on disease dynamics have been understudied. Mice have been widely used for modeling ZIKV infection, yet there are few ZIKV-susceptible immunocompetent mouse models and none of these have been used to investigate sexual transmission. In an effort to identify a small immunocompetent animal model to characterize sexual transmission of ZIKV, we attempt experimental infection of multimammate mice, New Zealand white rabbits, and Hartley guinea pigs. The multimammate mouse is the natural reservoir of Lassa fever virus and has been identified to harbor other human pathogens. Likewise, while NZW rabbits are susceptible to West Nile virus, they have not yet been examined for their susceptibility to infection with ZIKV. Guinea pigs have been successfully used as models for ZIKV infection, but only in immunocompromised life stages (young or pregnant). Here, it was found that the multimammate mouse and New Zealand White (NZW) rabbits are not susceptible ZIKV infection as determined by a lack viral RNA in tissues and fluids collected. Sexually mature male Hartley guinea pigs were inoculated subcutaneously and by mosquito bite, but found to be refractory to ZIKV infection, contrary to findings of other studies in young and pregnant guinea pigs. Interestingly, here it is shown that adult male guinea pigs are not susceptible to ZIKV infection, even when infected by natural route (e.g., mosquito bite). Although a new small animal model for the sexual transmission for ZIKV was not established through this study, these findings provide information on outbred animal species that are not permissive to infection (NZW rabbits and multimammate mice) and new information surrounding limitations of a previously established animal model (guinea pigs).

Aging, But Not Sex and Genetic Diversity, Impacts the Pathobiology of Bacterial Endophthalmitis

Purpose

Age, sex, and genetics are important biological variables in determining an individual's susceptibility or response to infectious agents; however, their role has not been evaluated in intraocular infections. In this study, we comprehensively examined the impact of these host biological factors in the pathogenesis of experimental bacterial endophthalmitis.

Methods

Endophthalmitis was induced by intravitreal injection of bacteria (Staphylococcusaureus) in the eyes of male and female C57BL/6 mice of different ages: group I (young, 6–8 weeks), group II (mid-age, 18–20 weeks), and group III (old, 1 year). Highly heterogeneous outbred J:DO mice were used for genetic diversity analysis. Eyes were subjected to clinical examination, retinal function testing using electroretinography (ERG), histopathological analysis (hematoxylin and eosin staining), and bacterial burden estimation. The levels of inflammatory mediators were measured using qPCR and ELISA, and the infiltration of neutrophils was determined by flow cytometry.

Results

Both inbred C57BL/6 and diversity outbred (J:DO) mice were equally susceptible to S. aureus endophthalmitis, as evidenced by a time-dependent increase in clinical scores, bacterial burden, intraocular inflammation, and retinal tissue damage, in addition to decreased retinal function. However, no significant differences were observed in disease severity and innate responses in male versus female mice. Older mice (group III) exhibited higher clinical scores coinciding with increased bacterial proliferation and intraocular inflammation, resulting in enhanced disease severity. Moreover, bone-marrow-derived macrophages from old mice exhibited reduced phagocytic activity but increased inflammatory response toward S. aureus challenge.

Conclusions

Age, but not sex, is an important biological variable in bacterial endophthalmitis. Identification of pathways underlying altered innate immunity and impaired bacterial clearance in aging eyes could provide new insights into the pathobiology of intraocular infections in elderly patients.

Deep learning predicts gene expression as an intermediate data modality to identify susceptibility patterns in Mycobacterium tuberculosis infected Diversity Outbred mice

BACKGROUND

Machine learning sustains successful application to many diagnostic and prognostic problems in computational histopathology. Yet, few efforts have been made to model gene expression from histopathology. This study proposes a methodology which predicts selected gene expression values (microarray) from haematoxylin and eosin whole-slide images as an intermediate data modality to identify fulminant-like pulmonary tuberculosis ('supersusceptible') in an experimentally infected cohort of Diversity Outbred mice (n=77).

METHODS

Gradient-boosted trees were utilized as a novel feature selector to identify gene transcripts predictive of fulminant-like pulmonary tuberculosis. A novel attention-based multiple instance learning model for regression was used to predict selected genes' expression from whole-slide images. Gene expression predictions were shown to be sufficiently replicated to identify supersusceptible mice using gradient-boosted trees trained on ground truth gene expression data.

FINDINGS

The model was accurate, showing high positive correlations with ground truth gene expression on both cross-validation (n = 77, 0.63 ≤ ρ ≤ 0.84) and external testing sets (n = 33, 0.65 ≤ ρ ≤ 0.84). The sensitivity and specificity for gene expression predictions to identify supersusceptible mice (n=77) were 0.88 and 0.95, respectively, and for an external set of mice (n=33) 0.88 and 0.93, respectively.

IMPLICATIONS

Our methodology maps histopathology to gene expression with sufficient accuracy to predict a clinical outcome. The proposed methodology exemplifies a computational template for gene expression panels, in which relatively inexpensive and widely available tissue histopathology may be mapped to specific genes' expression to serve as a diagnostic or prognostic tool.

FUNDING

National Institutes of Health and American Lung Association.

One Size Fits All? Not in In Vivo Modeling of Tuberculosis Chemotherapeutics

Tuberculosis (TB) remains a global health problem despite almost universal efforts to provide patients with highly effective chemotherapy, in part, because many infected individuals are not diagnosed and treated, others do not complete treatment, and a small proportion harbor Mycobacterium tuberculosis (Mtb) strains that have become resistant to drugs in the standard regimen. Development and approval of new drugs for TB have accelerated in the last 10 years, but more drugs are needed due to both Mtb's development of resistance and the desire to shorten therapy to 4 months or less. The drug development process needs predictive animal models that recapitulate the complex pathology and bacterial burden distribution of human disease. The human host response to pulmonary infection with Mtb is granulomatous inflammation usually resulting in contained lesions and limited bacterial replication. In those who develop progressive or active disease, regions of necrosis and cavitation can develop leading to lasting lung damage and possible death. This review describes the major vertebrate animal models used in evaluating compound activity against Mtb and the disease presentation that develops. Each of the models, including the zebrafish, various mice, guinea pigs, rabbits, and non-human primates provides data on number of Mtb bacteria and pathology resolution. The models where individual lesions can be dissected from the tissue or sampled can also provide data on lesion-specific bacterial loads and lesion-specific drug concentrations. With the inclusion of medical imaging, a compound's effect on resolution of pathology within individual lesions and animals can also be determined over time. Incorporation of measurement of drug exposure and drug distribution within animals and their tissues is important for choosing the best compounds to push toward the clinic and to the development of better regimens. We review the practical aspects of each model and the advantages and limitations of each in order to promote choosing a rational combination of them for a compound's development.

Lactate Metabolism and Signaling in Tuberculosis and Cancer: A Comparative Review

Infection with Mycobacterium tuberculosis (Mtb) leading to tuberculosis (TB) disease continues to be a major global health challenge. Critical barriers, including but not limited to the development of multi-drug resistance, lack of diagnostic assays that detect patients with latent TB, an effective vaccine that prevents Mtb infection, and infectious and non-infectious comorbidities that complicate active TB, continue to hinder progress toward a TB cure. To complement the ongoing development of new antimicrobial drugs, investigators in the field are exploring the value of host-directed therapies (HDTs). This therapeutic strategy targets the host, rather than Mtb, and is intended to augment host responses to infection such that the host is better equipped to prevent or clear infection and resolve chronic inflammation. Metabolic pathways of immune cells have been identified as promising HDT targets as more metabolites and metabolic pathways have shown to play a role in TB pathogenesis and disease progression. Specifically, this review highlights the potential role of lactate as both an immunomodulatory metabolite and a potentially important signaling molecule during the host response to Mtb infection. While long thought to be an inert end product of primarily glucose metabolism, the cancer research field has discovered the importance of lactate in carcinogenesis and resistance to chemotherapeutic drug treatment. Herein, we discuss similarities between the TB granuloma and tumor microenvironments in the context of lactate metabolism and identify key metabolic and signaling pathways that have been shown to play a role in tumor progression but have yet to be explored within the context of TB. Ultimately, lactate metabolism and signaling could be viable HDT targets for TB; however, critical additional research is needed to better understand the role of lactate at the host-pathogen interface during Mtb infection before adopting this HDT strategy.

Automatic discovery of clinically interpretable imaging biomarkers for Mycobacterium tuberculosis supersusceptibility using deep learning

BACKGROUND

Identifying which individuals will develop tuberculosis (TB) remains an unresolved problem due to few animal models and computational approaches that effectively address its heterogeneity. To meet these shortcomings, we show that Diversity Outbred (DO) mice reflect human-like genetic diversity and develop human-like lung granulomas when infected with Mycobacterium tuberculosis (M.tb) .

METHODS

Following M.tb infection, a "supersusceptible" phenotype develops in approximately one-third of DO mice characterized by rapid morbidity and mortality within 8 weeks. These supersusceptible DO mice develop lung granulomas patterns akin to humans. This led us to utilize deep learning to identify supersusceptibility from hematoxylin & eosin (H&E) lung tissue sections utilizing only clinical outcomes (supersusceptible or not-supersusceptible) as labels.

FINDINGS

The proposed machine learning model diagnosed supersusceptibility with high accuracy (91.50 ± 4.68%) compared to two expert pathologists using H&E stained lung sections (94.95% and 94.58%). Two non-experts used the imaging biomarker to diagnose supersusceptibility with high accuracy (88.25% and 87.95%) and agreement (96.00%). A board-certified veterinary pathologist (GB) examined the imaging biomarker and determined the model was making diagnostic decisions using a form of granuloma necrosis (karyorrhectic and pyknotic nuclear debris). This was corroborated by one other board-certified veterinary pathologist. Finally, the imaging biomarker was quantified, providing a novel means to convert visual patterns within granulomas to data suitable for statistical analyses.

IMPLICATIONS

Overall, our results have translatable implication to improve our understanding of TB and also to the broader field of computational pathology in which clinical outcomes alone can drive automatic identification of interpretable imaging biomarkers, knowledge discovery, and validation of existing clinical biomarkers.

FUNDING

National Institutes of Health and American Lung Association.