Independent genomic polymorphisms in the PknH serine threonine kinase locus during evolution of the Mycobacterium tuberculosis Complex affect virulence and host preference

RESEARCH ARTICLE Independent genomic polymorphisms in the PknH serine threonine kinase locus during evolution of the Mycobacterium tuberculosis Complex affect virulence and host preference Elena Mata 1,2 , Damien FarrellID 3 , Ruoyao MaID 3 , Santiago UrangaID 1,2 , Ana Belen Gomez 1,2 , Marta Monzon 4 , Juan BadiolaID 4 , Alberto Anel 5 , Jesu ´s GonzaloAsensio 1,2,6 , Carlos MartinID 1,2,7 , Stephen V. GordonID 3‡ , Nacho AguiloID 1,2‡ * 1Grupo de Gene ´tica de Micobacterias, Universidad de Zaragoza/ISS Aragon, Zaragoza, Spain, 2CIBER Enfermedades Respiratorias, Instituto de Salud Carlos III, Madrid, Spain, 3School of Veterinary Medicine, Veterinary Science Centre, University College Dublin, Dublin, Ireland, 4Research Centre for Encephalopathies and Transmissible Emerging Diseases, Universidad de Zaragoza, Zaragoza, Spain, 5Grupo Apoptosis, Inmunidad y Ca ´ncer, IIS Arago ´n. Dpto. Bioquı´mica y Biologı´a Molecular y Celular, Fac. Ciencias, Universidad de Zaragoza, Zaragoza, Spain, 6Instituto de Biocomputacio ´n y Fı ´sica de Sistemas Complejos (BIFI), Zaragoza, Spain, 7Servicio de Microbiologı ´a, Hospital Universitario Miguel Servet, ISS Aragon, Spain ‡ These authors are joint senior authors on this work. *[email protected] Abstract Species belonging to the Mycobacterium tuberculosis Complex (MTBC) show more than 99% genetic identity but exhibit distinct host preference and virulence. The molecular genetic changes that underly host specificity and infection phenotype within MTBC members have not been fully elucidated. Here, we analysed RD900 genomic region across MTBC members using whole genome sequences from 60 different MTBC strains so as to determine its role in the context of MTBC evolutionary history. The RD900 region comprises two homologous genes, pknH1 and pknH2, encoding a serine/threonine protein kinase PknH flanking the tbd2 gene. Our analysis revealed that RD900 has been independently lost in different MTBC lineages and different strains, resulting in the generation of a single pknH gene. Importantly, all the analysed M.bovis and M.caprae strains carry a conserved deletion within a proline rich-region of pknH, independent of the presence or absence of RD900. We hypothesized that deletion of pknH proline rich-region in M.bovis may affect PknH function, having a potential role in its virulence and evolutionary adaptation. To explore this hypothesis, we constructed two M.bovis ‘knock-in’ strains containing the M.tuberculosis pknH gene. Evaluation of their virulence phenotype in mice revealed a reduced virulence of both M.bovis knock-in strains compared to the wild type, suggesting that PknH plays an important role in the differential virulence phenotype of M.bovis vs M.tuberculosis. PLOS PATHOGENS PLOS Pathogens | https://doi.org/10.1371/journal.ppat.1009061 December 21, 2020 1 / 24 a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 OPEN ACCESS Citation: Mata E, Farrell D, Ma R, Uranga S, Gomez AB, Monzon M, et al. (2020) Independent genomic polymorphisms in the PknH serine threonine kinase locus during evolution of the Mycobacterium tuberculosis Complex affect virulence and host preference. PLoS Pathog 16(12): e1009061. https://doi.org/10.1371/journal. ppat.1009061 Editor: Marcel A. Behr, McGill UniversityHealth Centre, CANADA Received: July 21, 2020 Accepted: October 9, 2020 Published: December 21, 2020 Peer Review History: PLOS recognizes the benefits of transparency in the peer review process; therefore, we enable the publication of all of the content of peer review and author responses alongside final, published articles. The editorial history of this article is available here: https://doi.org/10.1371/journal.ppat.1009061 Copyright: ©2020 Mata et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Author summary Tuberculosis is caused in humans and animals by organisms from the Mycobacterium tuberculosis Complex (MTBC), that share more than 99% genetic identity but exhibit distinct host preference and virulence. While Mycobacterium tuberculosis is the main causative agent of human TB, Mycobacterium bovis is responsible for bovine TB disease, that exacts a tremendous economic burden worldwide, as well as being a zoonotic threat. Unlike the human restriction of M.tuberculosis,M.bovis has a broader host range and it has been found to be more virulent than M.tuberculosis in different animal models. However, the molecular basis for host preference and virulence divergence between M.tuberculosis and M.bovis is not fully elucidated. Here we study the genetic variations of the genomic region RD900 in the context of MTBC phylogeny. RD900 contains two genes encoding orthologues of the serine/threonine kinase PknH, which is linked to the regulation of several bacterial processes including virulence. We found that M.bovis pknH genes show a conserved deletion that is not present in M.tuberculosis strains, and we evaluated the potential impact of these variations in the regulation of M.bovis vs M.tuberculosis virulence through the construction and in vivo characterization of M.bovis pknH mutant strains. Introduction The Mycobacterium tuberculosis Complex (MTBC) is composed of several highly genetically related mycobacterial species (>99% nucleotide sequence identity) that infect and cause tuberculosis (TB) in a range of mammalian hosts. MTBC species can be broadly classified into human-adapted mycobacteria, comprising the obligate human pathogens Mycobacterium tuberculosis sensu stricto and Mycobacterium africanum, and animal-adapted mycobacteria, referring to those that can propagate and transmit in a range of wild and domesticated animal hosts. Despite their high degree of genomic identity, MTBC members exhibit important differences in relation to host range, transmissibility, pathogenesis and virulence [1,2]. Mycobacterium bovis is the causative agent of bovine TB, which is responsible for high economic losses in livestock productivity and disease control costs and remains a major problem in many developed livestock-producing countries as well as in developing countries [3]. Unlike the human restriction of M.tuberculosis,M.bovis has a broader host range. Although it mainly sustains infection in cattle, M.bovis infection affects many other mammalian species and can be found in a range of maintenance hosts such as deer, possums and badgers that act as wildlife reservoirs. The route of transmission from these reservoir species to cattle and vice versa is not fully elucidated, but it is proposed that close contact, aerosol and faecal-oral routes play a role [3,4]. Furthermore, M.bovis poses a risk as a zoonosis for humans, mainly through the consumption of raw milk or close contact with infected cattle. However, transmission of M.bovis among immunocompetent human hosts is uncommon, representing a dead end or spillover host [5]. Conversely, M.tuberculosis has been independently shown to be attenuated in cattle [6,7], while M.bovis strains show greater virulence than M.tuberculosis in animal models such as mice or rabbits [8–10]. M.tuberculosis has evolved an extensive signalling network that includes, among others, 11 serine threonine protein kinases (STPKs) involved in the regulation of many aspects of bacterial physiology. STPKs have been shown to play a crucial role in the growth and survival of M. tuberculosis during infection, controlling a variety of cellular processes related to cell division, cell wall biosynthesis, cellular metabolism, transcriptional regulation and virulence [11]. PknH is a transmembrane STPK that presents a typical conserved N-terminal kinase domain PLOS PATHOGENS PknH locus differences in the context of Mycobacterium tuberculosis Complex evolution PLOS Pathogens | https://doi.org/10.1371/journal.ppat.1009061 December 21, 2020 2 / 24 Data Availability Statement: The raw sequencing data are available in the Sequence Read Archive (SRA) under the bioproject ID PRJNA645286. Funding: CM and NA received a grant from Spanish Ministry of “Ciencia, Innovacio ´n y Universidades” (www.ciencia.gob.es/portal/site/ MICINN/) [grant number RTI2018-097625-B-I00]. SVG acknowledges funding from Science Foundation Ireland (www.sfi.ie/) (grant number 15/ IA/3154). RM was a recipient of a China Scholarship Council studentship (www. chinesescholarshipcouncil.com/). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Competing interests: The authors have declared that no competing interests exist. followed by a proline-rich region, a predicted transmembrane single helix domain and an extracellular C-terminal sensor domain. It has been shown that PknH phosphorylates itself and other substrates, and it plays a role in cell wall biosynthesis, response to nitric oxide stress and the dormancy response [12–15]. PknH has also been implicated in virulence, as an M. tuberculosis H37Rv pknH knock-out mutant was shown to be hypervirulent in a mouse model [16]. The organization of the genome region that contains the pknH gene exhibits substantial variation across MTBC strains. The M.tuberculosis H37Rv genome contains a single copy of the pknH gene which is located downstream of embR, one of the substrates phosphorylated by PknH [12]. Genome sequencing of the M.africanum GM041182 strain revealed a different genetic organization to that observed in M.tuberculosis H37Rv, showing two different pknH genes (pknH1 and pknH2) that flank a potential ABC transporter named tbd2. This observation led to the description of a new Region of Difference, named RD900, that was first described as a lineage specific locus. The RD900 locus is 3141 bp long and contains a single complete gene (MAF12860, tbd2) and the 3’ end of another (MAF12870, pknH2) [17]. This region was initially thought to be deleted in M.bovis and M.tuberculosis “modern” lineages. However, although RD900 locus was not present in the original M.bovis AF2122/97 genome annotation, resequencing of this genome revealed that the RD900 locus is actually present in this M.bovis strain [18]. Both pknH1 and pknH2 genes present in M.africanum GM041182 and M.bovis AF2122/97 exhibit a high degree of identity between their kinase domains, but they have different sensor domains; this may affect sensing of environment signals by PknH1 and PknH2, and consequently may result in differential cellular responses. The proline-rich region of M.africanum pknH2 has a deletion of about 100 bp compared to pknH1 and to the M.tuberculosis H37Rv pknH orthologue. The same proline-rich region is deleted in M.bovis AF2122/97 pknH2, but is also deleted in pknH1 [17,18]. Here, we analyzed the RD900 genomic region across MTBC members to determine its plasticity across MTBC evolutionary history, using whole genome sequencing data from 60 representative MTBC strains. In addition, we performed a comparative study between M. tuberculosis,M.bovis and M.africanum strains regarding their virulence and dissemination capacity in mice. We also hypothesized that the PknH proline rich-region deletion found in M.bovis could have an impact on the enhanced virulence associated with these strains. To elucidate this hypothesis, we constructed two M.bovis strains in which we introduced the M. tuberculosis pknH1 gene, and determined their virulence phenotype in mice. We finally used transcriptomics to assess the impact of introduction of the M.tuberculosis pknH1 gene on global gene expression in the recombinant M.bovis. Results Independent RD900 deletion events across MTBC and a conserved deletion in M.bovis and M.caprae pknH genes In order to delineate variation in the RD900 locus and pknH genes across the different MTBC lineages, and the potential evolutionary impact of this variation, we performed an analysis of this genomic region using WGS data from 60 isolates representative of the different MTBC lineages (S1 Table). The original sequence reads of the different MTBC strains were downloaded in fastq format from GenBank and aligned to the reference sequence of M.africanum GM041182 RD900 locus, including the tbd2 gene, both pknH genes (pknH1 and pknH2) and the two flanking genes embR and MAF_RS06695. The resulting alignments were first analysed PLOS PATHOGENS PknH locus differences in the context of Mycobacterium tuberculosis Complex evolution PLOS Pathogens | https://doi.org/10.1371/journal.ppat.1009061 December 21, 2020 3 / 24 for the presence or absence of RD900, and for the presence or absence of the deletion within the proline-rich region in the pknH genes. The RD900 region was found to be present in all analysed genomes of M.canettii, a ‘smooth’ colony species thought to be similar to the progenitor of the MTBC. The M.canettii genome shows the same genomic organization as M.africanum: the tbd2 gene flanked by two pknH genes and a deletion within the proline-rich region of pknH2 but not pknH1 (Fig 1A). The same locus organization was present in the analysed M.tuberculosis Lineage 7 isolates. However, the presence of RD900 was found to be variable among M.tuberculosis Lineage 1 strains. Some of the analysed isolates contained an intact RD900 region whereas others had a single complete pknH gene containing an intact proline region sequence. As previously described, RD900 was found to be deleted in all the analysed isolates from M.tuberculosis “modern” lineages (Lineages 2, 3 and 4) which are distributed worldwide. In contrast, RD900 was found to be present in all the analysed M.africanum Lineage 5 and Lineage 6 isolates, although variability was observed within the proline-rich region sequence of pknH1 which was intact in some isolates and deleted in others. In the animal adapted species, the RD900 region was found to be present in the analysed isolates from M.mungi,M.microti and M.pinnipedii, with the sequence encoding the prolinerich region deleted in pknH2 but not pknH1. In contrast, RD900 was found to be deleted in M. orygis, which presented a single pknH gene containing the complete proline-rich encoding region. Finally, the presence of RD900 in M.caprae,M.bovis and BCG strains was found to be variable among the different isolates analysed. In all of these latter strains, both with or without the RD900 region, we found that the different pknH gene copies had the proline-rich region deleted. The only exception came from one M.bovis strain (M.bovis B2 7505) in which RD900 was deleted and the resulting single copy of pknH had a complete proline rich region. This strain was isolated from a human patient in Uganda [19] and was subsequently described to show M.tuberculosis RD patterns [20]; hence, it appears actually to be an M.tuberculosis strain. In order to better understand the evolutionary process that this particular region underwent across the MTBC members, we analysed the presence of point mutations (synonymous, missense and frameshift mutations) in the RD900 locus across the analysed strains through a variant and consequence calling analysis. The results for the missense and frameshift mutations are represented in a clustered heat map showing the presence/absence data for each mutation (Fig 1B). Mutations present in only a single isolate from the same lineage were excluded. In the clustered heatmap it can be observed that some isolates of the same lineage are clustered together whereas other isolates from the same lineage are not clustered, reflecting the interstrain variability of this region across the different MTBC lineages. The M.bovis and M.caprae strains with an intact RD900 region were more closely related than those strains in which the RD900 region was deleted, reflecting their different evolutionary pathways. Finally, in order to confirm the results obtained from the genomic analysis of pknH genes, and focusing on the differences between M.tuberculosis and M.bovis strains, we designed specific primers to verify the presence or absence of the proline-rich region in M.bovis pknH genes compared to M.tuberculosis. To confirm the presence of this deletion, a PCR was performed using a primer pair flanking the region of the deletion. This PCR was performed using DNA extracts from M.tuberculosis H37Rv and different M.bovis strains: the reference strains AF2122/97 and AN5; the multidrug resistant strain M.bovis B, responsible for large TB outbreaks in Spain [21]; and a BCG strain. An amplification product of lower molecular weight was obtained from the DNA extracts of all the tested M.bovis strains compared to M.tuberculosis H37Rv, reflecting the deletion of the proline-rich region in M.bovis strains (Fig 1C). PLOS PATHOGENS PknH locus differences in the context of Mycobacterium tuberculosis Complex evolution PLOS Pathogens | https://doi.org/10.1371/journal.ppat.1009061 December 21, 2020 4 / 24 PLOS PATHOGENS PknH locus differences in the context of Mycobacterium tuberculosis Complex evolution PLOS Pathogens | https://doi.org/10.1371/journal.ppat.1009061 December 21, 2020 5 / 24 M.bovis disseminates more efficiently than M.tuberculosis and M. africanum in a mouse model We next evaluated in vivo infection and dissemination differences between MTBC members by performing a comparative virulence experiment in mice using M.tuberculosis,M.bovis and M.africanum strains. C57BL/6 mice were infected by the intranasal route with a low dose challenge (*200 CFUs). We used GFP-expressing strains so that we could assess cell-to-cell spread by flow cytometry. The mycobacterial strains used were the M.tuberculosis reference laboratory strain H37Rv, the M.tuberculosis clinical isolate Mt103 [22], the M.bovis reference strain AF2122/97 and the M.africanum clinical isolate HCU2828 (Fig 2A). Our results showed a higher bacterial load in lungs from mice infected with M.bovis compared to those infected with M.tuberculosis or M.africanum strains. This was observed both at one and four months post-challenge (Fig 2B). Accordingly, lung histopathology analysis revealed more extensive inflammatory foci with inflammatory infiltrate in lung parenchyma along with tissue degeneration in the M.bovis-infected group compared to other groups (Fig 2C). We also assessed bacterial dissemination into spleen, liver and kidneys. We found a significantly higher bacillary load in kidneys at both time points in M.bovis compared to M.tuberculosis and M.africanum infected mice. In the case of the liver, significantly higher M.bovis dissemination was evident at 4 months post-infection. In general, these results suggested a greater ability of M.bovis to disseminate and persist in these non-lymphoid organs. No significant differences were observed in CFU counts from spleen, although bacillary loads were slightly higher in M.bovis infected mice (Fig 2D). We next studied the ability of the different MTBC species to disseminate cell-to-cell within the lung. We analysed by flow cytometry the percentages of GFP-positive cells in cellular suspensions from lungs. At one month post-infection, the proportion of GFP+ cells was found to be higher in the M.bovis group, which reflected a greater ability of M.bovis to establish at the primary site of infection than M.tuberculosis or M.africanum strains (Fig 2E and 2F). Total and GFP+ infected lung macrophage and neutrophil populations were analysed using specific surface markers for each population at one month post-infection. In accordance with the enhanced virulence phenotype of the M.bovis strain, we found a strongly increased neutrophil influx in this group (Fig 2G and 2H). Interestingly, different studies have shown an association, both in humans and animal models, between an increase of neutrophil infiltration in the lungs and a higher degree of TB disease [23–26]. Introduction of the M.tuberculosis pknH allele into M.bovis results in lower virulence in mice Our results indicated that all the MTBC species, except M.bovis and M.caprae, contained a pknH gene that included the proline-rich region. Thus, considering our findings showing the higher virulence of M.bovis, together with a previous study that described a more virulent phenotype of a pknH knock-out M.tuberculosis H37Rv strain [16], we hypothesized that the Fig 1. RD900 analysis across MTBC members. (A) Schematic representation of RD900 locus organization in MTBC. pknH genes are represented with the kinase domain in purple, the proline rich region in blue and the different sensor domain in orange for pknH1 and white for pknH2. The presence of the deletion in the proline rich region is represented with an “X”. The number of analysed isolates (n) for each MTBC specie is indicated. (B) Presence/absence of missense and frameshift mutations in RD900 locus across MTBC. Clustered heat map constructed from the presence/absence data of missense and frameshift mutations in RD900 locus from the MTBC analysed isolates. The presence or absence of RD900 and the deletion within the proline-rich region in PknH1 is also indicated. (C) Verification of the presence of proline-rich region deletion in M.bovis AF2122/97, AN5, MBZ and BCG and M.tuberculosis H37Rv pknH genes by PCR using a primer pair flanking the deleted region. https://doi.org/10.1371/journal.ppat.1009061.g001 PLOS PATHOGENS PknH locus differences in the context of Mycobacterium tuberculosis Complex evolution PLOS Pathogens | https://doi.org/10.1371/journal.ppat.1009061 December 21, 2020 6 / 24 PLOS PATHOGENS PknH locus differences in the context of Mycobacterium tuberculosis Complex evolution PLOS Pathogens | https://doi.org/10.1371/journal.ppat.1009061 December 21, 2020 7 / 24 deletion within the proline-rich region of M.bovis pknH genes may lead to an inactive or differently regulated protein that in turn affects virulence. To address this hypothesis, we constructed two M.bovis strains in which the pknH gene from M.tuberculosis H37Rv was introduced into the chromosome using the integrative plasmid pMV361. The complemented M.bovis strains were the reference strain AF2122/91 and the AN5 strain. Both of them are RD900+ M.bovis strains, and therefore carry two homologous pknH copies, deleted for the proline-rich region (Fig 1C). To achieve comparable levels of pknH expression to those observed in H37Rv, the pknH gene was cloned under the control of its own promoter. The introduction of the pMV361 plasmid carrying the M.tuberculosis H37Rv pknH gene (pknH TB ) into the M.bovis strains was verified by colony PCR (Fig 3A). Using a primer pair flaking the proline-rich region deletion, the recombinant colonies showed two amplification products, one corresponding to the introduced complete pknH from M.tuberculosis H37Rv (442 bp) and the other one corresponding to the M.bovis pknH genes (352 bp). Once it was confirmed that the pknH TB gene was correctly integrated into the knock-in M.bovis strains (AF2122::pknH TB and AN5::pknH TB ), pknH TB expression was evaluated and showed comparable levels in both M.bovis knock-in strains and M.tuberculosis H37Rv (S1 Fig). Then, C57BL/6 mice were i.n. infected with a low dose (*200 CFUs) of the M.bovis AF2122/97 and AN5 WT strains and their corresponding pknH TB knock-in strains. Lung histopathology and bacterial burden determination in lungs and spleen were performed after 4 weeks of infection (Fig 3B). A significant reduction in bacterial load in lungs and spleen was observed for both AF2122::pknH TB and AN5::pknH TB strains compared to their wild type M. bovis parental strains (Fig 3C). In accordance with expectations, histopathology revealed lower inflammatory damage and tissue degeneration in lungs from the pknH TB knock-in strains (Fig 3D). To further characterize the phenotype of M.bovis::pknH TB in vivo, we constructed a GFPexpressing version of the strain, and we thereafter infected mice with these strains. We first confirmed that lung bacterial burden profile was similar to that observed with non-fluorescent strains at one month post challenge, to check that introduction of the new plasmid did not alter strain virulence (Fig 3F). Analyses of lung cellular suspensions revealed a significantly higher percentage of GFP+ infected lung cells in M.bovis AF2122 infected mice than in M. bovis AF2122::pknH TB and M.tuberculosis H37Rv infected mice (Fig 3G). The analysis of lung infected populations also showed a higher percentage of infected neutrophils in mice infected with the AF2122 wild-type strain (Fig 3H). In this experiment we evaluated antigen-specific IFNγresponses following ex vivo stimulation of lung cells with a preparation of tuberculosis antigens (Purified Protein Derivative, PPD). We found a stronger response in cellular suspensions from M.bovis AF2122-infected mice compared to M.bovis pknH TB knock-in and M. tuberculosis groups (Fig 3I). Lung IFNγresponses have been described as a surrogate marker Fig 2. M.bovis exhibits greater virulence than M.tuberculosis and M.africanum in the mouse model. (A) Schematic representation of the infection model in C57BL/6 mice. (B) Bacterial burden in lungs at one at four months post-infection. (C) Representative images (1.5x) of heamtoxilin-eosin staining of lungs from mice infected with the different pathogens at one and four months post-infection. (D) Bacterial burden in spleen, liver and kidneys at one and four months post-infection. (E) Representative dot-plots of GFP+-infected lung cells from lung cell suspensions of M.tuberculosis,M.bovis and M.africanum infected mice at one month post-infection. (F) Percentages of GFP+-infected lung cells from lung cell suspensions of M.tuberculosis,M.bovis and M.africanum infected mice at one month post-infection. (G) Cell number of GFP+-infected lung macrophages and neutrophils from lung cell suspensions of M. tuberculosis,M.bovis and M.africanum infected mice at one month post-infection. (H) Percentages of total CD45+ macrophages and neutrophils from lung cell suspensions of M.tuberculosis,M.bovis and M.africanum infected mice at one month post-infection. Data are represented as mean ±SEM from two independent experiments (n = 6 mice/group). Statistical analysis was performed by two-way ANOVA (B,D), one-way ANOVA (F) and multiple unpaired T-test (G,H,I). �p<0.05; ��p<0.01; ���p<0.001; ����p<0.0001. https://doi.org/10.1371/journal.ppat.1009061.g002 PLOS PATHOGENS PknH locus differences in the context of Mycobacterium tuberculosis Complex evolution PLOS Pathogens | https://doi.org/10.1371/journal.ppat.1009061 December 21, 2020 8 / 24 PLOS PATHOGENS PknH locus differences in the context of Mycobacterium tuberculosis Complex evolution PLOS Pathogens | https://doi.org/10.1371/journal.ppat.1009061 December 21, 2020 9 / 24 Ultimately however, an evolutionary successful strategy depends on host factors as well as pathogen virulence, the interaction of which drives transmission and maintenance of the pathogen in the host population. Indeed, M.bovis is able to spread and sustain across different animal populations, both domesticated (e.g. cattle) and wild (e.g. badgers and deer), yet humanto-human transmission of M.bovis is a rare event. Thus, one could hypothesize that M.bovis has evolved to exploit dissemination routes distinct from respiratory transmission. Conversely, M.tuberculosis has evolved towards human adaptation to which respiratory dissemination is integral and which may rely on a functional PknH. In relation to the evolution of the distinct human vs animal adapted MTBC, the two component PhoPR system seems to have a crucial role. M.bovis strains contain point mutations in the region encoding the extracellular domain of PhoR, which impairs the signalling of this system. Thus, as described by Gonzalo-Asensio et al (34), the expression of phoP regulon genes is divergent in M.bovis as compared to M.tuberculosis, which could again play a role in their compromised capacity to spread via human-to-human. Interestingly, an M.bovis strain, called the ‘B strain’ caused a severe TB outbreak in Spain in the 1990s and disseminated efficiently among humans. This latter strain overexpressed the phoP gene and subsequently expressed the PhoP regulon at levels similar to M.tuberculosis, highting the linkages between this regulon and efficient human-to-human transmission. In summary, our results suggest a role for PknH in the differential regulation of virulence and infection in M.bovis and M.tuberculosis. The apparent alteration, or lack, of PknH activity in M.bovis leads to increased virulence in animal hosts. Indeed, this finding may help explain the reduced evidence for latency in M.bovis infection in animals, whereby infected livestock or wild mammals usually exhibit a more progressive and disseminated disease presentation as compared to human disease [53]. Our findings therefore have implications for our understanding of the genetic underpinnings and evolution of host adaptation in the MTBC. Materials and methods Ethics statement Experimental work was conducted in agreement with the Spanish Policy for Animal Protection RD53/2013 and the European Union Directive 2010/63 for the protection of animals used for experimental and other scientific purposes and experimental procedures were approved by the Ethics Committee for Animal Experiments of University of Zaragoza (CEA). Animals C57BL/6JR mice were purchased from Janvier Biolabs. All mice were housed and maintained in specific pathogen-free conditions and observed for any sign of disease in the facilities of Centro de Investigacio ´n de Encefalopatı ´as y Enfermedades Transmisibles Emergentes (ES 50 297 0012 009). Male and female mice between the ages of 8 to 10 weeks were used for all the experiments. Bacterial strains and culture conditions Mycobacterial strains were grown at 37˚C in Middlebrook 7H9 broth (BD Difco) liquid medium supplemented with 0.05% Tween 80 (Sigma) and 10% (v/v) Middlebrook albumin dextrose catalase enrichment (ADC; BD Biosciences 0.2% dextrose, 0.5% bovine serum albumin, 0.085% NaCl and 0.0003% beef catalase). For cultures in solid media, Middlebrook 7H10 agar (BD Difco) supplemented with 10% ADC (BD Biosciences) was used. When required, medium was supplemented with 20 μg/ml of kanamycin (Km) or 20 μg/ml of streptomycin PLOS PATHOGENS PknH locus differences in the context of Mycobacterium tuberculosis Complex evolution PLOS Pathogens | https://doi.org/10.1371/journal.ppat.1009061 December 21, 2020 16 / 24 (Sm). For the culture of M.bovis strains liquid and solid media was supplemented with sodium pyruvate (Sigma) 4 mg/ml. All GFP-expressing carry with the replicative plasmid pJKD6 encoding the green fluorescent protein (GFP), a kind gift from Luciana Leite, Butantan Institute, Brazil. This plasmid carries de gfp gene under the control of a strong promoter generated through error-prone PCR of the PL5 promoter from the mycobacteriophage L5 resulting in high levels of GFP expression [54]. Bacterial suspensions for intranasal infections were prepared in PBS from quantified glycerol stock solutions. Construction of M.bovis::pknH TB knock-in strains For the construction of the M.bovis::pknH TB knock-in strains, the pknH gene from M.tuberculosis H37Rv (pknH TB ) was cloned using the integrative plasmid pMV361. The gene insert was synthesized and subcloned in NheI site of pMV361 by Genscript Company (Piscataway, USA). M.bovis AF2122/97 and AN5 competent cells were obtained and electroporated with pMV361pknH TB plasmid. Oligonucleotides used for PCR verifications are detailed in S3 Table. GFP-expressing strains of M.tuberculosis H37Rv and Mt103, M.bovis AF2122/97 and M. africanum HCU2828 were generated by transformation by electroporation with the replicative plasmid pJKD6 carrying a kanamycin resistance cassette. GFP-expressing AF2122::pknH TB was generated through transformation by electroporation with the replicative plasmid pJKD6 carrying a streptomycin resistance cassette. Intranasal infection Mice were infected with a low dose (�100–200 CFUs) of the different mycobacterial strains by the intranasal route. The animals were anesthetized by inhalation route with Isofluorane (Isboa Vet) using a vaporizer and intranasal administration was performed with two instillations of 20 μl of the bacterial suspension prepared in PBS. Bacterial suspensions for infection were plated in solid agar medium to determine the CFUs used for in vivo challenges. Bacterial burden determination For bacterial burden determination in lung, spleen, liver and kidneys, the organs were aseptically removed and homogenized in 1 ml of H 2 O using a GentleMacs dissociator (Miltenyi Biotec). CFUs were determined by plating serial dilutions on solid 7H10 medium supplemented with 10% ADC and sodium pyruvate when required. Histological analysis For lung histopathology analysis, left lung was removed and fixed in 4% formaldehyde for 24 hours prior to hematoxylin-eosin (HE) staining. Histological staining was performed in the Pathological Anatomy Service from CIBA (Zaragoza, Spain) and images were obtained with a Leica DM5000B optical microscope. Lung single cell suspensions For lung cell suspensions, lungs were aseptically removed and homogenized in HEPES buffer (HEPES 10 mM, NaCl, 150 mM, KCl 5 mM, MgCl2 1 mM, CaCl2 1,8 mM pH 7,4) containing DNaseI (AppliChem) 40 IU/ml and Collagenase D (Roche) 2 mg/ml using a GentleMacs dissociator (Miltenyi Biotec) according to manufacturer instructions. Lungs were incubated at 37˚C for 30 minutes and further homogenized with the GentleMacs dissociator. The PLOS PATHOGENS PknH locus differences in the context of Mycobacterium tuberculosis Complex evolution PLOS Pathogens | https://doi.org/10.1371/journal.ppat.1009061 December 21, 2020 17 / 24 homogenates were filtered through a 70 μm cell strainer (MACS SmartStainers, Miltenyi Biotec) and centrifuge at 1500 rpm for 5 min. Red blood cells were lysed using Red Blood Cell Lysing Buffer (Sigma-Aldrich) and finally, cells were pelleted and resuspended in RPMI (Gibco) culture medium. Antibody staining and flow cytometry analysis Lung single cell suspensions were plated in U-bottom 96-well plates and incubated for 15 minutes at 4˚C with FcR blocking reagent (Miltenyi Biotec). Surface staining was performed for 20 min at 4˚ C using different combinations of antibodies to define lung myeloid populations of macrophages and neutrophils. Then, cells were washed and fixed with 4% paraformaldehyde for 30 minutes. Cells were acquired using a Gallios flow cytometer (Beckman Coulter) and the results were analysed using Weasel Software. For the analysis of lung myeloid populations, total lung macrophages were defined as CD45 +Ly6G-CD11c+ and neutrophils as CD45+Ly6G+CD11c-CD11b+ (S3 Fig). Cytokine determination Lung cell suspensions were plated in U-bottom 96-well plates and incubated for 48 h in the presence or absence of PPD 10 μg/ml. Supernatants were collected to determine IFNγ response. Quantification of IFNγwas performed using a specific commercial ELISA kit (Mabtech Biotech) according to manufacturer instructions. Statistics Mice were distributed in groups of at least 6 animals per cage prior to experimental procedures. Results were not blinded for analysis and randomization was not applicable to these studies. Rstudio and GraphPad Prism software (version 6) was used for graphical representation and statistical analysis. Statistical test used in each experiment are indicated in the figure legends. Outlier values were determined applying Grubb’s test. A p value of <0.05 was considered significant (�p<0.05, ��p<0.01,���p<0.001, ����p<0.0001). Comparative genomics Computational analysis of RD900 region across MTBC was performed using WGS data of 60 different MTBC isolates (S1 Table). We did not use annotations based on existing assemblies since misannotations are likely present in this locus due to the fact that homologous domains shared between pknH genes cannot be uniquely aligned with short reads. Such reads are marked as multi-mappers and ignored; resulting assemblies are therefore split into contigs over such a region and annotation software may miss genes. To avoid this, we downloaded the original raw sequence reads in fastq format from GenBank for each available isolate (S1 Table). The raw reads were aligned to a reference sequence (M.africanum GM041182 RD900 locus) representing the most complete version of the RD900 across the MTBC. In this way we could most readily see missing elements in the other species. Reads were aligned using bwa (mem algorithm) [55] and the resulting bam files were sorted and indexed using samtools. The alignments were visualized in IGV [56] to identify the presence or absence of RD900 and the presence of the deletion in the proline-rich region within pknH genes. Finally, a variant calling analysis was performed to identify point mutations present in this region from the different PLOS PATHOGENS PknH locus differences in the context of Mycobacterium tuberculosis Complex evolution PLOS Pathogens | https://doi.org/10.1371/journal.ppat.1009061 December 21, 2020 18 / 24 MTBC analysed species. Variant calling, filtering and consequence calling were performed using freebayes (https://arxiv.org/abs/1207.3907) and bcftools. RNA-seq analysis M.bovis AF2122 wild type and M.bovis AF2122::pknHMtb knock-in were grown in triplicate standard 7H9 medium with pyruvate (see Bacterial strains and culture conditions above) to mid-log phase (OD0.4–0.6). RNA was extracted by QIAzol lysis Reagent and RNeasy Mini Kit (Qiagen). DNA was removed by TURBO DNase (Thermo Fisher) and RNA Clean-Up and Concentrator (Zymo). The concentration and quality of all the RNA samples were checked using Nanodrop, Qubit (Thermo Fisher) and Agilent 2100 bioanalyzers. RNA was sent for commercial sequencing (Novogene) for library preparation and 150 bp paired-end reads sequencing on an Illumina HiSeq. RNA-seq data were first processed using FastQC (https://www.bioinformatics.babraham. ac.uk/projects/fastqc/) tool to check the quality of the raw reads. The raw reads were trimmed by Trim-galore (https://www.bioinformatics.babraham.ac.uk/projects/trim_galore/) to trim off reads with bad quality and adapters used for sequencing. Alignment to the reference genome M.bovis AF2122/97 was done using bwa mem [55]. Gene expression was determined from the resulting bam files using featureCounts [57]. The count data was then analysed for differential expression of genes between the wild type and mutant with the DESeq2 R package [58]. Significant genes were selected at a log2 fold change threshold of 2. Supporting information S1 Fig. Expression of pknH TB in M.bovis::pknH TB knock-in strains. (A, B) Confirmation of pknH TB presence in M.bovis::pknH TB strains. cDNA was analyzed by PCR, either with the reverse primer included inside the proline-rich region (A), or with both primers flanking this sequence (B). In the first case, cDNA amplification was only observed when pknH TB was expressed, whereas in the second case amplification occurred in both cases, with a different fragment size depending on the pknH expressed. (C) pknH TB expression was measured by qRT-PCR using the proline-rich region flanking primers. Graph represents mean±SD from one experiment with three replicates, corresponding to the fold-change value in comparison to H37Rv expression. (TIF) S2 Fig. Functional classification of differentially-expressed (DE) genes between AF2122 wild-type and pknH TB knock-in strains. (TIF) S3 Fig. Flow cytometry analysis for the identification of lung macrophages and neutrophil populations. (TIF) S1 Table. Genomic data of MTBC isolates used for the genomic analysis of RD900 locus. (DOCX) S2 Table. Filtered DE gene overlap between M.bovis::pknH TB vs M.bovis WT, and M.bovis WT vs M.tuberculosis.�Gene expression data from M.tuberculosis were obtained from Malone et al. [38]. (DOCX) PLOS PATHOGENS PknH locus differences in the context of Mycobacterium tuberculosis Complex evolution PLOS Pathogens | https://doi.org/10.1371/journal.ppat.1009061 December 21, 2020 19 / 24 S3 Table. Primer sequences used for PCR and qRT-PCR. (DOCX) Acknowledgments The authors acknowledge the Scientific and Technical Services from Instituto Aragone ´s de Ciencias de la Salud and Universidad de Zaragoza for their assistance. We thank Jeff Chen for advice and discussion. Author Contributions Conceptualization: Carlos Martin, Stephen V. Gordon, Nacho Aguilo. Data curation: Elena Mata, Damien Farrell, Ruoyao Ma. Formal analysis: Elena Mata, Damien Farrell, Stephen V. Gordon, Nacho Aguilo. Funding acquisition: Alberto Anel, Carlos Martin, Stephen V. Gordon, Nacho Aguilo. Investigation: Elena Mata, Damien Farrell, Ruoyao Ma, Santiago Uranga, Ana Belen Gomez. Methodology: Elena Mata, Nacho Aguilo. Project administration: Stephen V. Gordon, Nacho Aguilo. Resources: Ruoyao Ma, Marta Monzon, Juan Badiola, Carlos Martin, Stephen V. Gordon, Nacho Aguilo. Software: Elena Mata, Damien Farrell, Ruoyao Ma. Supervision: Alberto Anel, Carlos Martin, Stephen V. Gordon, Nacho Aguilo. Validation: Elena Mata. Visualization: Elena Mata, Stephen V. Gordon, Nacho Aguilo. Writing – original draft: Elena Mata, Alberto Anel, Jesu ´s Gonzalo-Asensio, Carlos Martin, Stephen V. Gordon, Nacho Aguilo. Writing – review & editing: Elena Mata, Carlos Martin, Stephen V. Gordon, Nacho Aguilo. References 1. Brites D, Gagneux S. The Nature and Evolution of Genomic Diversity in the Mycobacterium tuberculosis Complex. Advances in experimental medicine and biology. 2017; 1019:1–26. https://doi.org/10.1007/ 978-3-319-64371-7_1 PMID: 29116627. 2. Brosch R, Gordon SV, Marmiesse M, Brodin P, Buchrieser C, Eiglmeier K, et al. A new evolutionary scenario for the Mycobacterium tuberculosis complex. Proceedings of the National Academy of Sciences of the United States of America. 2002; 99(6):3684–9. https://doi.org/10.1073/pnas.052548299 PMID: 11891304; PubMed Central PMCID: PMC122584. 3. Malone KM, Gordon SV. Mycobacterium tuberculosis Complex Members Adapted to Wild and Domestic Animals. Advances in experimental medicine and biology. 2017; 1019:135–54. Epub 2017/11/09. https://doi.org/10.1007/978-3-319-64371-7_7 PMID: 29116633. 4. Allen AR. One bacillus to rule them all?—Investigating broad range host adaptation in Mycobacterium bovis. Infection, genetics and evolution: journal of molecular epidemiology and evolutionary genetics in infectious diseases. 2017; 53:68–76. https://doi.org/10.1016/j.meegid.2017.04.018 PMID: 28434972. 5. Berg S, Smith NH. Why doesn’t bovine tuberculosis transmit between humans? Trends in microbiology. 2014; 22(10):552–3. https://doi.org/10.1016/j.tim.2014.08.007 PMID: 25174642. 6. Villarreal-Ramos B, Berg S, Whelan A, Holbert S, Carreras F, Salguero FJ, et al. Experimental infection of cattle with Mycobacterium tuberculosis isolates shows the attenuation of the human tubercle bacillus for cattle. Sci Rep. 2018; 8(1):894. https://doi.org/10.1038/s41598-017-18575-5 PMID: 29343690; PubMed Central PMCID: PMC5772528. PLOS PATHOGENS PknH locus differences in the context of Mycobacterium tuberculosis Complex evolution PLOS Pathogens | https://doi.org/10.1371/journal.ppat.1009061 December 21, 2020 20 / 24 7. Whelan AO, Coad M, Cockle PJ, Hewinson G, Vordermeier M, Gordon SV. Revisiting host preference in the Mycobacterium tuberculosis complex: experimental infection shows M. tuberculosis H37Rv to be avirulent in cattle. PloS one. 2010; 5(1):e8527. Epub 2010/01/06. https://doi.org/10.1371/journal.pone. 0008527 PMID: 20049086; PubMed Central PMCID: PMC2795854. 8. Nedeltchev GG, Raghunand TR, Jassal MS, Lun S, Cheng QJ, Bishai WR. Extrapulmonary dissemination of Mycobacterium bovis but not Mycobacterium tuberculosis in a bronchoscopic rabbit model of cavitary tuberculosis. Infection and immunity. 2009; 77(2):598–603. https://doi.org/10.1128/IAI.0113208 PMID: 19064634; PubMed Central PMCID: PMC2632025. 9. Dunn PL, North RJ. Virulence ranking of some Mycobacterium tuberculosis and Mycobacterium bovis strains according to their ability to multiply in the lungs, induce lung pathology, and cause mortality in mice. Infection and immunity. 1995; 63(9):3428–37. Epub 1995/09/01. https://doi.org/10.1128/IAI.63.9. 3428-3437.1995 PMID: 7642273; PubMed Central PMCID: PMC173472. 10. Medina E, Ryan L, LaCourse R, North RJ. Superior virulence of Mycobacterium bovis over Mycobacterium tuberculosis (Mtb) for Mtb-resistant and Mtb-susceptible mice is manifest as an ability to cause extrapulmonary disease. Tuberculosis (Edinburgh, Scotland). 2006; 86(1):20–7. Epub 2005/10/29. https://doi.org/10.1016/j.tube.2005.04.003 PMID: 16253563. 11. Greenstein AE, Grundner C, Echols N, Gay LM, Lombana TN, Miecskowski CA, et al. Structure/function studies of Ser/Thr and Tyr protein phosphorylation in Mycobacterium tuberculosis. Journal of molecular microbiology and biotechnology. 2005; 9(3–4):167–81. https://doi.org/10.1159/000089645 PMID: 16415590. 12. Molle V, Kremer L, Girard-Blanc C, Besra GS, Cozzone AJ, Prost JF. An FHA phosphoprotein recognition domain mediates protein EmbR phosphorylation by PknH, a Ser/Thr protein kinase from Mycobacterium tuberculosis. Biochemistry. 2003; 42(51):15300–9. https://doi.org/10.1021/bi035150b PMID: 14690440. 13. Sharma K, Chandra H, Gupta PK, Pathak M, Narayan A, Meena LS, et al. PknH, a transmembrane Hank’s type serine/threonine kinase from Mycobacterium tuberculosis is differentially expressed under stress conditions. FEMS microbiology letters. 2004; 233(1):107–13. Epub 2004/03/27. https://doi.org/ 10.1016/j.femsle.2004.01.045 PMID: 15043876. 14. Sharma K, Gupta M, Pathak M, Gupta N, Koul A, Sarangi S, et al. Transcriptional control of the mycobacterial embCAB operon by PknH through a regulatory protein, EmbR, in vivo. Journal of bacteriology. 2006; 188(8):2936–44. https://doi.org/10.1128/JB.188.8.2936-2944.2006 PMID: 16585755; PubMed Central PMCID: PMC1446986. 15. Zheng X, Papavinasasundaram KG, Av-Gay Y. Novel substrates of Mycobacterium tuberculosis PknH Ser/Thr kinase. Biochemical and biophysical research communications. 2007; 355(1):162–8. Epub 2007/02/09. https://doi.org/10.1016/j.bbrc.2007.01.122 PMID: 17286964. 16. Papavinasasundaram KG, Chan B, Chung JH, Colston MJ, Davis EO, Av-Gay Y. Deletion of the Mycobacterium tuberculosis pknH gene confers a higher bacillary load during the chronic phase of infection in BALB/c mice. Journal of bacteriology. 2005; 187(16):5751–60. Epub 2005/08/04. https://doi.org/10. 1128/JB.187.16.5751-5760.2005 PMID: 16077122; PubMed Central PMCID: PMC1196067. 17. Bentley SD, Comas I, Bryant JM, Walker D, Smith NH, Harris SR, et al. The genome of Mycobacterium africanum West African 2 reveals a lineage-specific locus and genome erosion common to the M. tuberculosis complex. PLoS neglected tropical diseases. 2012; 6(2):e1552. Epub 2012/03/06. https://doi.org/ 10.1371/journal.pntd.0001552 PMID: 22389744; PubMed Central PMCID: PMC3289620. 18. Malone KM, Farrell D, Stuber TP, Schubert OT, Aebersold R, Robbe-Austerman S, et al. Updated Reference Genome Sequence and Annotation of Mycobacterium bovis AF2122/97. 2017; 5(14). https:// doi.org/10.1128/genomeA.00157-17 PMID: 28385856. 19. Wanzala SI, Nakavuma J, Travis DA, Kia P, Ogwang S, Sreevatsan S. Draft Genome Sequences of Mycobacterium bovis BZ 31150 and Mycobacterium bovis B2 7505, Pathogenic Bacteria Isolated from Archived Captive Animal Bronchial Washes and Human Sputum Samples in Uganda. Genome announcements. 2015; 3(5). https://doi.org/10.1128/genomeA.01102-15 PMID: 26450718; PubMed Central PMCID: PMC4599077. 20. Zimpel CK, Brandao PE, de Souza Filho AF, de Souza RF, Ikuta CY, Ferreira Neto JS, et al. Complete Genome Sequencing of Mycobacterium bovis SP38 and Comparative Genomics of Mycobacterium bovis and M. tuberculosis Strains. Frontiers in microbiology. 2017; 8:2389. https://doi.org/10.3389/ fmicb.2017.02389 PMID: 29259589; PubMed Central PMCID: PMC5723337. 21. Sagasti S, Millan-Lou MI, Soledad Jimenez M, Martin C, Samper S. In-depth analysis of the genome sequence of a clinical, extensively drug-resistant Mycobacterium bovis strain. Tuberculosis (Edinburgh, Scotland). 2016; 100:46–52. https://doi.org/10.1016/j.tube.2016.06.005 PMID: 27553409. 22. Jackson M, Raynaud C, Laneelle MA, Guilhot C, Laurent-Winter C, Ensergueix D, et al. Inactivation of the antigen 85C gene profoundly affects the mycolate content and alters the permeability of the PLOS PATHOGENS PknH locus differences in the context of Mycobacterium tuberculosis Complex evolution PLOS Pathogens | https://doi.org/10.1371/journal.ppat.1009061 December 21, 2020 21 / 24 Mycobacterium tuberculosis cell envelope. Molecular microbiology. 1999; 31(5):1573–87. https://doi. org/10.1046/j.1365-2958.1999.01310.x PMID: 10200974. 23. Berry MP, Graham CM, McNab FW, Xu Z, Bloch SA, Oni T, et al. An interferon-inducible neutrophildriven blood transcriptional signature in human tuberculosis. Nature. 2010; 466(7309):973–7. https:// doi.org/10.1038/nature09247 PMID: 20725040; PubMed Central PMCID: PMC3492754. 24. Eruslanov EB, Lyadova IV, Kondratieva TK, Majorov KB, Scheglov IV, Orlova MO, et al. Neutrophil responses to Mycobacterium tuberculosis infection in genetically susceptible and resistant mice. Infection and immunity. 2005; 73(3):1744–53. https://doi.org/10.1128/IAI.73.3.1744-1753.2005 PMID: 15731075; PubMed Central PMCID: PMC1064912. 25. Eum SY, Kong JH, Hong MS, Lee YJ, Kim JH, Hwang SH, et al. Neutrophils are the predominant infected phagocytic cells in the airways of patients with active pulmonary TB. Chest. 2010; 137(1):122– 8. https://doi.org/10.1378/chest.09-0903 PMID: 19749004; PubMed Central PMCID: PMC2803122. 26. J NH, Das S, Tripathy SP, Hanna LE. Role of neutrophils in tuberculosis: A bird’s eye view. Innate immunity. 2019:1753425919881176. https://doi.org/10.1177/1753425919881176 PMID: 31735099. 27. Kumar P. IFNgamma-producing CD4(+) T lymphocytes: the double-edged swords in tuberculosis. Clinical and translational medicine. 2017; 6(1):21. https://doi.org/10.1186/s40169-017-0151-8 PMID: 28646367; PubMed Central PMCID: PMC5482791. 28. Lin PL, Rodgers M, Smith L, Bigbee M, Myers A, Bigbee C, et al. Quantitative comparison of active and latent tuberculosis in the cynomolgus macaque model. Infection and immunity. 2009; 77(10):4631–42. https://doi.org/10.1128/IAI.00592-09 PMID: 19620341; PubMed Central PMCID: PMC2747916. 29. Dijkman K, Sombroek CC, Vervenne RAW, Hofman SO, Boot C, Remarque EJ, et al. Prevention of tuberculosis infection and disease by local BCG in repeatedly exposed rhesus macaques. Nature medicine. 2019; 25(2):255–62. https://doi.org/10.1038/s41591-018-0319-9 PMID: 30664782. 30. Aguilo N, Alvarez-Arguedas S, Uranga S, Marinova D, Monzon M, Badiola J, et al. Pulmonary but Not Subcutaneous Delivery of BCG Vaccine Confers Protection to Tuberculosis-Susceptible Mice by an Interleukin 17-Dependent Mechanism. The Journal of infectious diseases. 2016; 213(5):831–9. https:// doi.org/10.1093/infdis/jiv503 PMID: 26494773. 31. Gordon SV, Brosch R, Billault A, Garnier T, Eiglmeier K, Cole ST. Identification of variable regions in the genomes of tubercle bacilli using bacterial artificial chromosome arrays. Molecular microbiology. 1999; 32(3):643–55. https://doi.org/10.1046/j.1365-2958.1999.01383.x PMID: 10320585. 32. Gonzalo-Asensio J, Malaga W, Pawlik A, Astarie-Dequeker C, Passemar C, Moreau F, et al. Evolutionary history of tuberculosis shaped by conserved mutations in the PhoPR virulence regulator. Proceedings of the National Academy of Sciences of the United States of America. 2014; 111(31):11491–6. https://doi.org/10.1073/pnas.1406693111 PMID: 25049399; PubMed Central PMCID: PMC4128152. 33. Hunt DM, Sweeney NP, Mori L, Whalan RH, Comas I, Norman L, et al. Long-range transcriptional control of an operon necessary for virulence-critical ESX-1 secretion in Mycobacterium tuberculosis. Journal of bacteriology. 2012; 194(9):2307–20. https://doi.org/10.1128/JB.00142-12 PMID: 22389481; PubMed Central PMCID: PMC3347062. 34. Gomez-Velasco A, Bach H, Rana AK, Cox LR, Bhatt A, Besra GS, et al. Disruption of the serine/threonine protein kinase H affects phthiocerol dimycocerosates synthesis in Mycobacterium tuberculosis. Microbiology. 2013; 159(Pt 4):726–36. https://doi.org/10.1099/mic.0.062067-0 PMID: 23412844; PubMed Central PMCID: PMC3709824. 35. Pinto R, Tang QX, Britton WJ, Leyh TS, Triccas JA. The Mycobacterium tuberculosis cysD and cysNC genes form a stress-induced operon that encodes a tri-functional sulfate-activating complex. Microbiology. 2004; 150(Pt 6):1681–6. https://doi.org/10.1099/mic.0.26894-0 PMID: 15184554. 36. Fenn K, Wong CT, Darbari VC. Mycobacterium tuberculosis Uses Mce Proteins to Interfere With Host Cell Signaling. Frontiers in molecular biosciences. 2019; 6:149. https://doi.org/10.3389/fmolb.2019. 00149 PMID: 31998747; PubMed Central PMCID: PMC6961568. 37. Gioffre A, Infante E, Aguilar D, Santangelo MP, Klepp L, Amadio A, et al. Mutation in mce operons attenuates Mycobacterium tuberculosis virulence. Microbes and infection. 2005; 7(3):325–34. https://doi.org/ 10.1016/j.micinf.2004.11.007 PMID: 15804490. 38. Malone KM, Rue-Albrecht K, Magee DA, Conlon K, Schubert OT, Nalpas NC, et al. Comparative ’omics analyses differentiate Mycobacterium tuberculosis and Mycobacterium bovis and reveal distinct macrophage responses to infection with the human and bovine tubercle bacilli. Microbial genomics. 2018; 4 (3). https://doi.org/10.1099/mgen.0.000163 PMID: 29557774; PubMed Central PMCID: PMC5885015. 39. Said-Salim B, Mostowy S, Kristof AS, Behr MA. Mutations in Mycobacterium tuberculosis Rv0444c, the gene encoding anti-SigK, explain high level expression of MPB70 and MPB83 in Mycobacterium bovis. Molecular microbiology. 2006; 62(5):1251–63. https://doi.org/10.1111/j.1365-2958.2006.05455.x PMID: 17064366. PLOS PATHOGENS PknH locus differences in the context of Mycobacterium tuberculosis Complex evolution PLOS Pathogens | https://doi.org/10.1371/journal.ppat.1009061 December 21, 2020 22 / 24 40. Garnier T, Eiglmeier K, Camus JC, Medina N, Mansoor H, Pryor M, et al. The complete genome sequence of Mycobacterium bovis. Proceedings of the National Academy of Sciences of the United States of America. 2003; 100(13):7877–82. Epub 2003/06/06. https://doi.org/10.1073/pnas. 1130426100 PMID: 12788972; PubMed Central PMCID: PMC164681. 41. Boritsch EC, Frigui W, Cascioferro A, Malaga W, Etienne G, Laval F, et al. pks5-recombination-mediated surface remodelling in Mycobacterium tuberculosis emergence. Nature microbiology. 2016; 1:15019. https://doi.org/10.1038/nmicrobiol.2015.19 PMID: 27571976. 42. Lowe DM, Redford PS, Wilkinson RJ, O’Garra A, Martineau AR. Neutrophils in tuberculosis: friend or foe? Trends in immunology. 2012; 33(1):14–25. https://doi.org/10.1016/j.it.2011.10.003 PMID: 22094048. 43. Malone KM. An integrative ’omics approach to define functional variation between the human and bovine tubercle bacilli. p.69: University College Dublin. School of Veterinary Medicine.; 2016. 44. BRIAN K. KAY MPWAMS. The importance of being proline: the interaction of proline-rich motifs in signaling proteins with their cognate domains. The FASEB Journal. 2000; 14:213–41. 45. Williamson MP. The structure and function of proline-rich regions in proteins. The Biochemical journal. 1994; 297 (Pt 2):249–60. Epub 1994/01/15. https://doi.org/10.1042/bj2970249 PMID: 8297327; PubMed Central PMCID: PMC1137821. 46. Zarrinpar A, Bhattacharyya RP, Lim WA. The structure and function of proline recognition domains. Science’s STKE: signal transduction knowledge environment. 2003; 2003(179):RE8. https://doi.org/10. 1126/stke.2003.179.re8 PMID: 12709533. 47. Groschel MI, Sayes F, Simeone R, Majlessi L, Brosch R. ESX secretion systems: mycobacterial evolution to counter host immunity. Nature reviews Microbiology. 2016; 14(11):677–91. https://doi.org/10. 1038/nrmicro.2016.131 PMID: 27665717. 48. Pang X, Samten B, Cao G, Wang X, Tvinnereim AR, Chen XL, et al. MprAB regulates the espA operon in Mycobacterium tuberculosis and modulates ESX-1 function and host cytokine response. Journal of bacteriology. 2013; 195(1):66–75. https://doi.org/10.1128/JB.01067-12 PMID: 23104803; PubMed Central PMCID: PMC3536182. 49. Munoz-Elias EJ, Upton AM, Cherian J, McKinney JD. Role of the methylcitrate cycle in Mycobacterium tuberculosis metabolism, intracellular growth, and virulence. Molecular microbiology. 2006; 60 (5):1109–22. https://doi.org/10.1111/j.1365-2958.2006.05155.x PMID: 16689789. 50. Baer CE, Iavarone AT, Alber T, Sassetti CM. Biochemical and spatial coincidence in the provisional Ser/Thr protein kinase interaction network of Mycobacterium tuberculosis. The Journal of biological chemistry. 2014; 289(30):20422–33. https://doi.org/10.1074/jbc.M114.559054 PMID: 24928517; PubMed Central PMCID: PMC4110253. 51. Voskuil MI, Schnappinger D, Visconti KC, Harrell MI, Dolganov GM, Sherman DR, et al. Inhibition of respiration by nitric oxide induces a Mycobacterium tuberculosis dormancy program. The Journal of experimental medicine. 2003; 198(5):705–13. https://doi.org/10.1084/jem.20030205 PMID: 12953092; PubMed Central PMCID: PMC2194188. 52. Chao JD, Papavinasasundaram KG, Zheng X, Chavez-Steenbock A, Wang X, Lee GQ, et al. Convergence of Ser/Thr and two-component signaling to coordinate expression of the dormancy regulon in Mycobacterium tuberculosis. The Journal of biological chemistry. 2010; 285(38):29239–46. Epub 2010/ 07/16. https://doi.org/10.1074/jbc.M110.132894 PMID: 20630871; PubMed Central PMCID: PMC2937955. 53. Russell DG. Highlighting the parallels between human and bovine tuberculosis. Journal of veterinary medical education. 2003; 30(2):140–2. https://doi.org/10.3138/jvme.30.2.140 PMID: 12970858. 54. Kanno AI, Goulart C, Rofatto HK, Oliveira SC, Leite LCC, McFadden J. New Recombinant Mycobacterium bovis BCG Expression Vectors: Improving Genetic Control over Mycobacterial Promoters. Applied and environmental microbiology. 2016; 82(8):2240–6. https://doi.org/10.1128/AEM.03677-15 PMID: 26850295; PubMed Central PMCID: PMC4959472. 55. Li H, Durbin R. Fast and accurate long-read alignment with Burrows-Wheeler transform. Bioinformatics. 2010; 26(5):589–95. https://doi.org/10.1093/bioinformatics/btp698 PMID: 20080505; PubMed Central PMCID: PMC2828108. 56. Li H, Durbin R. Fast and accurate short read alignment with Burrows-Wheeler transform. Bioinformatics. 2009; 25(14):1754–60. https://doi.org/10.1093/bioinformatics/btp324 PMID: 19451168; PubMed Central PMCID: PMC2705234. 57. Liao Y, Smyth GK, Shi W. featureCounts: an efficient general purpose program for assigning sequence reads to genomic features. Bioinformatics. 2014; 30(7):923–30. https://doi.org/10.1093/bioinformatics/ btt656 PMID: 24227677. PLOS PATHOGENS PknH locus differences in the context of Mycobacterium tuberculosis Complex evolution PLOS Pathogens | https://doi.org/10.1371/journal.ppat.1009061 December 21, 2020 23 / 24 58. Love MI, Huber W, Anders S. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome biology. 2014; 15(12):550. https://doi.org/10.1186/s13059-014-0550-8 PMID: 25516281; PubMed Central PMCID: PMC4302049. PLOS PATHOGENS PknH locus differences in the context of Mycobacterium tuberculosis Complex evolution PLOS Pathogens | https://doi.org/10.1371/journal.ppat.1009061 December 21, 2020 24 / 24