Kupyaphores are counter regulatory zinc homeostatic metallophores required for Mycobacterium tuberculosis colonization

Tuberculosis (TB) patients suffer from progressive and debilitating loss of muscle mass and function, referred to as cachexia. Though a multifactorial condition, cachexia in cancer is promoted by systemic zinc redistribution and accumulation in muscles. Clinical studies with TB patients indeed show zinc dyshomeostasis. We therefore set out to understand mechanisms by which Mycobacterium tuberculosis ( Mtb ) govern zinc metallostasis at the host-pathogen interface. Here, we report a novel zinc metallophore from Mtb that restores zinc metabolic imbalance. These diisonitrile lipopeptides, named kupyaphores are transiently induced early-on during macrophage infection and also in infected mice lungs. Kupyaphores protects bacteria from host-mediated nutritional deprivation and intoxication. Kupyaphore Mtb mutant strain cannot mobilize zinc and shows reduced fitness in mice. Further, we characterize Mtb encoded isonitrile hydratase that could mediate intracellular zinc release through covalent modification of kupyaphores. Our studies could provide a molecular link between TB-induced altered zinc homeostasis and associated cachexia.

The adaptive response strategy of bacterial pathogens to host-imposed zinc scarcity and poisoning is primarily governed by metal-sensing metalloregulatory proteins, metal transporters and pumps. As in most bacteria, Mtb contains two key zinc sensor proteins that co-ordinately function as uptake or efflux repressors. Mtb Zinc uptake repressor (Zur, Rv2359) regulate expression of several genes, including those involved in zinc uptake 18 , while the efflux regulator ZntR (Rv3334) regulates transcription of genes encoding three types of exporters 19 .
Mtb has been shown to utilize P1-type ATPases to neutralize the toxic effects of zinc in macrophages 14 . Mtb also employs Zn sparing to overcome zinc starvation, e.g. remodeling of the ribosome 70S subunit 20 . Similar mechanisms of metallostasis also govern the intracellular levels of iron during the host-pathogen interaction 21,22 . However, the essential bacterial strategy of iron acquisition involving siderophore-mediated scavenging is yet to be recognized for other transition metals. We therefore attempted to identify mechanisms that will actively facilitate Mtb to acquire zinc during intracellular survival and in-turn modulate host zinc dynamics.
In this study, we identify a zinc-specific metallophore, which is produced on demand by Mtb to maintain bacterial fitness under varying zinc environments. These diacyl-diisonitrile lipopeptides, named kupyaphores are specifically induced during infection and can be move in and out of cells scavenging zinc. Further, we identify a novel isonitrile hydratase homologue in Mtb that is expressed in low zinc conditions and probably facilitates zinc release from kupyaphores. Detection of kupyaphores from infected mouse lungs and Mtb clinical strains propose pathophysiological relevance in zinc mobilization that could mediate cachectic metabolic end state.

Transcriptional responses maintains zinc homeostasis
To examine mycobacterial adaptive responses to conditions of varying zinc levels, we performed transcriptome analysis. The Middlebrook 7H9 medium used for culturing Mtb contains 6M concentration of ZnSO4 and we assessed Mtb growth kinetics at a log-order lower-and higher-zinc concentrations in the minimal Sauton's medium. Mtb growth curves at 0.1M (low), 6M (optimal) and 50M (high) of zinc showed no significant differences [ Fig   1a]. Mtb intracellular total zinc levels, as measured by inductively coupled plasma mass spectrometry (ICP-MS), were also buffered at all three conditions [Fig 1b]. RNA was isolated from log phase cultures and transcriptome analysis was performed using Illumina MiSeq sequencing platform for three biological replicates and the quality of reads was assessed [ Fig   S1a]. Principal component analysis (PCA) with biological replicates showed good clustering, except for one from optimal conditions, which was not taken into account for further analysis.
Interestingly, greater correlation was observed between transcriptional features of low and high zinc condition data sets, than with the optimal zinc condition [Fig 1c]. We compared differential expression analysis between 50M and 0.1M with cut off log2-fold change >0.5 and p-value <0.05 and visualized the data using MA plot [Fig 1d]. This analysis identified a small set of 39 genes to be differentially regulated. Interestingly, several of these genes have been previously known to be zinc (or metal) responsive genes. For example, upregulated genes included Zn transcription repressors smtb and zur 18 and efflux zinc pump ctpC 15 , Rv2025c 23 and cadI 24 . The metal storage proteins bfrB 25 , Rv0958 and rpoB 26 were downregulated. Circos plot analysis of 30 important operonic loci for all the three conditions of low-, high-and optimal-zinc levels were performed to understand co-regulated genes from Mtb [Fig 1e]. It is interesting to note dramatic increase in the levels of ctpC operon at high-zinc and downregulation at low-zinc. This is concordant with the established role of ctpC in zinc detoxification. Genes belonging to esx-1 and esx-3 export system show highest downregulation at both high-and low-zinc, when compared to optimal levels. Previous study had suggested role of these clusters in metal-dependent regulation 27 . We also observed regulation of biosynthetic clusters that are known to produce mycobactin and pthiocerol dimycocerosate.
Careful examination of genes from the cryptic biosynthetic cluster Rv0097 to Rv0101 showed modest upregulation at both low-and high-zinc conditions. Since this cluster has been proposed to produce unknown metabolite, we assessed the induced expression of these genes by qRT-PCR analysis. This cluster indeed showed upregulation at both low and high conditions, when compared with optimum levels of zinc [Fig 1f]. We set out to understand the relevance of this cluster in zinc metallostasis.

Role of Mtb nrps in mycobacterial physiology
Rv0097-Rv0101 Mtb gene cluster spans 10.8 kbp, which over the years has been implicated in Mtb virulence, without exactly delineating their biochemical function 28,29,30 . The largest open reading frame (Rv0101) of Rv0097-Rv0101 gene cluster codes for a bimodular non-ribosomal peptide synthetase (NRPS). We generated a NRPS knockout strain in Mtb (nrps) using phage-based transduction 31 to assess the role of this gene cluster in metal homeostasis [ Fig   S2a]. The mutant strain showed no growth profile differences under planktonic conditions in the 7H9 medium [ Fig S2b]. However, nrps Mtb mutant exhibited differences under biofilm growth conditions as reported earlier 30

Computational and Biochemical analysis of Mtb nrps biosynthetic gene cluster
To predict metabolite produced by the biosynthetic operon Rv0097-Rv0101, we carried out detailed in silico analysis and also confirmed biochemical functions for three enzymes by The four other genes from this cluster encode for an oxidoreductase (Rv0097), a thioesteraselike protein (Rv0098), a fatty acyl-AMP ligase (FAAL) (Rv0099) and an acyl carrier protein (ACP) (Rv0100). Our previous studies showed FAAL10 (Rv0099) to activate long-chain fatty acids as acyl-adenylates 35 and transfer them onto the thiol group of Ppant ACP (Rv0100) protein 32 . We confirmed similar activity for unsaturated long chain fatty acids. This ACPbound lipid chain is likely to be acted upon by Rv0097 and Rv0098. We cloned and expressed Rv0097 and Rv0098 in E. coli and purified the proteins using affinity chromatography. while the second A2 domain shows specificity for phenylalanine 36,37 . Finally, the acyl chain would be reductively released to the corresponding alcohol, thus producing a novel isonitrile lipopeptide.

Identification and Characterization of Diisonitrile Lipopetides from Mtb
In order to isolate these molecules from Mtb, we extracted metabolites from Mtb cells grown

Characterization of Kupyaphores in Infected Mouse Lungs and Human Clinical Strains
Finally, we set out to understand the temporal regulation of this zinc acquisition machinery in the mouse infection model. After aerosol challenge in Balb/C mice with WT Mtb, lungs were harvested from these mice on day 1, and weeks 1, 2, 4 and 6. We performed CFU analysis with part of lung tissue to confirm infection [ Fig S8a]  Since zinc is redox-inert metal, the release of zinc from metallophores cannot follow the classical reductive release mechanism known for iron-siderophore. In the case of enterobactin, siderophore hydrolase is known to modify the scaffold mediating iron release 40 . We therefore analyzed the Mtb genome to investigate the presence of putative isonitrile modifying enzyme.
This enzyme activity (InhA) was first reported from Pseudomonas putida 41 Fig 6c]. This modification of isonitrile is likely to reduce the Lewis base character and thus will diminish formation of stable coordinate complex bond with zinc. Interestingly, Rv0052 expression was found to be significantly upregulated by qRT-PCR only in low zinc condition and showed no induction under high zinc condition [Fig 6d]. We thus propose that this could be a putative mechanism of zinc release from kupyaphores under low-zinc conditions.

DISCUSSION
Given that there is virtually no free zinc in the cell and that this micronutrient cannot be produced de-novo, it becomes mandatory for pathogens to acquire these from the host pools.
Bacteria have thus evolved rather intricate molecular mechanisms of metal ion sensing, uptake, efflux and allocation to maintain homeostasis 43 . We here decipher a novel mechanism of zinc acquisition whereby Mtb secretes zinc-specific metallophores on demand to support its intracellular survival. While siderophore-mediated iron acquisition is well established in several bacterial and fungal systems 44,45 , metabolites that selectively chelate and transport zinc have not been characterized thus far. As pathogen proliferate and establish infection, the demand of zinc continue to accrue and micronutrients like zinc becomes a shared resource for both host and pathogen. This competitive conflict culminates into an altered host zinc homeostasis 46, 47, 48 . In addition, the inflammatory response of host during Mtb pathogenesis also mediate redistribution of metabolic nutrients. Together, these events could play a pivotal role in the cachexia development, as recognized in cancer.
Our studies show that diisonitrile lipopeptides, named kupyaphores produced by Mtb are critical to maintain zinc metallostasis. The chemical features of kupyaphores are typical of Mtb metabolic repository. Long-fatty acyl chains facilitates these molecules to transverse complex Mtb cell envelope. The electron rich isonitrile functionality forms coordinate bond and the affinity for chelation to zinc metal ions is probably governed by the structural architecture. In the low zinc conditions, zinc-kupyaphore complex is shuttled into mycobacterial cells by using transporters that are yet to be characterized. The intracellular release of zinc ions may be mediated by Mtb-coded isonitrile hydratase (Rv0052) that modifies isonitriles to corresponding formamides. It is possible that kupyaphores are then recycled back into the system by regenerating the active isonitrile groups. Recently, opine-type metallophores staphylopine from Staphylococcus aureus 49 and pseudopaline from pseudomonas aeruginosa 50 that chelate broad range of metal ions including zinc have been reported. This family of metallophores have been referred to as zincophore 51 despite the fact that these metabolites possess a wide range of metal ion specificity.
Kupyaphores show strict specificity for zinc and kupya biosynthetic cluster is transcriptionally activated during both the low-and high-levels of zinc. We therefore propose that kupyaphores may be potentially playing a role in both zinc acquisition as well as quenching of toxic zinc levels [Fig 6e]. Secretion of kupyaphores under limiting zinc conditions would allow Mtb to scavenge zinc from environment with concomitant intracellular release through isonitrile hydratase. While detailed mechanism needs to be worked out, based on RNAseq data we propose that classical zinc suppressor Zur may be involved in this activation during the low zinc conditions. The expression of zur mRNA are substantially higher when Mtb is grown at high zinc levels as compared to low levels. Indeed, classical Zur-binding sites can be mapped (61bp upstream of Rv0096 and 354bp upstream of Rv0099) in the promoter region of kupya biosynthetic cluster. As previously recognized, binding of zinc to Zur stabilizes the formation of dimers that binds strongly to these binding sites, thus suppressing the genes in their  Mtb biofilms were grown in Sauton's medium (without tween 80 or tyloxapol) by incubation without shaking at 37°C for 5 weeks under humidified conditions. The dishes were wrapped with parafilm during incubation.

RNA-seq library preparation
Total DNA-free RNA sample was depleted of bacteria rRNA with Ambion's MICROBExpress kit (AM1905) per the manufacturer's instructions. Total bacterial RNA ~1ug was processed using Truseq Stranded Total RNA seq protocol (Illumina Inc). Depletion of Bacterial rRNA was done using biotinylated, target specific oligos combined with Ribo-Zero Gold rRNA removal beads. The rRNA depleted total RNA was purified, fragmented and primed for cDNA 1st-and 2nd-strand synthesis followed by A-tailing and ligation of adapters with multiplexing indexes. The products were amplified with 15 PCR cycles and purified using Agencourt AMPure XP magnetic beads (Beckman Coulter, Brea, CA, USA) according to the manufacturer's instructions. The quality of cDNA libraries was checked with Agilent DNA1000 chips (2100 Bioanalyzer; Agilent Technologies, Santa Clara, CA, USA). 300x2bp paired end sequencing was performed using V3 flow cell on Illumina MiSeq. A total of 23 Million reads were generated for the three biological samples(each in triplicates)

RNA-seq data analysis
Mycobacterium tuberculosis genome and its annotations (gff/fna/ptt/rnt) were downloaded using an R package, Progenome (v0.0.7) 54 and the gist used for extraction is in the following reference 55 . Further, raw reads were quality trimmed using Trim-Galore (v0.6.4) 56 and the resulting trimmed reads were aligned using Salmon (v0.14.1) 57 and estimated differential gene expression using DESeq2 (v1.26.0) 58 after filtering genes with count of 5 or more for at least 4 or more samples. For predicting the operons from the RNA-Seq data we used Rockhopper (v2.03) 59

Acid catalyzed hydrolysis for isonitrile functional group confirmation
Ethyl acetate metabolite extracts of WT Mtb biofilm culture were treated with formic acid (5% v/v) (Sigma Aldrich, Cat# 64186) for 4 hours at room temperature. Extracted ion analysis was performed for the parent lipopeptide peak (m/z for [M+H] + ion 848.6987) and synthesized -amine lipopetide (m/z for [M+H] + ion 828.7300) using the LC-MS protocol described in the earlier section.

Gene expression analysis by RT-PCR
Using NucleoSpin RNA kit (Macherey-Nagel, Cat# 740955), total RNA from desired cultures of the indicated strains of Mtb were isolated from 10 mL cultures obtained from mid-exponential growth phase in 7H9-OADC, Sauton's medium or biofilm. Using the PrimeScript First Strand cDNA Synthesis Kit (Takara Bio, Cat# 6110A), cDNA was generated from 1g of total RNA from each specified sample.
RT-qPCR reactions were prepared using 1 μL of cDNA reaction mixture for each gene-specific primer per reaction with SYBR Green Master Mix (Applied Biosystems, Cat# A25742) as per the manufacturer's instruction in a Roche LightCycler 480 instrument II. Template normalization was performed by dividing the absolute gene expression of specific genes by the absolute gene expression of 16S rRNA. The sequences of primers used for qRT-PCR are provided in SI Table 1. Reactions without the cDNA were used as no-template negative control.

Cell culture and infection
The murine macrophage cell line RAW264.7 or primary Bone-marrow derived macrophages were micro-centrifuge tubes. Volume were made up to 1ml using MS grade water and samples were filtered before running on ICP-MS (Thermo Xcaliber II). Absolute ppb counts were normalized to protein concentration estimated for each sample and plotted.

Mtb metabolite isolation upon murine TB infection
For Mtb metabolite extraction from mice, 0.1g of lung tissue were taken from left apical lobe of uninfected and infected mice at indicated time points. 5 times the volume of ethyl acetate was added.
The sample were then homogenized using 0.1zirconium bead in a bead beater followed by centrifugation at 13000g for 5 minutes. The organic layer was collected, transferred to fresh tube, dried and subjected to targeted LC-HRMS studies as described previously. All the species analyzed were quantified using the multiple reaction monitoring high resolution (MRM-HR) LC-MS method on a Sciex X500R QTOF mass spectrometer fitted with an ExionLC UHPLC system. All data was collected and analyzed using the SciexOS software. All metabolite estimations were performed using an ESI source, with the following MS parameters: curtain gas = 20 L/min, ion spray voltage = 5500 V, temperature = 500 °C. Details of mice infection studies with WT Mtb are provided in Supplementary Information.

Statistical Analysis
GraphPad Prism 8 software was used for statistical analysis. Statistical significance was analyzed by Student's t-test or one-way or two-way ANOVA with p > 0.05 (not significant), * p < 0.05, and * * p < 0.01 ***p<0.005 when applicable. Data were plotted as the mean, with error bars representing SEM of three biological replicates.

Ethics Declaration
All mouse studies described in this paper received formal approval from the National Institute of Immunology -Institutional Animal Ethics Committee (NII-IAEC 440/17) following the guidelines outlined by the Committee for the Purpose of Control and Supervision of Experiments on Animals (CPCSEA), Government of India. Studies on clinical strains isolated from TB subjects described in this paper has approval from Christian Medical College, Vellore and CSIR-Institute of Genomics and Integrative Biology.
Detailed codes used for the analysis are submitted in github repository (https://github.com/viv3kanand/MTU-Manuscript/). The datasets generated during and/or analysed during the current study can be accessed using the NCBI BioProject ID PRJNA701877.