Constraints-based analysis identifies NAD+ recycling through metabolic reprogramming in antibiotic resistant Chromobacterium violaceum

In the post genomic era, high throughput data augment stoichiometric flux balance models to compute accurate metabolic flux states, growth and energy phenotypes. Investigating altered metabolism in the context of evolved resistant genotypes potentially provide simple strategies to overcome drug resistance and induce susceptibility to existing antibiotics. A genome-scale metabolic model (GSMM) for Chromobacterium violaceum, an opportunistic human pathogen, was reconstructed using legacy data. Experimental constraints were used to represent antibiotic susceptible and resistant populations. Model predictions were validated using growth and respiration data successfully. Differential flux distribution and metabolic reprogramming were identified as a response to antibiotics, chloramphenicol and streptomycin. Streptomycin resistant populations (StrpR) redirected tricarboxylic acid (TCA) cycle flux through the glyoxylate shunt. Chloramphenicol resistant populations (ChlR) resorted to overflow metabolism producing acetate and formate. This switch to fermentative metabolism is potentially through excess reducing equivalents and increased NADH/NAD ratios. Reduced proton gradients and changed Proton Motive Force (PMF) induced by antibiotics were also predicted and verified experimentally using flow cytometry based membrane potential measurements. Pareto analysis of NADH and ATP maintenance showed the decoupling of electron transfer and ATP synthesis in StrpR. Redox homeostasis and NAD+ cycling through rewiring metabolic flux was implicated in re-sensitizing antibiotic resistant C. violaceum. These approaches can be used to probe metabolic vulnerabilities of resistant pathogens. On the verge of a post-antibiotic era, we foresee a critical need for systems level understanding of pathogens and host interaction to extend shelf life of antibiotics and strategize novel therapies.

.000 Propionic acid --0.000 * Experimental evidence exists, # Conflicting literature evidence    Gene is not present in CV but reaction essential for biomass production. Gene with known function is present in unrelated organism.
Homology in phylogenetically unrelated organism 1.2 Enzyme not present in CV but indirect biochemical evidence of existence of the reaction is present. <30%. Homology 1.3 Inferred from homology (as per UNIPROT data). Gene present in CV but very low match with other organism. 30% < 1.3 < 60%. Homology 1.4 Gene present in CV and reaction essential for biomass production. Gene with known function present in related organism.
Homology in phylogenetically related organism 1.5 Gene present in CV and reaction essential for biomass production. Gene with known function present in well-known organism -E. coli       The 64.8% GC content is split evenly between dGTP and dCTP. MWs taken from ChEBI website. The relative weights/mol are the % prevalances multipled by the molecular weights. % by weight is calculated by divding the relative weights/mol by the sum of the relative weights/mol. mmol/gDW is calculated by: (% by weight/100)*(X grams of DNA/gDW)*(1/molecular weight)*(1000 mmol/mol)

RNA Composition
Reference -ORF DNA sequences (proxy for RNA) for C. violaceum used from Haselkorn

Material A: SEED Draft model limitation
The SEED-derived in silico C. violaceum was unable to produce twenty six out of 74 biomass precursors as reported in literature (7,8) with glucose as the sole carbon source. A total of 69 reactions ( Table E in  Pathways with gaps and missing reactions included lysine, phenylalanine, tyrosine and tryptophan biosynthesis and glycine, serine and threonine metabolism pathways. C. violaceum does not require any amino acids for its growth (8) (13). Many genes were not identified, annotated or mis-annotated. In case of lysine biosynthesis a lumped reaction had to be added for the conversion of L-aspartate semi-aldehyde to meso-2,6-diaminopimelate (rDB00050_c) in order to form lysine in the reconstruction. For arginine biosynthesis reaction (rDB00003_c) was added for formation of arginine (glutamate and 2acetamido-5oxopentanoate to oxoglutarate and N-acetylornithine). Tryptophan metabolism in C.
violaceum is unique as compared to other bacteria due to its oxidative conversion to natural bis-indole compound violacein (14-16) through a five gene violacein operon (17)(18)(19). In the SEED reconstruction the reaction that converts chorismate to prephenate (Chorismate mutase or prephenate dehydratase, CV_2355, pheA) was missing ( Figure

Material B: In silico representation of metabolic genome features of Chromobacterium violaceum
In 2003, the C. violaceum ATCC 12472 genome sequencing project was executed by the Brazilian National Genome Sequencing Consortium that included 25 sequencing laboratories, 1 bioinformatics center, and 3 coordination laboratories spread across Brazil (21). Of the 4407 protein coding genes only 61.3% could be assigned a putative function whereas 21.6% were identified as conserved hypothetical proteins and 17.1% other hypothetical proteins.
On comparison with other sequenced organisms, C. violaceum has been reported to be most similar to (17.4%) Ralstonia solanacearum, a free living phytopathogen (21). Although phylogenetically similar to the serious human pathogen N. meningitidis serogroup A (9.75%), and that of the Clusters of Orthologous Groups (COG) of ribosomal structure, biogenesis and translation.
(1), similarity of COGs of inorganic ion transporters to Ralstonia predicted C.
violaceum to be free living rather than a commensal.
C. violaceum was proposed to be well adapted to glucose, nitrogen, phosphate and amino acid starvation and is resistant to toxic agents such as hydrogen peroxide, arsenic (22), UV radiation, oxidative damage due to presence of several ORFs that act in response to such stress like pho regulon, peptide utilization and heat shock related ORFs. Around 251 genes incorporated in the model had direct literature evidence. The model iDB858 was able to successfully predict the physiology of C. violaceum as per legacy data ( Table 2).

Central Carbon Metabolism
The in silico C. violaceum, iDB858, was able to synthesize all the necessary amino acids for its survival and also showed ability to synthesize cyanide (23). As previously reported (24) during aerobic growth on glucose it was able to use glycolysis, tricarboxylic acid and glyoxylate cycle to produce cellular energy required for cell survival. The model was able to utilize amino acids, lipids and acetonitrile as sole carbon sources (25). The latter was utilized by the presence of homologous nitrilase (CV_2097) that allowed utilization of nitriles compounds such as indole acetonitrile, benzonitrile, phenylacetonitrile as suggested in literature (26). All the genes required for nucleotide salvage pathway were accounted for in the model. oxidase was also reported in C. violaceum known to play a role in oxidative stress and to create electrochemical membrane gradient for energetic requirements (28). Cyanide formation in C. violaceum is a distinguishing feature among violacein producing bacteria. In general, cyanide binds with the respiratory electron chain molecule and inhibits respiration and kills the cells. Therefore, there must be an evolved respiratory system that is resistant to cyanide production as reported (29). cioA (CV_3658), a cyanide insensitive terminal cytochrome oxidase in the respiratory electron transport (30) and cioB (CV_3657) were genes present in in silico C. violaceum model homologous to the cytochrome bd (EC 1.10.3.14) oxidases and may belong to cyanide insensitive oxidases (CIO) as observed in Pseudomonas aeruginosa (31) suggesting terminal branching of the respiratory system in C. violaceum with one pathway resistant to cyanide inhibition (or azide, CO inhibition) while the other being sensitive (29). C. violaceum being a facultative anaerobe reactions involving nitrate (denitrification) or fumarate as terminal electron acceptors for growth under anaerobic conditions were present that convert glucose into acetic acid and formic acid under anaerobic condition (7,32,33).

Macromolecular Biosynthesis
The  (41). These genes are known to be related to virulence and pathogenicity in C. violaceum (37). The specific polysaccharide or O-antigen has been reported to be composed of D-glycero-D-mannoheptose and D-fucosamine (42) along with galactose, glucose, glucosamine (3). Peptidoglycan degree of cross-linking and O-acetylation appeared to be associated with the genetic background of the strains and was found to be quite similar to N. meningitidis. The percentage of O-acetylation per disaccharide was on average 14.7% compared to 33% in N. meningitidis (43). Other studies show that peptidoglycan structures are recognized by the innate immune system (39). Several studies have shown that a direct correlation exists between the extent of O-acetylation and susceptibility to lysozymecatalyzed hydrolysis of peptidoglycan to protect the bacterium from a host immune response (43). In our model genes including CV_4346 and CV_4349 were present that are known to be related to virulence and pathogenicity in the peptidoglycan biosynthesis subsystem. C.
violaceum has strong ability to adapt to stress condition due to presence of different types of transporter proteins. 25% of extracellular proteins have been reported to be involved in transport or metabolism (44,45).

Cyanide Formation
Cyanide is produced by C. violaceum (46) as a secondary metabolite that has application in pharmaceuticals industry to gold recovery from electronic scrap materials (22,23,47,48).
Cyanide formation in C. violaceum is also used as a distinguishing feature among violacein producing bacteria. Culture conditions such as pH, temperature regulate cyanide production (49). 14 C studies have showed carbon atom of cyanide being derived from glycine (50). In silico model of C. violaceum was able to produce small amounts of cyanide without any additives in the media with either only glucose or succinate as carbon source and ammonium salts as nitrogen source. Glutamate on the other hand served as both carbon and nitrogen source with best cyanide yield (23) ( Table 2). The reactions involved in utilization of cyanide to form β-cyanoalanine (49,50) and β -cyano-α-amino butyric acid were added to the model along with other reactions shown in Table E in this file. There is no inhibition of cytochrome c oxidase with the level of cyanide produced (30).

Violacein biosynthesis
The main precursor metabolite for the synthesis of violacein is tryptophan (51). Violacein biosynthetic pathway has a five gene vioABCDE operon structure (17-19, 52, 53). The operon is reported to be negatively regulated by VioS (54) and it is positively regulated by the CviI/R quorum sensing system. The first step is the oxidative dimerization of two molecules of tryptophan to indole-3-pyruvic acid (IPA) imine catalyzed by flavoenzyme Ltryptophan oxidase, VioA (CV_3274) in presence of oxygen (16). IPA imine formed is