Published online : 30 April 2023 Article Outline
Data Release
Mycobacterial metabolic model development for drug target identification
Bridget P. Bannerman
* Bridget P. Bannerman
Lucy Cavendish College, University of Cambridge, Lady Margaret Rd, Cambridge, CB3 0BU, UK
Science Resources Foundation, 128 City Road, London, EC1V 2NX, UK
Science Resources Foundation, 128 City Road, London, EC1V 2NX, UK
Corresponding author. E-mail: bpc28@cam.ac.uk
Find this author on Google Scholar
Find this author on PubMedAlexandru Oarga
Alexandru Oarga
Department of Computer Science and Systems Engineering, University of Zaragoza, C/María de Luna n° 1, 50018, Zaragoza, Spain
Jorge Júlvez
Jorge Júlvez
Department of Computer Science and Systems Engineering, University of Zaragoza, C/María de Luna n° 1, 50018, Zaragoza, Spain
264
46
Cite this article as...
Bridget P. Bannerman, Alexandru Oarga, Jorge Júlvez, Mycobacterial metabolic model development for drug target identification, Gigabyte, 2023 https://doi.org/10.46471/gigabyte.80
Article Information
Journal title: Gigabyte
Publisher name: GigaScience Press
Publisher location: Sha Tin, New Territories, Hong Kong SAR
Received: 12 November 2022
Accepted: 24 April 2023
Published: 30 April 2023
Preprint submitted to: https://doi.org/10.1101/2023.03.31.534705
Copyright © The Author(s) 2023.
https://creativecommons.org/licenses/by/4.0/
Data description
Context
Mycobacterium leprae (NCBI:txid1769) and Mycobacterium tuberculosis (NCBI:txid1773) are two related pathogenic mycobacteria responsible for leprosy and tuberculosis in humans. Another related mycobacterium, Mycobacteroides abscessus (NCBI:txid36809), causes opportunistic infections in healthcare-related settings. Previous analyses of the metabolic models of M. tuberculosis have supported studies demonstrating the evolutionary drivers of antibiotic resistance and the identification of novel drug targets against mycobacteria [1, 2]. Here, we demonstrate newly built genome-scale metabolic models of M. leprae and My. abscessus, including curation, simulation, and model-optimisation strategies [3–5]. To ensure the development of standardised metabolic models for the global systems biology community, we have implemented the recently released community standards and used Metabolic Model Testing MEMOTE quality control software to evaluate our models [6, 7].
Methods
Genome-scale metabolic model reconstruction, curation, and simulation
Automated draft reconstructions of M. leprae and My. abscessus were downloaded from BioModels and evaluated against other organism-specific databases, such as BioCyc, (RRID:SCR_002298) and Kyoto Encyclopedia of Genes and Genomes (KEGG, RRID:SCR_012773) [8, 9]. COBRApy (RRID:SCR_012096), a Python toolbox for the construction, manipulation, and analysis of constraint-based models [3], and GNU Linear Programming Kit (GLPK: RRID:SCR_012764), a software package that solves efficiently large-scale linear programming problems [10], were used to manipulate and simulate the models.
To create new genome-scale metabolic models (GEMs) of My. abscessus and M. leprae, additional reactions, gene-to-reaction associations, and pathways (that were not in the automated model) were integrated from KEGG and BioCyc [8, 9]. The annotations of genes and metabolites were improved by comparing and transferring annotations from the related M. tuberculosis models in the iEKVIII model and BioCyc database [1, 8]. Further improvements to the models were made by comparing them with the compound formula and charge from the MetaNetX database, and by mapping the genes and reactions of the GEMs to the BiGG (RRID:SCR_005809), ChEBI (RRID:SCR_002088), KEGG (RRID:SCR_012773), and MetaCyc databases (RRID:SCR_007778) [11–13]. Figure 1 describes the full process of the mycobacterial metabolic models (iMab22 and iMlep22) of the pathogens My. abscessus and M. leprae. The revised model reconstructions of M. leprae (iMlep22) and My. abscessus (iMab22) can be instantiated without error on the COBRA software (version 0.16.0) [3].
Figure 1.
Mycobacterial metabolic model development for drug target identification: My. abscessus and M. leprae.
Biomass reactions for M. leprae and My. abscessus
We generated biomass reactions for My. abscessus and M. leprae using the methodology in the pathway tools software [14] and the corresponding BioCyc database [15]. The software tool findCPcli, which implements the computational method for identifying bottleneck reactions as drug targets, is available in GitHub [16].
Data validation and quality control
Model optimisation
The dead-end metabolite reactions that were previously present in the automated model were eliminated to enhance the models’ quality. The models were then iteratively evaluated, considering model-specific reactions for the My. abscessus and M. leprae organisms and comparisons with the BioCyc, KEGG, MetaNetX 4.2, BiGG, and ChEBI databases [8, 11–13].
To improve the quality of the models, the reactions with dead-end metabolites previously found in the automated models were removed. MEMOTE, a standardised genome-scale metabolic model testing programme, was used to undertake quality control checks during the models’ iterations and optimisation [7]. In the process, Systems Biology Ontology (SBO) annotations and gene annotations from the KEGG database and 728 new formulae from the MetaNetX database were added [13]. As a result, the MEMOTE score increased from 49% to 63% on the M. leprae model and from 48% to 66% on the My. abscessus model.
The new GEMs for M. abscessus and M. leprae are encoded in the Systems Biology Markup Language (SBML) [17] and designated as follows:
iMlep22 for the M. leprae model: i for in silico, Mlep for M. leprae, and published in 2022. iMlep22 consists of 5,625 reactions, 4,016 metabolites, and 871 genes.
iMab22 for the My. abscessus model: i for in silico, Mab for My. abscessus, and published in 2022. iMab22 consists of 8,580 reactions, 6,273 metabolites, and 1,837 genes.
Standardisation and curation have been done according to the community standards for the development of metabolic models, as described in Carey et al. and Lieven et al. [6, 7], to produce gold-standard metabolic network reconstructions of My. abscessus and M. leprae. The development of My. abscessus and M. leprae (iMab22 and iMlep22) metabolic models is illustrated in Figure 1, and the overall capability of the models has been summarised in Figure 2.
Figure 2.
Overall capability of the models: My. abscessus (iMab22) and M. leprae and (iMlep22).
Comparative analysis
My. abscessus, an opportunistic pathogen, has a large genome size of 5.1 MB [18], with more biochemical and bottleneck reactions in the iMab22 model compared with the obligate pathogens M. tuberculosis (with a genome size of 4.4 MB) and M. leprae (with an even smaller genome size of 3.3 bp) [19]. The fewer metabolic and bottleneck reactions in the iMlep22 model can be attributed to the reductive evolution that occurred in M. leprae and the subsequent loss of genes. An illustration of the distribution of unique enzymes and bottleneck reactions in each of the models (iMab22 and iMlep22) in comparison with each other and with M. tuberculosis is demonstrated in Figure 3. Alternative enzymes are not included in this analysis because they catalyse the same biochemical reactions and do not fit into the category of bottleneck reactions, which are defined as unique reactions responsible for the survival and growth of the organisms in the metabolic network [4, 5, 20].
Figure 3.
Unique/bottleneck reactions in My. abscessus and M. leprae in relation to M. tuberculosis.
Data Availability
Data supporting this work are openly available in the GigaDB repository [21]. The models can be retrieved from:
My. abscessus https://www.ebi.ac.uk/biomodels/MODEL2203300002 and https://www.patmedb.org/Bacteria/Mabscessus.
M. leprae https://www.ebi.ac.uk/biomodels/MODEL2203300001 and https://www.patmedb.org/Bacteria/Mleprae.
The findCPcli tool (RRID:SCR_023391, biotools:findcpcli) [4] used for analysing the models can be retrieved from GitHub [16].
Declarations
List of abbreviations
GEM: Genome-scale metabolic model; GLPK: GNU Linear Programming Kit; KEGG: Kyoto Encyclopedia of Genes and Genomes; MEMOTE: Metabolic Model Testing; CObraPy: COnstraints-based reconstruction and analysis for Python; SBO: Systems Biology Ontology.
Ethics approval and consent to participate
The authors declare that ethical approval was not required for this type of research.
Competing Interests
The authors declare no competing interests.
Authors’ contributions
BPB: conceptualization, data curation, software, formal analysis, investigation, methodology, and writing—original draft, review, and editing.
JJ: software, formal analysis and writing—review, and editing.
AO: data curation, software, formal analysis, and writing—original draft, review, and editing.
Funding
The authors of this manuscript are supported by the Spanish Ministry of Science, Innovation, and Universities (JJ and AO).
Acknowledgements
The authors would like to thank Ebirien Nte and Jo Chukualim for the graphic designs and scripting.
References
1KavvasE, SeifY, YurkovichJ Updated and standardized genome-scale reconstruction of Mycobacterium tuberculosis H37Rv, iEK1011, simulates flux states indicative of physiological conditions. BMC Syst. Biol., 2018; 12: 25.
2BannermanBP, VedithiS, JulvezJ Comparative pathway analysis of mycobacterial species provides insight into the identification of new drug targets. bioRxiv. 2019; doi:10.1101/535856.
3EbrahimA, LermanJA, PalssonBO COBRApy: COnstraints-based reconstruction and analysis for Python. BMC Syst. Biol., 2013; 7: 74. doi:10.1186/1752-0509-7-74.
4OargaA, BannermanBP, JúlvezJ. Growth dependent computation of chokepoints in metabolic networks. In:
AbateA, PetrovT, WolfV (eds), Computational Methods in Systems Biology. CMSB 2020. Lecture Notes in Computer Science. Cham: Springer, 2020; doi:10.1007/978-3-030-60327-4_6.
5BannermanBP, JúlvezJ, OargaA Integrated human/SARS-CoV-2 metabolic models present novel treatment strategies against COVID-19. Life Sci. Alliance, 2021; 4(10): e202000954. doi:10.26508/lsa.202000954.
6CareyMA, DrägerA, BeberME Community standards to facilitate development and address challenges in metabolic modeling. Mol. Syst. Biol., 2020; 16: e9235. doi:10.15252/msb.20199235.
7LievenC, BeberME, OlivierBG MEMOTE for standardized genome-scale metabolic model testing. Nat. Biotechnol., 2020; 38: 272–276. doi:10.1038/s41587-020-0446-y.
8CaspiR, AltmanT, BillingtonR The MetaCyc database of metabolic pathways and enzymes and the BioCyc collection of pathway/genome databases. Nucleic Acids Res., 2014; 42: D459–D471. doi:10.1093/nar/gkt1103.
9KanehisaM, FurumichiM, SatoY KEGG: Integrating viruses and cellular organisms. Nucleic Acids Res., 2021; 49: D545–D551. doi:10.1093/nar/gkaa970.
10OkiE. GLPK (GNU Linear Programming Kit). 2012; https://www.gnu.org/software/glpk/.
11HastingsJ, OwenG, DekkerA ChEBI in 2016: Improved services and an expanding collection of metabolites. Nucleic Acids Res., 2016; 44: D1214–D1219. doi:10.1093/nar/gkv1031.
12NorsigianCJ, PusarlaN, McConnJL BiGG models 2020: Multi-strain genome-scale models and expansion across the phylogenetic tree. Nucleic Acids Res., 2020; 48: D402–D406. doi:10.1093/nar/gkz1054.
13MorettiS, TranVDT, MehlF MetaNetX/MNXref: Unified namespace for metabolites and biochemical reactions in the context of metabolic models. Nucleic Acids Res., 2021; 49: D570–D574. doi:10.1093/nar/gkaa992.
14KarpPD, MidfordPE, PaleySM Pathway Tools version 23.0 update: software for pathway/genome informatics and systems biology. Brief Bioinform., 2021; 22(1): 109–126. doi:10.1093/bib/bbz104.
15KarpPD, BillingtonR, CaspiR The BioCyc collection of microbial genomes and metabolic pathways. Brief Bioinform., 2019; 20(4): 1085–1093. doi:10.1093/bib/bbx085.
16OargaA. findCPcli - find ChokePoint reactions in genome-scale metabolic models. GitHub. 2021; https://github.com/findCP/findCPcli.
17KeatingSM, WaltemathD, KönigM SBML level 3: An extensible format for the exchange and reuse of biological models. Mol. Syst. Biol., 2020; 16: e9110. doi:10.15252/msb.20199110.
18RipollF, PasekS, SchenowitzC Non mycobacterial virulence genes in the genome of the emerging pathogen Mycobacterium abscessus. PLoS One, 2009; 4(6): e5660–e5660.
19SinghP, ColeS. Mycobacterium leprae: genes, pseudogenes and genetic diversity. Future Microbiol., 2011; 6(1): 57–71.
20YehI, HanekampT, TsokaS Computational analysis of plasmodium falciparum metabolism: Organizing genomic information to facilitate drug discovery. Genome Res., 2004; 14: 917–924. doi:10.1101/gr.2050304.
21BannermanBP, OargaA, JúlvezJ. Supporting data for “Mycobacterial metabolic model development for drug target identification”. GigaScience Database, 2023; http://dx.doi.org/10.5524/102383.
Stencila view