Journal of Natural Science, Biology and Medicine

: 2020  |  Volume : 11  |  Issue : 2  |  Page : 194--197

Uncovering physical interactions among human and Mycobacterium tuberculosis proteins

Dhammapal Bharne, Bhagyashri Tawar, Vaibhav Vindal 
 Department of Biotechnology and Bioinformatics, School of Life Sciences, University of Hyderabad, Hyderabad, Telangana, India

Correspondence Address:
Vaibhav Vindal
Department of Biotechnology and Bioinformatics, School of Life Sciences, University of Hyderabad, Hyderabad - 500 046, Telangana


Background: Pathogens usually evade and manipulate host immune pathways through host-pathogen protein interactions. Uncovering these interactions is crucial for determining the mechanisms underlying pathogen infection and the defense system. The growing prevalence of tuberculosis (TB) infection in the world necessitated advances in TB research. With the rising information from several divisions of biosciences, computational approaches are promising to analyze and interpret the data at the system level. Methods: In the present study, in silico two-hybrid systems is employed on model organisms to predict physical interactions among proteins of Human and Mycobacterium tuberculosis (Mtb). Consistent protein interactions are identified by the Interlog method. Co-expression analysis and functional annotations are performed to infer significant Human and Mtb protein physical interactions (HMIs). Results: The interactions identified in this study support the current TB research through an improved understanding of the pathogen infection and survival mechanism. A network of HMIs highlighted dnaK as the most highly interacting protein. Further, dnaK, eno, tuf, and gap proteins are found to trigger toll-like receptor signaling pathways and initiate pathogenesis. Conclusion: The interactions proteins identified in this study may incline the researchers to explore for novel therapeutic intervention strategies.

How to cite this article:
Bharne D, Tawar B, Vindal V. Uncovering physical interactions among human and Mycobacterium tuberculosis proteins.J Nat Sc Biol Med 2020;11:194-197

How to cite this URL:
Bharne D, Tawar B, Vindal V. Uncovering physical interactions among human and Mycobacterium tuberculosis proteins. J Nat Sc Biol Med [serial online] 2020 [cited 2021 Jun 13 ];11:194-197
Available from:

Full Text


Recent technological advances have paved the way for substantial progress in tuberculosis (TB) research. However, achievements do not have any noticeable impact on the current global trends of TB disease.[1] TB is a chronic disease that could develop over many years. During this period, the causative agent, Mycobacterium tuberculosis (Mtb) could not be isolated.[2] Further, a long generation time of the pathogen makes it difficult to grow in vitro. All these factors prolong the invention of new therapeutic interventions and slow down the advances in TB research. Nevertheless, the improved interest in research and funding from various public sectors and organizations is giving optimism. Recently, the Stop TB Partnership led by the WHO has defined a global plan to eradicate TB by 2050. This is possible to be accomplished only with the enlarged interdisciplinary research and development. Computational biology is one of the interdisciplinary research fields that can elucidate some of the critical aspects of TB through various studies such as host-pathogen protein interactions and drug discovery and development. The study of host-pathogen protein interactions is one of the fundamental approaches in understanding the virulence and infection mechanism of the pathogen. The past attempts for the prediction of Human and Mtb protein interactions have generated only a few results.[3] Therefore, the investigation of novel physical interactions of Mtb proteins with the proteins of the Human host is crucial for a systematic understanding of the pathogen. In the present study, in silico two hybrids (I2H) system[4] is employed to predict Human and Mtb protein physical interactions (HMIs) through the correlated mutations. The consistent HMIs are identified by mapping the interacting proteins with the protein interactions from Database of Interacting Proteins (DIP),[5] STRING[6] and HitPredict[7] databases through the Interlog method. These HMIs are enriched with functional annotations and expression data to derive significant interactions that are essential for infection and progression of the disease. Further, an interaction network is built from the significant HMIs to expose critical proteins for novel drug targets.

 Materials and Methods

Identification of HMIs using in silico two hybrid system

Complete proteome sets of Mtb and Homo sapiens (Human) along with seven evolutionary diverse model organisms, viz., Escherichia coli, Dictyostelium discoideum, Saccharomyces cerevisiae, Caenorhabditis elegans, Drosophila melanogaster, Danio rerio, and Mus musculus, were retrieved from the NCBI genome database. Reciprocal best BLAST hit[8] was performed on the model organisms to obtain orthologs of Mtb proteins and the orthologs of Human proteins. Multiple sequence alignments (MSAs)[9] from orthologs of an Mtb protein was reduced to at least five coincident species in orthologs of a Human protein. I2H was performed on such MSAs of Mtb and Human proteins to predict HMIs. To extract consistent HMIs, the predicted HMIs are filtered with the Mtb and Human interlogs[10] derived using complete protein-protein interaction (PPIs) from DIP and HitPredict databases and binding, experimental and co-expressing PPIs from STRING database. These HMIs were processed through functional annotations[11] using common ontology terms for Mtb and Human proteins.[12] Further, co-expressing Mtb proteins and Human proteins were integrated[13],[14] to infer significant HMIs. The significant HMIs were used to construct a protein interactions network. A core HMIs network was generated using Cytoscape software.[15] Further, the topological properties of the core interaction network were analyzed. Gene Ids of Human proteins in the core interactions network were mapped in Kyoto Encyclopedia of Genes and Genomes (KEGG)[16] pathways to identify critical proteins involved in TB pathways.


Predicted Human - Mycobacterium tuberculosis protein interactions

Using I2H, as discussed in the methods section, 20,891 HMIs were predicted. 73 Mtb and Human interlogs were identified from DIP database PPIs data, 140 from HitPredict database and 7,711 Mtb and Human interlogs from String database PPIs data. Filtering the predicted HMIs with Mtb and Human interlogs, as represented in [Figure 1], generated 485 consistent HMIs. It was observed from ontology terms of Mtb proteins and Human proteins in the consistent HMIs that there were four common biological processes, three common molecular functions, and two common cellular components. The extraction of these proteins from consistent HMIs produced 283 HMIs as significant interactions. In the consistent HMIs, there were 290 Human proteins with at least one common interacting Mtb protein partner and 22 Mtb proteins with at least one common interacting Human protein partner. Co-expression of these Human proteins in multiple microarray experiments substantiated 230 consistent HMIs as significant interactions, while the co-expression of Mtb proteins substantiated 177 consistent HMIs as significant interactions. Overall, the total number of unique and significant HMIs derived through the analysis was 419.{Figure 1}

Human - Mycobacterium tuberculosis protein interactions network

A protein interaction network generated for the significant HMIs is shown in the following [Figure 2]. It was observed from the interaction network that there were 251 Human proteins and 20 Mtb proteins. 11 of the Mtb proteins were previously revealed to play crucial roles during infection;[3],[17] the other nine Mtb proteins, viz., accD1, dnaJ2, echA8, gmdA, Mdh, Nuo, pckA, Rv2970A, and sahH are the new and significant proteins from the present study. The top highest degree nodes of Mtb were observed to be dnaK with a value of 147 followed by rpoC with 59 and atpD with a value of 47. The number of interactions for these nodes was measured to be >50% of the total number of interactions in the HMIs network. Therefore, the removal of such nodes causes fatality.[18] The top highest degree nodes of Human were observed to be DNJA4 with a value of seven, followed by ENO, TUFM, and MRPS12 with a value of six and ATP5A1 and HSPA9 with a value of five. These nodes were measured to contribute more than 8% of the total number of interactions in the HMIs network. Therefore, the study of these nodes would help in understanding the regulatory mechanisms involved in TB disease.{Figure 2}

Hypothetical protein

In the interactions network, Rv2970A, a conserved hypothetical protein of Mtb, was found to interact with four Human proteins, viz., ADH1B, ADH1C, ADH1A, and ADH4. These Human proteins were different isoforms of alcohol dehydrogenase enzyme. Recently, it was demonstrated that the alcohol metabolizing enzyme is genetically associated with the risk of causing TB.[19] Therefore, Rv2970A would have properties similar to the alcohol dehydrogenase enzyme and would play crucial functions during TB infection.

Pathway analysis

Gene Ids of 251 Human proteins in the significant HMIs were extracted from the protein table of the NCBI genome and were mapped for pathways with pink and blue color-coding at the KEGG database. It indicated that the six Human proteins, namely, HSPA9 (HSP70 gene family), HSPD1 (HSP65 gene family) and ATP6VOB, ATP6VOC, ATP6VOD1 and ATP6VOD2 (V-ATPase gene family) of the HMIs network are involved in TB pathway. It was reported that HSPA9 plays a role in the proliferation and maintenance of the cells[20] while HSPD1 plays a role in the survival and inflammatory responses.[21] ATP6V0C, ATP6V0B, ATP6V0D1, and ATP6V0D2 are the V-type proton ATPase subunit isoforms[22] that involve in phagosome maturation arrest[23] which cause the inhibition of antigen presentation. Thus, these Human proteins of the HMIs network provide insights into the TB infection and successive pathways.


The TB pathway indicated that dnaK chaperone of Mtb interacts with HSPD1/HSPA9 of Human proteins, which triggers the TLR4 and TLR2 and TLR1/6 protein to initiate toll-like receptor signaling pathway. The eno (enolase), tuf (elongation factor Tu), gap (glyceraldehyde 3-phosphate dehydrogenase) and atpA (ATP synthase subunit alpha) genes of Mtb trigger toll-like receptor pathway through interaction with HSPA9 that interacts with TLR2 and TLR1/6. Both atpA and aptD (ATP synthase subunit beta) of Mtb interacts with V-ATPase proteins such as ATP6V0D1, ATP6V0D2, and ATP6V0B for phagosome maturation arrest while atpD also interact with ATP6V0C. From the DEG database,[24] all these Mtb proteins were observed to be essential gene products. It was observed from the Tuberculist[25] database that dnaK is involved in virulence, detoxification and adaptation processes, eno, gap, atpA and atpD are involved in intermediary metabolism and respiration processes while tuf is involved in information pathways. The significance of Mtb proteins in TB pathway is represented in [Table 1]. The atpD, atpA, dnaK and eno were the already known drug targets.[3] Therefore, tuf and gap would guide the researchers to investigate these as the novel drug targets.{Table 1}


Using I2H system, the present study predicted 20,891 HMIs. Integration of HMIs with the Human and Mtb interlogs identified 485 consistent HMIs. These HMIs were further processed through functional annotations and co-expression analysis to transform into 419 significant HMIs. During this process, nine novel and significant Mtb proteins involved in TB infection were identified. An interaction network of the significant HMIs revealed few critical and hypothetical Mtb proteins. Further, the present study uncovered tuf and gap as novel Mtb proteins to investigate and develop therapeutic approaches against TB.


We acknowledge Bioinformatics Infrastructure Facility, University of Hyderabad, India.

Financial support and sponsorship


Conflicts of interest

There are no conflicts of interest.


1Comas I, Gagneux S. The past and future of tuberculosis research. PLoS Pathog 2009;5:e1000600.
2Gideon HP, Flynn JL. Latent tuberculosis: What the host “sees”? Immunol Res 2011;50:202-12
3Rapanoel HA, Mazandu GK, Mulder NJ. Predicting and analyzing interactions between Mycobacterium tuberculosis and its human host. PLoS One 2013;8:e67472.
4Pazos F, Valencia A. In silico two-hybrid system for the selection of physically interacting protein pairs. Proteins Struct Funct Genet 2002;47:219-27.
5Xenarios I, Salwínski L, Duan XJ, Higney P, Kim SM, Eisenberg D. DIP, the database of interacting proteins: A research tool for studying cellular networks of protein interactions. Nucleic Acids Res 2002;30:303-5.
6von Mering C, Jensen LJ, Snel B, Hooper SD, Krupp M, Foglierini M, et al. STRING: Known and predicted protein-protein associations, integrated and transferred across organisms. Nucleic Acids Res 2005;33:D433-7.
7López Y, Nakai K, Patil A. HitPredict version 4: Comprehensive reliability scoring of physical protein-protein interactions from more than 100 species. Database (Oxford) 2015;2015.pii: bav117.
8Moreno-Hagelsieb G, Latimer K. Choosing BLAST options for better detection of orthologs as reciprocal best hits. Bioinformatics 2008;24:319-24.
9Sievers F, Wilm A, Dineen D, Gibson TJ, Karplus K, Li W, et al. Fast, scalable generation of high-quality protein multiple sequence alignments using Clustal Omega. Mol Syst Biol 2011;7:539.
10Walhout AJ, Sordella R, Lu X, Hartley JL, Temple GF, Brasch MA, et al. Protein interaction mapping in C. elegans using proteins involved in vulval development. Science 2000;287:116-22.
11Mi H, Poudel S, Muruganujan A, Casagrande JT, Thomas PD. PANTHER version 10: Expanded protein families and functions, and analysis tools. Nucleic Acids Res 2016;44:D336-42.
12Dyer MD, Murali TM, Sobral BW. Computational prediction of host-pathogen protein-protein interactions. Bioinformatics 2007;23:i159-66.
13Lee HK, Hsu AK, Sajdak J, Qin J, Pavlidis P. Coexpression analysis of human genes across many microarray data sets. Genome Res 2004;14:1085-94.
14Wang Y, Cui T, Zhang C, Yang M, Huang Y, Li W, et al. Global protein-protein interaction network in the human pathogen Mycobacterium tuberculosis H37Rv. J Proteome Res 2010;9:6665-77.
15Shannon P, Markiel A, Ozier O, Baliga NS, Wang JT, Ramage D, et al. Cytoscape: A software environment for integrated models of biomolecular interaction networks. Genome Res 2003;13:2498-504.
16Kanehisa M, Furumichi M, Tanabe M, Sato Y, Morishima K. KEGG: New perspectives on genomes, pathways, diseases and drugs. Nucleic Acids Res 2017;45:D353-61.
17Cui T, Li W, Liu L, Huang Q, He ZG. Uncovering new pathogen-host protein-protein interactions by pairwise structure similarity. PLoS One 2016;11:e0147612.
18Jeong H, Tombor B, Albert R, Oltvai ZN, Barabasi AL. Lethality and centrality in protein networks. Nature 2011;411:41-2.
19Park SK, Park CS, Lee HS, Park KS, Park BL, Cheong HS, et al. Functional polymorphism in aldehyde dehydrogenase-2 gene associated with risk of tuberculosis. BMC Med Genet 2014;15:40.
20Wadhwa R, Yaguchi T, Hasan MK, Mitsui Y, Reddel RR, Kaul SC. Hsp70 family member, mot-2/mthsp70/GRP75, binds to the cytoplasmic sequestration domain of the p53 protein. Exp Cell Res 2002;274:246-53.
21Choi B, Choi M, Park C, Lee EK, Kang DH, Lee DJ, et al. Cytosolic Hsp60 orchestrates the survival and inflammatory responses of vascular smooth muscle cells in injured aortic vessels. Cardiovasc Res 2015;106:498-508.
22Pietrement C, Sun-Wada GH, Silva ND, McKee M, Marshansky V, Brown D, et al. Distinct expression patterns of different subunit isoforms of the V-ATPase in the rat epididymis. Biol Reprod 2006;74:185-94.
23Lu N, Zhou Z. Membrane trafficking and phagosome maturation during the clearance of apoptotic cells. Int Rev Cell Mol Biol 2012;293:269-309.
24Luo H, Lin Y, Gao F, Zhang CT, Zhang R. DEG 10, an update of the database of essential genes that includes both protein-coding genes and noncoding genomic elements. Nucleic Acids Res 2014;42:D574-80.
25Lew JM, Mao C, Shukla M, Warren A, Will R, Kuznetsov D, et al. Database resources for the tuberculosis community. Tuberculosis (Edinb) 2013;93:12-7.