Figures
Abstract
Background
Bacterial infections remain a major threat and a leading cause of death worldwide. Most of the bacterial infections are caused by gram-positive and negative bacteria, which are recognized by Toll-like receptor (TLR) 2 and 4, respectively. Activation of these TLRs initiates multiple pathways that subsequently lead to effective immune response. Although, both the TLRs share common signaling mechanism yet they may exhibit specificity as well, resulting in the release of diverse range of inflammatory mediators which could be used as candidate biomolecules for bacterial infections.
Results
We adopted systems biological approach to identify signaling pathways mediated by TLRs to determine candidate molecules associated with bacterial infections. We used bioinformatics concepts, including literature mining to construct protein-protein interaction network, prioritization of TLRs specific nodes using microarray data and pathway analysis. Our constructed PPI network for TLR 2 (nodes: 4091 and edges: 66068) and TLR 4 (node: 4076 and edges: 67898) showed 3207 common nodes, indicating that both the TLRs might share similar signaling events that are attributed to cell migration, MAPK pathway and several inflammatory cascades. Our results propose the potential collaboration between the shared signaling pathways of both the receptors may enhance the immune response against invading pathogens. Further, to identify candidate molecules, the TLRs specific nodes were prioritized using microarray differential expressed genes. Of the top prioritized TLR 2 molecules, 70% were co-expressed. A similar trend was also observed within TLR 4 nodes. Further, most of these molecules were preferentially found in blood plasma for feasible diagnosis.
Conclusions
The analysis reveals the common and unique mechanism regulated by both the TLRs that provide a broad perspective of signaling events in bacterial infections. Further, the identified candidate biomolecules could potentially aid future research efforts concerning the possibility in differential diagnosis of gram-positive and negative bacterial infections.
Citation: R M, V A, Thangam B, Ahmed SSSJ (2014) A Systems Biological Approach Reveals Multiple Crosstalk Mechanism between Gram-Positive and Negative Bacterial Infections: An Insight into Core Mechanism and Unique Molecular Signatures. PLoS ONE 9(2): e89993. https://doi.org/10.1371/journal.pone.0089993
Editor: Markus M. Heimesaat, Charité, Campus Benjamin Franklin, Germany
Received: December 4, 2013; Accepted: January 24, 2014; Published: February 28, 2014
Copyright: © 2014 R, 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.
Funding: The authors have no support or funding to report.
Competing interests: The authors have declared that no competing interests exist.
Introduction
Innate immune system is the first line of defense against invading pathogens that are highly conserved across species [1]. The principal characteristic of innate immunity is to recognize and to eliminate the invading pathogens. Recognition of pathogens is mediated by germ-line encoded pattern recognition receptors (PRRs) present in the outer membrane of innate immune cells [2]. These PRRs bind to evolutionarily conserved pathogen-associated molecular patterns (PAMPs) that play a key role in host defense mechanism. Among several PRRs, the toll-like receptors (TLRs) are the best characterized, in terms of the PAMPs recognition and activation of its corresponding pathways.
Several PAMPs have been noticed in gram-positive (gram +ve) and gram-negative (gram -ve) bacteria such as flagellin, lipoarabinomannan, lipoteichoic acid, lipopolysaccharide and peptidoglycan. In general, wide range of gram-positive bacteria containing peptidoglycan, lipoarabinomannan and lipoteichoic acid are predominantly recognized by TLR 2, and gram-negative bacteria with lipopolysaccharide by TLR 4 [3]. Upon PAMPs recognition, TLR 2 and 4 initiate multiple pathways that subsequently lead to complex inflammatory responses. Although, both the TLRs share common signaling mechanism yet they may exhibit specificity as well, resulting in the release of diverse range of inflammatory mediators like cytokines, cell adhesion molecules and growth factors [4], [5]. Importantly, cytokines play a pivotal role in the inflammatory cascade in response to TLRs activation.
Several studies have reported the altered levels of cytokines such as TNF α, IL-6, IL-8 and PCT in bacterial infection, which are suggested as diagnostic markers [6], [7]. However, the altered levels of these cytokines are not specific but also attributed to other disease conditions, irrespective of bacterial infection. In addition, the current gold standard culture positive method is time consuming, meantime the inflammatory cascade progresses to cause multi-organ failure leading to morbidity and mortality [8]. Hence, the identification of biomarkers specific to bacterial infection is needed for early and rapid diagnosis. Understanding the molecular mechanism between host recognition and immune response at the system level is a prerequisite for biomarker discovery [9]–[11].
In this study, we propose an integrative systems biological approach to uncover common and unique mechanisms between the receptors TLR 2 and TLR 4 that are activated with bacterial ligands. In particular, the potential application of this study is to identify early diagnostic markers for bacterial infections and to establish the bioavailability of these markers in body fluids for rapid detection.
Results and Discussion
In this study, a systems biological framework (Fig. 1) was developed to illustrate toll-like receptor 2 and 4 mediated signaling mechanism in gram-positive and negative bacterial infections. We used Medline ranker to retrieve articles relevance to the queries (query 1:“TLR 2 or “Toll-like receptor 2”); (query 2: “TLR 4 or “Toll-like receptor 4”) to obtain 968 and 988 efficient articles associated to search terms, respectively. From the retrieved results, the abstracts with p-value≤0.05 were selected and subjected to PESCADOR to extract 191 and 186 potential proteins for TLR 2 and 4. Further, the proteins were manually curated to remove duplicates and to create a potential seed list to generate PPI network for TLR 2 (Fig. 2) and TLR 4 (Fig. 3).
The framework consists of four major compartments. The first compartment provides the seed genes/proteins from data mining and output as a network. The second compartment involves gene expression and prioritization of TLRs network proteins. The third compartment takes the top prioritized proteins as input to generate co-expression map, ontology and pathways. The final compartment of bioavailability network provides the proteins expressed in various body fluids for feasible detection.
Literature mining tools extract the seed genes/proteins from research articles, and cytoscape generated the protein interaction network. The nodes (blue) represent seed proteins and orange-colored nodes correspond to seed interacting proteins. All the edges (red) represent interactions between the nodes.
Literature mining tools extract the seed genes/proteins from research articles and cytoscape generated the protein interaction network. The nodes (blue) represent seed proteins and orange colored nodes represents seed interacting proteins. All the edges (red) represent interactions between the nodes.
Protein-Protein interaction network for the TLRs
The constructed TLR networks reveal the complex signaling interactions in TLR 2 and 4 in response to pathogen invasion. In TLR 2 network, there were 4091 nodes with 66068 edges, whilst there were 4076 nodes connected by 67898 edges in TLR 4 network. Further, the topological properties of the networks were analyzed for comparison, which include fundamental characteristics such as centrality, connectivity, heterogeneity and shortest path (Table 1) [12]–[14]. Topological analysis of both the networks showed that TLR 4 possesses more connecting components compared to TLR 2. For instance, in Table 1 the TLR 4 network had measurements of characteristic path length 2.363, network diameter 6 and a larger network centralization value 0.749, whilst the TLR 2 network was 2.388, 6 and 0.737, respectively. In particular, the network heterogeneity of TLR 2 was 2.080, while the heterogeneity of TLR 4 was measured to be 2.066. Overall, the variation in topological parameters describes the complexity and the existence of possible differential mechanism in TLR 2 and 4. Also, several common nodes were noticed between the networks that inclined to be crowded with higher interactions, suggesting the presence of collective regulatory mechanism in both TLRs [15].
Network association with bacterial infection
To further, investigate the relevance of the networks in bacterial infections, the microarray data set was obtained from GEO, accession number GSE6535 [16]. Gene Spring GX 7.3 microarray software was used to identify the differentially expressed genes in patients infected with bacteria (1) gram-positive vs. control, (2) gram-negative vs. control with the fold change≥2 and p-value≤0.05 as cut-off criteria [17], [18]. A total of 968 genes were significantly differentially expressed in gram +ve and 1024 in gram -ve bacterial infection (Table S1). Some of these differentially expressed genes encoding proteins were reported to influence the signaling mechanism. For instance, the differentially expressed gene, SF3B4 in TLR 2 was observed to interact with 49 neighbors in the network which was associated with Myd88 signaling pathway [19]. Also, PTPN1B in TLR 4 network was reported to have an essential role in regulation of MAPKs, NF-kappaB and IRF3 [20] which are key regulatory components against invading pathogens. These result evidence the potential significant relationship between network connectivity and bacterial infections.
Protein prioritization with differential gene expression
In order to identify potential proteins in TLRs network, we used a strategy of protein prioritization based on functional characteristic features [21], [22]. The Toppgene tool [23] was used to prioritize candidate proteins by adopting the template of known gene-diseases association data as training set to rank the proteins in the test set. In this context, the differentially expressed genes in gram +ve infected patients were used as training set and the TLR 2 specific proteins were considered as test set. The 50 top ranked proteins (Table 2) were identified and subsequently taken for further analysis. Similar procedure was followed to determine the top 50 proteins (Table 2) in TLR 4 with their respective training set (gram -ve infected patients) and test set. Overall, these highly ranked candidate proteins play a vital role in bacterial pathogenesis that may provide insights into the pathways associated to TLR 2 and 4.
Co-expression and crosstalk mechanisms
Pathway analysis is crucial in understanding the molecular events involved in pathological conditions [24]. Most of these processes are the results of multiple genes working co-operatively in pathways. Several studies suggest that the genes co-expression is essential for pathway regulation [25]–[27]. Hence, prior to pathway analysis, we demonstrated the gene\protein co-expression pattern for the prioritized proteins using Genefriends [28] tool. Among the analyzed protein (Table 2), 33 of TLR2 and 32 of TLR 4 were co-expressed (Fig. 4 and 5). These co-expressed proteins were further considered for gene ontology (Fig. 6) and pathway analysis to reveal the molecular and signaling mechanism in bacterial infection. The nature curated PID database was executed to identify the pathway mechanisms for TLR 2, TLR 4 and TLR2U4 (common proteins between TLR 2 and TLR 4) which showed an existence of key signaling events behind the receptors (Fig. 7 and 8). In addition, TLR2U4 resulted pathways suggest the occurrence of a common downstream process in both the receptors (Table S2) [29], [30]. For instance, the regulations of p38-alpha and p38-beta pathways were found to be linked with both the TLRs [31]. In addition, these pathways were associated with mitogen activated protein kinase (MAPK) pathway [32] that attributes to molecular cascades such as apoptosis, inflammation, cell migration and differentiation. MAPK cascades are signaling modules that transduce extra stimuli into intra cellular response through NFκB in the host [32]. Activation of MAPKs is an initial signaling event after PAMPs recognition that mediates NFκB, which in turn induce genes of IL-1 family of inflammatory cytokines [33]. Interestingly, IL1 mediated pathway was identified as a common pathway between both the receptors. Overall, the pathway analysis establishes that the recognition of specific PAMPs by respective TLRs will activate separate and shared pathways. It is suggested that the potential collaboration between the shared signaling pathways of both the receptors may enhance the transcriptional and cellular response against invading pathogenic microbe. Activation of the responses exhibits several inflammatory mediators which have been extensively studied and reported to be the candidate markers for bacterial infections [34]. Though many markers have been reported, their application towards clinical diagnosis is still juvenile. In addition, most of these markers failed to differentiate bacterial infections from other immunological conditions. Hence, a system-level approach would be promising in understanding complex mechanisms of host response to determine candidate markers for diagnosis. Furthermore, the bioavailability of the candidate markers driven from such approach need to be validated in body fluids (plasma and urine) to develop rapid diagnostic procedures. In this context, the analysis of co-expressed proteins in gram +ve (TLR 2) and gram -ve (TLR 4) bacterial infections were found in plasma and urine (Fig. 9). In gram -ve infection, 26 proteins were expressed in plasma and two were found in urine. Similarly, for gram +ve infection, 26 proteins were expressed in plasma whereas no candidate markers were found in urine indicating that blood plasma could be a suitable medium for detection of gram +ve and gram -ve bacterial infections. Though, the identified individual candidate markers could help in differential diagnosis, the combination of these markers would promote the development of multi-marker system for effective detection and diagnosis of bacterial infections.
The map represents the co-expressed of prioritized top 50 genes of TLR 2 network. The plot represent the wide range of expression pattern between the genes indicated as co-expression (yellow), no co-expression (orange) and self-expression (green).
The map represents the co-expressed of prioritized top 50 genes of TLR 4 network. The plot represent the wide range of expression pattern between the genes indicated as co-expression (yellow), no co-expression (orange) and self-expression (green).
The ontological classification of proteins associated to TLR2 (A) and TLR4 (B) receptors were extracted from the gene ontology database. The inputted 33 proteins of TLR2 and 32 proteins of TLR4 showed a range of functional distributions. Of which, protein binding activity was mostly attributed by the genes that influence signal transduction in bacterial infections.
The prioritized co-expressed genes\proteins were assigned to biological pathways which showed twenty one significant pathways associated to TLR 2 network.
The prioritized co-expressed genes\proteins were assigned to biological pathways which showed thirty significant pathways associated to TLR 4 network.
Bioavailability network provides the proteins expressed in plasma and urine for feasible detection. The proteins expressed in fluids represented as triangle for plasma, colored yellow (Gram positive), green (Gram negative) and proteins expressed in urine represented as ellipse for Gram negative. No specific proteins were found in urine for gram positive infections.
Conclusions
In conclusion, a systems biological knowledge about the host signaling mechanism upon activation of TLR 2 and 4 will pave the way towards identification of new candidate biomolecules associated to gram-positive and negative bacterial infections. In addition, our result demonstrates that the existence of common and unique pathway mechanism, which could be used to formulate new therapeutic approaches to improve disease condition. Moreover, experimental and clinical studies are a prerequisite to accomplish these observations.
Materials and Methods
Text mining
To extract a more efficient set of PubMed abstracts relevant to our study, Medline Ranker [35], a text mining tool was used. Medline Ranker computes most discriminating words for the given query, and compares with the abstracts available at Medline database and ranks them based on word relevancy. In this context, we retrieved a list of top ranked abstracts with its corresponding PubMed accession number (PMID) for the keywords, “TLR 2” or “Toll-like receptor 2”, “TLR 4” or “Toll-like receptor 4” with the search criteria between the years 2003 to 2013. Further, these selected PMID were imported into the PESCADOR [36], to extract genes/proteins associated to TLR 2 and 4, respectively. These extracted proteins were curated to remove duplicates from the identified potential seed proteins of TLR 2 and 4.
Network generation and topological analysis
Seed proteins were subjected to Cytoscape software [37] to generate protein-protein interaction networks (PPI) for TLR 2 and 4, respectively. Each network was created using BisoGenet plug-in [38] with experimentally validated PPIs from six databases (BIOGRID, INTACT, MINT, DIP, BIND and HPRD) to expand the seed proteins into a network. However, these networks were limited to first interacting neighbors of seed proteins to have the direct impact on TLR 2 and TLR 4 activation and regulation. In addition, self-loops and duplicated edges were removed, and the topological features were calculated by the network analyzer to determine the network property.
Gene expression data
Microarray data was retrieved from NCBI, Gene Expression Omnibus (GEO) repository [39] using the accession number GSE 6535. The data include four sample sets (i) healthy individuals (ii) sepsis patients infected with gram +ve bacteria (iii) sepsis patients infected with gram -ve bacteria and (iv) sepsis patients infected with co-bacterial infection. All samples in the data set was log transformed, normalized (perchip: normalization to the 75 percentile shift; per gene: normalization to median across all samples) and filtered based on fold change (two fold). Further, we applied analysis of variance (ANOVA) to determine the statistical significance (p≤0.05) of differentially expressed genes using Gene Spring GX.7.3 (Agilent Technologies Inc., Santa Clara, California) software.
Gene-prioritization
Toppgene [23] was used to prioritize the key molecules that play a vital role in TLR 2 and 4 signaling mechanism. Toppgene tools prioritize the candidate genes/proteins in the test set based on molecular functional similarity against training gene list. The microarray differentially expressed genes were considered as training set and systems biology driven proteins that are specific to TLR 2 and TLR 4 were input as test sets. Further, the top fifty prioritized proteins in the test set of TLR 2 and TLR 4 were identified and subjected to co-expression analysis.
Co-expression analysis
The co-expression analysis was carried out using GeneFriends [28] tool for the identified top 50 ranked proteins. The GeneFriends uses a vote-counting method to rank co-expressed genes by creating co-expression map from the genomic data at NCBI GEO data set. In this context, the top ranked proteins for TLR 2 and TLR 4 were filtered based on co-expression and subjected to gene ontology and pathway analysis.
Pathway analysis
In order to understand the functional and signaling mechanism of the proteins, we input the co-expressed proteins into nature curated Pathway Interaction Database (PID) [40]. The pathway interaction database identifies canonical pathways associated with a given list of proteins to determine the key regulating molecules in TLR 2 and TLR 4 signaling mechanism.
Candidate proteins in body fluids
To facilitate feasible diagnosis, we evaluated the availability of co-expressed genes encoding proteins in various biological fluids [41]. Protein expressed in plasma and urine were retrieved from the human protein reference database (HPRD) [42] and mapped to the co-expressed protein for the feasible diagnosis.
Supporting Information
Table S1.
Significantly differentially expressed genes. Microarray analysis of bacterial infections patients showed a significantly differentially expressed for Gram-positive and Gram-negative bacterial infection, which considered as training set for gene prioritization.
https://doi.org/10.1371/journal.pone.0089993.s001
(PDF)
Table S2.
Common pathways between TLR 2 and TLR 4. Identified common pathways (TLR2U4) mediated by both the TLRs suggests the occurrence of a common downstream process in bacterial infections.
https://doi.org/10.1371/journal.pone.0089993.s002
(PDF)
Acknowledgments
We thank Profs. R. Murugesan, R.M Pitchappan, M. Vairamani and S. Thyagarajan for their intellectual input. Authors thank Chettinad Academy of Research and Education (CARE), and SRM University for the infrastructure support and encouragement. AV specially thanks CARE for the research fellowship.
Author Contributions
Conceived and designed the experiments: AV SSJ. Performed the experiments: MR AV BT SSJ. Analyzed the data: AV SSJ. Contributed reagents/materials/analysis tools: BT MR AV SSJ. Wrote the paper: MR AV SSJ.
References
- 1. Gardy JL, Lynn DJ, Brinkman FS, Hancock RE (2009) Enabling a systems biology approach to immunology: focus on innate immunity. Trends Immunol 30: 249–262.
- 2. Erridge C (2010) Endogenous ligands of TLR2 and TLR4: agonists or assistants? J Leukoc Biol 87: 989–999.
- 3. Takeuchi O, Hoshino K, Kawai T, Sanjo H, Takada H, et al. (1999) Differential roles of TLR2 and TLR4 in recognition of Gram-negative and Gram-positive bacterial cell wall components. Immunity 11: 443–451.
- 4. Hajishengallis G, Lambris JD (2010) Crosstalk pathways between Toll-like receptors and the complement system. Trends Immunol 31: 154–163.
- 5. Elson G, Dunn-Siegrist I, Daubeuf B, Pugin J (2007) Contribution of toll-like receptors to the innate immune response to Gram-negative and Gram-positive bacteria. Blood 109: 1574–1583.
- 6. Schrag B, Roux-Lombard P, Schneiter D, Vaucher P, Mangin P, et al. (2012) Evaluation of C-reactive protein, procalcitonin, tumor necrosis factor alpha, interleukin-6, and interleukin-8 as diagnostic parameters in sepsis-related fatalities. Int J Legal Med 126: 505–512.
- 7. Verboon-Maciolek MA, Thijsen SF, Hemels MA, Menses M, van Loon AM, et al. (2006) Inflammatory mediators for the diagnosis and treatment of sepsis in early infancy. Pediatr Res 59: 457–461.
- 8. Andaluz-Ojeda D, Bobillo F, Iglesias V, Almansa R, Rico L, et al. (2012) A combined score of pro- and anti-inflammatory interleukins improves mortality prediction in severe sepsis. Cytokine 57: 332–336.
- 9. Vandenbon A, Teraguchi S, Akira S, Takeda K, Standley DM (2012) Systems biology approaches to toll-like receptor signaling. WIREs Syst Biol Med 4: 497–507.
- 10. Aderem A, Adkins JN, Ansong C, Galagan J, Kaiser S, et al. (2011) A systems biology approach to infectious disease research: innovating the pathogen-host research paradigm. MBio 2(1): e00325–10.
- 11. Jegga AG, Schneider L, Ouyang X, Zhang J (2011) Systems biology of the autophagy-lysosomal pathway. Autophagy 5: 477–489.
- 12. Aittokallio T, Schwikowski B (2006) Graph-based methods for analysing networks in cell biology. Brief Bioinform 7: 243–255.
- 13. Estrada E (2010) Quantifying network heterogeneity. Phys RevE Stat Nonlin Soft Matter Phys 82(6 Pt 2): 066102.
- 14.
Xue M, Zhang S, Cai C, Yu X, Shan L, et al.. (2013) Predicting the drug safety for traditional Chinese medicine through a comparative analysis of withdrawn drugs using pharmacological network. Evid Based Complement Alternat Med 256782.
- 15. Sabroe I, Prince LR, Jones EC, Horsburgh MJ, Foster SJ, et al. (2003) Selective roles for toll-like receptor (TLR) 2 and TLR4 in the regulation of neutrophil activation and life span. J Immunol 170: 5268–5275.
- 16. Tang BM, McLean AS, Dawes IW, Huang SJ, Cowley MJ, et al. (2008) Gene-expression profiling of Gram-positive and Gram-negative sepsis in critically ill patients. Crit Care Med 36: 1125–1128.
- 17. Zhang BH, Liu J, Zhou QX, Zuo D, Wang Ys (2013) Analysis of differentially expressed genes inductal carcinoma with DNA microarray. Eur Rev Med Pharmacol Sci 17: 758–766.
- 18. Tu CT, Chen BS (2013) New measurement methods of network robustness and response ability via microarray data. PLoS One 8(1): e55230.
- 19. D Schlaepfer, K Jones, T Hunter (1998) Multiple Grb2-Mediated Integrin-Stimulated Signaling Pathways to ERK2/Mitogen-Activated Protein Kinase: Summation of Both c-Src- and Focal Adhesion Kinase-Initiated Tyrosine Phosphorylation Events. Mol Cell Biol 18: 2571–2585.
- 20. Xu H, An H, Hou J, Han C, Wang P, et al. (2008) Phosphatase PTP1B negatively regulates MyD88- and TRIF-dependent proinflammatory cytokine and type I interferon production in TLR-triggered macrophages. Mol Immunol 45: 3545–52.
- 21. Bromberg Y: Chapter 15 (2013) Disease gene prioritization. PLoS Comput Biol 9(4): e1002902.
- 22. Sookoian S, Pirola CJ (2013) Systems biology elucidates common pathogenic mechanisms between nonalcoholic and alcoholic-fatty liver disease. PLoS One 8(3): e5889.
- 23. Chen J, Bardes EE, Aronow BJ, Jegga AG (2009) ToppGene suite for gene list enrichment analysis and candidate gene prioritization. Nucleic Acids Res 37: 305–311.
- 24. Jia P, Kao CF, Kuo PH, Zhao Z (2011) A comprehensive network and pathway analysis of candidate genes in major depressive disorder. BMC Syst Biol 23: 5.
- 25. Bauer-Mehren A, Furlong LI, Sanz F (2009) Pathway databases and tools for their exploitation:benefits, current limitations and challenges. Mol Syst Biol 5: 290.
- 26. Krishnan J, Choi S (2012) Systems biological approaches reveal non-additive responses and multiple crosstalk mechanisms between TLR and GPCR signaling. Genomics Inform 10(3): 153–166.
- 27. Diaz-Beltran L, Cano C, Wall DP, Esteban FJ (2013) Systems biology as a comparative approach to understand complex gene expression in neurological diseases. Behav. Sci 3: 253–272.
- 28. Van Dam S, Cordeiro R, Craig T, van Dam J, Wood SH, et al. (2012) GeneFriends: An online co-expression analysis tool to identify novel gene targets for aging and complex diseases. BMC Genomics 13: 535.
- 29. Huang QY, Li GH, Cheung WM, Song YQ, Kung AW (2008) Prediction of osteoporosis candidate genes by computational disease-gene identification strategy. J Hum Genet 53: 644–655.
- 30. Patil S, Pincas H, Seto J, Nudelman G, Nudelman I, et al. (2010) Signaling network of dendritic cells in response to pathogens: a community-input supported knowledgebase. BMC Syst Biol 4: 137.
- 31. Chang JH, Park JY, Kim SK (2006) Dependence on p38 MAPK signalling in the up-regulation of TLR2,TLR4 and TLR9 gene expression in Trichomonas vaginalis-treated HeLa cells. Immunology 118: 164–170.
- 32. Vasselon T, Hanlon WA, Wright SD, Detmers PA (2002) Toll-like receptor 2 (TLR2) mediates activation of stress activated MAP kinase. J Leukoc Biol 71: 503–510.
- 33. Cahill CM, Rogers JT, Walker WA (2012) The role of phosphoinositide 3-kinase signaling in intestinal inflammation. J Signal Transduct 2012: 358476.
- 34. Derek CA, Tom P (2013) Severe sepsis and septic shock. N Engl J Med 369: 840–851.
- 35. Fontaine JF, Barbosa-Silva A, Schaefer M, Huska MR, Muro EM, et al. (2009) MedlineRanker: flexible ranking of biomedical literature. Nucleic Acids Res 37: 141–146.
- 36. Barbosa-Silva A, Soldatos TG, Magalhães IL, Pavlopoulos GA, Fontaine JF, et al. (2010) LAITOR–literature assistant for identification of terms co-occurrences and relationships. BMC.Bioinformatics 11: 70.
- 37. Smoot ME, Ono K, Ruscheinski J, Wang PL, Ideker T (2011) Cytoscape 2.8: new features for data integration and network visualization. Bioinformatics 27: 431–432.
- 38. Martin A, Ochagavia ME, Rabasa LC, Miranda J, Fernandez-de-Cossio J, et al. (2010) BisoGenet: a new tool for gene network building, visualization and analysis. BMC Bioinformatics 11: 91.
- 39. Edgar R, Domrachev M, Lash AE (2002) Gene expression omnibus: NCBI gene expression and hybridization array data repository. Nucleic Acids Res 30: 207–210.
- 40. Schaefer CF, Anthony K, Krupa S, Buchoff J, Day M, et al. (2009) PID: the pathway interaction database. Nucleic Acids Res 37: 674–679.
- 41. Ahmed SS, Ahameethunisa AR, Santosh W, Chakravarthy S, Kumar S (2011) Systems biological approach on neurological disorders: a novel molecular connectivity to aging and psychiatric diseases. BMC Syst Biol 5: 6.
- 42. Peri S, Navarro JD, Kristiansen TZ, Amanchy R, Surendranath V, et al. (2004) Human protein reference database as a discovery resource for proteomics. Nucleic Acids Res 32: 497–501.