Exploring beta-tubulin-protein interactome using computational techniques

 

B. Lokapriya Nandan, Velin Marita Sequeira, Kanika Verma, K. Ramanathan*

Industrial Biotechnology Division, School of Bio Sciences and Technology, VIT University,

Vellore- 632014, Tamil Nadu, India.

*Corresponding Author E-mail: kramanathan@vit.ac.in

 

ABSTRACT:

The control of cell division is a cardinal issue in the field of cancer therapy. Indeed, beta-tubulin is an essential protein in cell division process. Therefore it is the prime target of many cancer types such as breast cancer, ovarian cancer, lung cancer etc. Paclitaxel is one of the most widely used anti-cancer agents for the inhibition of beta tubulin function. Although treatment with paclitaxel often works well initially, many patients develop resistance after chronic exposure to an anti-cancer agent. Of note, hampering the activity of key proteins which interact with beta-tubulin will consequently hinder the rapid cell division during tumor formation. Keeping this in mind, the STRING tool was employed in the present investigation to derive a large network of genome level protein-protein interaction with proteins as nodes and interactions as edges. A computational approach to predict such interactions is recommended as it integrates information from several sources to generate a detailed interactome. Subsequently, the proteins obtained were divided into different groups based on their functional role by k-means clustering algorithm. The grouping into clusters assisted in picking out proteins that had regulatory roles and had connections with cancer pathways as well. We certainly hope that these proteins can be shortlisted as alternative targets to beta-tubulin in cancer chemotherapy.

 

KEYWORDS: Beta-tubulin; Interactome; STRING; Clustering.

 

 


INTRODUCTION:

Microtubules are a key component of the eukaryotic cytoskeleton and are vital for many cellular functions such as maintenance of cell shape, proliferation, motility, intracellular trafficking and vesicle transport1. The basic elements of microtubules are heterodimers of small globular alpha and beta tubulin subunits2. The tubulin heterodimers account for approximately 4% of the total protein content in mammalian cells. The two monomers have similar secondary and tertiary structures due to the over 40% identity in their sequences spanning above 445 amino acids3. A pronounced difference in microtubule associated proteins has been shown to perturb the normal cell cycle activity and lead towards malignancy4.

 

Therefore, a widely accepted approach of chemotherapy to kill malignant tumors is the disruption of the organization of microtubules in order to hinder mitotic spindle formation and hence progression to mitotic division5,6. Since microtubules play a central role in tumor cell motility, formation of long dynamic protrusions7 and elongation of invadopodia8, targeting microtubules is a viable tactic to inhibit cancer. In comparison to the different beta-tubulin isotypes, beta III-tubulin has particular distinctive characteristics, which probably accounts for its unique function(s)9. The beta-tubulin is by far the most extensively studied isotype in human malignancy.10 The altered expression of the III-tubulin during carcinogenesis cells was known from the 1990s11. Since then, beta-tubulin has often been considered a potent target for anti-cancer therapeutic interventions.12. Since then several clinical/translational and functional studies on tumors have contributed substantially to our understanding of beta-III-tubulin’s role in tumorigenesis, cancer progression and its resistance to chemotherapeutic drugs.13 Therefore, if we are able to identify proteins whose absence can interfere with tubulin activity and as a result, cell division process, we will have a repertoire of potential targets to choose from and further screen the ones which have minimal non-specific activity. For this, protein-protein interaction networks facilitate the arduous task of understanding complex interactions in a meaningful manner by providing a simple graphical representation. 14,15 Hence, in the present study, we have used STRING tool to analyze the functional interaction of beta-tubulin with other proteins.

 

METHODOLOGY:

A.    Construction of Protein–Protein Interaction (PPI) network

A search tool for retrieval of interacting genes/protein STRING database v10 (available at http://string-db.org/) 16 was used with beta-III tubulin gene (TUBB3) as seed for the construction of the protein interaction (PPI) network. The interactions pertaining to Homo sapiens were selected among the lot and the interaction network was made more robust using the “add more interactors” option in STRING database. The prediction methods selected for our analysis include Neighbourhood, Gene Fusion, Co-occurrence, Co-expression, Experiments, Databases and Text mining. It was then further refined to include only those interactions with a confidence score greater than 0.9 17. The disconnected nodes were removed. The result was a comprehensive network of proteins associated with beta-tubulin generated from sources such as databases of physical interactions and databases of curated biological pathway knowledge.

 

B. Protein Clustering and Annotation

The convolute network of proteins obtained was better comprehended by clustering the interactors into 9 non-overlapping clusters using STRING k-Means clustering algorithm18. The k-Means algorithm was opted because it is an unsupervised clustering algorithm based on adjacency matrix, which groups molecules based on pre-specified criteria.


 

Fig 1: Output protein network obtained when beta-tubulin gene was added as seed

 


The resulting clusters were separated manually for better visual representation and interpretation of the interaction network. The annotation information related to the protein molecule, interaction type and source of interaction were retrieved by the STRING database when we selected the particular molecule and its interaction. The STRING software has tools to derive information the proteins involved in different biological processes. The details of the protein interaction network are presented in the results.

 

RESULT:

The PPI network developed to unfold and analyse the interactions of beta-tubulin with other proteins formed a dense yet highly connected network. Figure 1 represents the network of proteins obtained. The interactions based only on text mining were verified using PubMed. The interaction network depicted the protein molecules as the nodes of the graph while the interactions as the edges. Many of the interactors were connected to one another by multiple lines suggesting interactions derived from more than one source of information. On segregating the network based on the k-Means clustering algorithm, the members within and between each cluster were observed to be highly interconnected, reflecting a high degree of functional association and therefore, indicating interplay between the myriad pathways that comprise the protein network. Figure 2 represents the proteins clustered based on k-Means algorithm.

 

The first cluster is a highly connected network of 30 molecules. The proteins in this cluster have been reported to possess a wide range of functional attributes. This network has been found to be associated with epidermal growth factor receptor (EGFR), transcriptional regulators and many ubiquitin proteins.

 

Cluster 2 included three proteins critical to the process of ubiquitination. While UBE2D1 and UBE2D2 are ubiquitin conjugating enzymes, STUB1 is an E3 uniquitin ligase enzyme. Ubiquitin-mediated proteolysis of key regulators of cell cycle such as cyclins and CDK inhibitors inhibits untimely activation of cyclin-dependent kinases (CDKs) thereby, regulating the cell division process 19. Ubiquitin ligases mediates the specific ubiquitylation of these regulators20.

 

Therefore, alterations in the ubiquitylation of cell-cycle regulators might result in uncontrolled proliferation and therefore cancer21.

 

Cluster 3 consists of 8 proteins which are primarily involved in histone modifications. Suz12 is responsible for histone methylation.


 

Fig 2: Clustering of Beta-Tubulin Network into 9 clusters

 


The other proteins in the cluster (HDAC6 22, RBBP4, RBBP7, SAP3023, MTA1, MTA224, SUZ12 and GATAD2A) are involved in the histone deacetylase complex. Histone modification is considered an epigenetic mechanism which involves coordinated regulation of cellular processes such as gene transcription, DNA replication, and DNA repair 25, 26.

 

Cluster 4 comprises of TP53 protein and proteins interacting with it during stress conditions. The p53 protein is a key transcription factor and regulates downstream genes involved in cell cycle arrest, DNA repair and programmed cell death 27. Alterations in p53 function confers genome instability, improper apoptosis and reduced cell cycle restraint28.

 

Therefore, p53 mutations are responsible for the critical characteristics of malignancy. Due to its pivotal role in the cell cycle, p53 alterations account for the most common mutation in human cancer28. 

 

During genomic stress, p53 represses target genes through several mechanisms. One such mechanism is the recruitment of chromatin remodelling complex that includes three proteins also found in the cluster- SIN3A, HDAC1 and HDAC2 complex29,30. Ribosomal protein S27a and Hypoxia-inducible factor 1α (HIF-1α) can bind independently to MDM2 and curb MDM2-mediated p53 ubiquitination to allow activation of p53 in response to stress31, 32. The heat shock protein HSP90 acts as a chaperone for p5333. CDK1NA encodes for p21 which is an inhibitor of cyclin-dependent kinases that are essential for cell cycle progression 34.Therefore, p53-mediated growth inhibition is dependent on induction of p21 and failure in transactivation of p21 may lead to uncontrolled proliferation35. SIRT1 is a deacetylase involved in stress responses, cellular metabolism and aging through deacetylating a variety of substrates including p5336,37.

 

Cluster 5 contains HSPA8 which is known to bind to p53, however the significance of this complex is yet to be elucidated38.

 

Cluster 6 possesses the SIRT2 protein. Unlike other sirtuins, SIRT2 is the only one localised at the cytoplasm and colocalizes with microtubules and specifically deacetylates lysine-40 of alpha-tubulin 39,40.

 

 Cluster 7 has the NOTCH1 receptor which binds to beta-tubulin to initiate notch signalling which controls fundamental cellular processes such as proliferation, stem cell maintenance and differentiation during embryonic and adult development41. This pathway has physiological significance with respect to cancer too42.

 

Cluster 8 has the genes coding for alpha tubulin. They are TUBA4A, TUBA1A and TUBA1B Alpha-tubulin along with beta-tubulin is one of the major constituents of microtubules.

 

In cluster 9, proteins encoded by FOX01, FOX03, AKT1 and RHOA are present which are associated with the regulation and timing of cell division. Akt is a protein kinase involved in the activation of Akt-PI3K pathway which regulates cell survival, growth, proliferation, cell migration and angiogenesis, by phosphorylating a range of intracellular proteins Aberrant hyperactivation of this pathway promotes malignancy due to its effects on cell migration43. FOXO1 and FOXO3 belong to the O subclass of the forkhead family of transcription factors which are characterized by a distinct fork head DNA-binding domain and play a role in the Akt-PI3K pathway44. RhoA is a prominent regulatory factor in other functions such as the regulation of cytoskeletal dynamics, transcription, cell cycle progression and cell transformation and it has found to be associated with gastric cancer45.


 

Table 1- Genes involved in different molecular and biological processes

S.No.

Molecular and Biological Process

Genes Involved

1.

Transcriptional regulation of cancer

FOXO1, RELA, TP53, CDKN1A, SIN3A, HDAC1, HDAC2

2.

Regulation of microtubule polymerization

HDAC6, CCNB1

3.

Pathways in cancer

FOXO1, RHOA, AKT1, XIAP, TRAF6, NFKBIA, ERBB2, EFGR, CBL, RELA, CCND1, HIF1A, HSP90AA1, TP53, CDKN1A, HDAC2, HDAC1

4.

Regulation of gene expression, epigenetics

SIRT1, SIRT2, SIN3A, HDAC1,HDAC2, HDAC6, SUZ12,MTA1, MTA2, GATAD2A, RBBP7, RBBP4, SAP30

 

Table 2 – Proteins involved in epigenetic regulation and cancer pathways in each cluster

Role

Cluster 1

Cluster 2

Cluster 3

Cluster 4

Cluster

5

Cluster 6

Cluster 7

Cluster 8

Cluster 9

Epigenetic Regulation

0

0

8

4

0

1

0

0

0

Cancer Pathway

8

0

0

6

0

0

0

0

3

Transcriptional misregulation in cancer

1

0

0

5

0

0

0

0

1

Regulation of microtubule

polymerization

1

0

1

0

0

0

0

0

0

 

 

Fig 3- Graph showing the number of critical proteins involved in epigenetic regulation and cancer pathways in each cluster

 


All the proteins present in cluster 4 are key players in epigenetic regulation and cancer pathways, especially in transcriptional mis-regulation of cancer. The results are shown in Table2 and represented in Fig3 as graph. Therefore, the proteins in this cluster can be considered as ideal targets for novel therapeutic approaches against cancer.

 

DISCUSSION:

Beta-tubulin is situated at the crossroads of a network of cancer pathways and epigenetic regulations that are essential for cell cycle and cell division. Although the mechanisms of cancer are not understood entirely, proteins involved in cell cycle are well known to be involved in cancer development. All the biological processes such as, cell cycle, metabolic pathways and signal transduction, are consequences of properly coordinated and regulated protein-protein interactions46,47. The protein-protein interactions will enable us to better understand information about the regulation of molecular and biological functions. Data inferred from protein functional interactions can be extended to genes. Of recent, novel candidate genes have been identified as drug targets using molecular network interactions based on the hypothesis that genes present in the vicinity of disease causing genes in the network are highly likely to be involved with a particular or a similar disease48. Several studies have identified novel disease genes through the use of protein-protein interaction networks and have revealed that proteins involved in cancer are extremely interconnected49,50.

 

In the present study, a computational approach was applied to analyze the interactions of beta-tubulin protein through the use of STRING database, as it is feasible to obtain information from verified experimental sources to literature; thus providing a wider base for analyzing the protein interactome. The output of our interaction network clearly indicates that cluster 4 proteins involved in cancer are namely HIF1A, HSP90AA1, TP53, CDKN1A, HDAC2 and HDAC1, of which HDAC1 and HDAC2 are also involved in epigenetics, transcriptional misregulation of cancer and cancer pathways. It is known that the HDAC family of proteins are key components of histone modification complexes. The results obtained from our analysis  clearly indicates that HDAC1 and HDAC2 have associated with multiple roles such as Transcriptional regulation of cancer, Pathways in cancer and epigenetics. Therefore, we conclude from our analysis that targeting HDCA1 and HDCA2 which only curbs the tumor activity while not disturbing the usual activity. Such drugs are the need of the hour considering the several side effects of chemotherapy in the current scenario. Hence, we conclude HDAC1 and HDAC2 could be used as as putative targets for anti-cancer drugs. 

 

ACKNOWLEDGMENTS:

The authors express deep sense of gratitude to the management of Vellore Institute of Technology for all the support, assistance and constant encouragements to carry out this work.

 

REFERENCES:

1.     Wang Y, O’Brate A, Zhou W, Giannakakou P. Resistance to microtubule-stabilizing drugs involves two events: beta-tubulin mutation in one allele followed by loss of the second allele. Cell Cycle. 4(12); 2005:1847–53.

2.     Jordan MA, Wilson L. Microtubules and actin filaments: dynamic targets for cancer chemotherapy. Curr Opin Cell Biol. 10(1); 1998:123–30.

3.     Nogales E, Wolf SG, Downing KH. Structure of the alpha beta tubulin dimer by electron crystallography. Nature. 391(6663); 1998:199–203.

4.     Pasquier E, Kavallaris M. Microtubules: a dynamic target in cancer therapy. IUBMB Life. 60(3); 2008:165–70.

5.     Dumontet C, Jordan MA. Microtubule-binding agents: a dynamic field of cancer therapeutics. Nat Rev Drug Discov. 9(10); 2010:790–803.

6.     Kanika Verma, K Ramanathan. Investigation of Paclitaxel Resistant R306C Mutation in βTubulin – A Computational Approach. J Cell Biochem. 1324; 2015:1318–24.

7.     Whipple RA, Cheung AM, Martin SS. Detyrosinated microtubule protrusions in suspended mammary epithelial cells promote reattachment. Exp Cell Res. 313(7); 2007:1326–36.

8.     Schoumacher M, Goldman RD, Louvard D, Vignjevic DM. Actin, microtubules, and vimentin intermediate filaments cooperate for elongation of invadopodia. J Cell Biol. 189(3); 2010:541–56.

9.     Magiera MM, Janke C. Post-translational modifications of tubulin. Curr Biol. 24(9); 2014:R351–4.

10.  Levallet G, Bergot E, Antoine M, Creveuil C, Santos AO, Beau-Faller M, et al. High TUBB3 expression, an independent prognostic marker in patients with early non-small cell lung cancer treated by preoperative chemotherapy, is regulated by K-Ras signaling pathway. Mol Cancer Ther. 11(5); 2012:1203–13.

11.  Sève P, Mackey J, Isaac S, Trédan O, Souquet P-J, Pérol M, et al. Class III beta-tubulin expression in tumor cells predicts response and outcome in patients with non-small cell lung cancer receiving paclitaxel. Mol Cancer Ther. 4(12); 2005:2001–7.

12.  Ferlini C, Raspaglio G, Cicchillitti L, Mozzetti S, Prislei S, Bartollino S, et al. Looking at drug resistance mechanisms for microtubule interacting drugs: does TUBB3 work? Curr Cancer Drug Targets. 7(8); 2007:704–12.

13.  Dumontet C, Jordan MA, Lee FFY. Ixabepilone: targeting betaIII-tubulin expression in taxane-resistant malignancies. Mol Cancer Ther. 8(1); 2009:17–25.

14.  Alkan F, Erten C. SiPAN: simultaneous prediction and alignment of protein-protein interaction networks. Bioinformatics. 31(14); 2015:2356–63.

15.  Wang 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. American Chemical Society. 9(12);2010:6665–77.

16.  Szklarczyk D, Franceschini A, Wyder S, Forslund K, Heller D, Huerta-Cepas J, et al. STRING v10: protein-protein interaction networks, integrated over the tree of life. Nucleic Acids Res. 43; 2015:D447–52.

17.  Li J, Zhu X, Chen JY. Building disease-specific drug-protein connectivity maps from molecular interaction networks and PubMed abstracts. PLoS Comput Biol. 5(7); 2009:e1000450.

18.  MacQueen J. Some methods for classification and analysis of multivariate observations. In: Proceedings of the Fifth Berkeley Symposium on Mathematical Statistics and Probability, Volume 1: Statistics. The Regents of the University of California; 1967.

19.  Reed SI. The ubiquitin-proteasome pathway in cell cycle control. Results Probl Cell Differ. 42; 2006:147–81.

20.  Teixeira LK, Reed SI. Ubiquitin ligases and cell cycle control. Annu Rev Biochem. 82; 2013:387–414.

21.  Nakayama KI, Nakayama K. Ubiquitin ligases: cell-cycle control and cancer. Nat Rev Cancer. 6(5); 2005:369–81.

22.  Boyault C, Sadoul K, Pabion M, Khochbin S. HDAC6, at the crossroads between cytoskeleton and cell signaling by acetylation and ubiquitination. Oncogene. 26(37); 2007:5468–76.

23.  Zhang Y, Sun Z-W, Iratni R, Erdjument-Bromage H, Tempst P, Hampsey M, et al. SAP30, a Novel Protein Conserved between Human and Yeast, Is a Component of a Histone Deacetylase Complex. Mol Cell. 1(7); 1998:1021–31.

24.  Yao Y-L, Yang W-M. The metastasis-associated proteins 1 and 2 form distinct protein complexes with histone deacetylase activity. J Biol Chem. 278(43); 2003:42560–8.

25.  Sawan C, Herceg Z. Histone modifications and cancer. Adv Genet. 70; 2010:57–85.

26.  Cheung P, Lau P. Epigenetic regulation by histone methylation and histone variants. Mol Endocrinol. 19(3); 2005:563–73.

27.  Smith ND, Rubenstein JN, Eggener SE, Kozlowski JM. The p53 tumor suppressor gene and nuclear protein: basic science review and relevance in the management of bladder cancer. J Urol. 169(4); 2003:1219–28.

28.  Chang F, Syrjänen S, Kurvinen K, Syrjänen K. The p53 tumor suppressor gene as a common cellular target in human carcinogenesis. Am J Gastroenterol. 88(2); 1993:174–86.

29.  Bansal N, Kadamb R, Mittal S, Vig L, Sharma R, Dwarakanath BS, et al. Tumor Suppressor Protein p53 Recruits Human Sin3B/HDAC1 Complex for Down-Regulation of Its Target Promoters in Response to Genotoxic Stress. Gartel AL, editor. PLoS One. 6(10); 2011:e26156.

30.  Kadamb R, Mittal S, Bansal N, Batra H, Saluja D. Sin3: Insight into its transcription regulatory functions. Eur J Cell Biol. 92(8-9); 2013:237–46.

31.  Sun X-X, DeVine T, Challagundla KB, Dai M-S. Interplay between Ribosomal Protein S27a and MDM2 Protein in p53 Activation in Response to Ribosomal Stress. J Biol Chem. 286(26); 2011:22730–41.

32.  Chen D, Li M, Luo J, Gu W. Direct Interactions between HIF-1 and Mdm2 Modulate p53 Function. J Biol Chem. 278(16); 2003:13595–8.

33.  Walerych D, Kudla G, Gutkowska M, Wawrzynow B, Muller L, King FW, et al. Hsp90 chaperones wild-type p53 tumor suppressor protein. J Biol Chem. 279(47); 2004:48836–45.

34.  Gartel AL, Tyner AL. The Role of the Cyclin-dependent Kinase Inhibitor p21 in Apoptosis. Mol Cancer Ther. 1(8); 2002:639–49.

35.  Elbendary AA, Cirisano FD, Evans AC, Davis PL, Iglehart JD, Marks JR, et al. Relationship between p21 expression and mutation of the p53 tumor suppressor gene in normal and malignant ovarian epithelial cells. Clin Cancer Res. 2(9); 1996:1571–5.

36.  Yi J, Luo J. SIRT1 and p53, effect on cancer, senescence and beyond. Biochim Biophys Acta - Proteins Proteomics. 1804(8); 2010:1684–9.

37.  Solomon JM, Pasupuleti R, Xu L, McDonagh T, Curtis R, DiStefano PS, et al. Inhibition of SIRT1 Catalytic Activity Increases p53 Acetylation but Does Not Alter Cell Survival following DNA Damage. Mol Cell Biol. 26(1); 2005:28–38.

38.  Elengoe A, Naser MA, Hamdan S. A Novel Protein Interaction between Nucleotide Binding Domain of Hsp70 and p53 Motif. Int J Genomics. 2015; 2015:1–6.

39.  Hammond JW, Cai D, Verhey KJ. Tubulin modifications and their cellular functions. Curr Opin Cell Biol. 20(1); 2008:71–6.

40.  Park S-H, Zhu Y, Ozden O, Kim H-S, Jiang H, Deng C-X, et al. SIRT2 is a tumor suppressor that connects aging, acetylome, cell cycle signaling, and carcinogenesis. Translational Cancer Research.1(1);2012: 15–21.

41.  Yeh T-S, Hsieh R-H, Shen S-C, Wang S-H, Tseng M-J, Shih C-M, et al. Nuclear betaII-tubulin associates with the activated notch receptor to modulate notch signaling. Cancer Res. 64(22); 2004:8334–40.

42.  Borggrefe T, Oswald F. The Notch signaling pathway: Transcriptional regulation at Notch target genes. Cell Mol Life Sci. 66(10); 2009:1631–46.

43.  Mahajan K, Mahajan NP. PI3K-independent AKT activation in cancers: a treasure trove for novel therapeutics. J Cell Physiol. 227(9); 2012:3178–84.

44.  Brunet A, Bonni A, Zigmond MJ, Lin MZ, Juo P, Hu LS, et al. Akt Promotes Cell Survival by Phosphorylating and Inhibiting a Forkhead Transcription Factor. Cell. 96(6); 1999:857–68.

45.  Zhang S, Tang Q, Xu F, Xue Y, Zhen Z, Deng Y, et al. RhoA regulates G1-S progression of gastric cancer cells by modulation of multiple INK4 family tumor suppressors. Mol Cancer Res. 7(4); 2009:570–80.

46.  Pawson T, Nash P. Protein-protein interactions define specificity in signal transduction. Genes andamp; Dev. 14(9); 2000:1027–47.

47.  Ozgür A, Vu T, Erkan G, Radev DR. Identifying gene-disease associations using centrality on a literature mined gene-interaction network. Bioinformatics. 24(13); 2008:i277–85.

48.  Wachi S, Yoneda K, Wu R. Interactome-transcriptome analysis reveals the high centrality of genes differentially expressed in lung cancer tissues. Bioinformatics. 21(23); 2005:4205–8.

49.  Goehler H, Lalowski M, Stelzl U, Waelter S, Stroedicke M, Worm U, et al. A protein interaction network links GIT1, an enhancer of huntingtin aggregation, to Huntington’s disease. Mol Cell. 15(6); 2004:853–65.

50.  Jonsson PF, Bates PA. Global topological features of cancer proteins in the human interactome. Bioinformatics. 22(18); 2006:2291–7.

 

 

 

 

 

 

 

Received on 30.10.2015             Modified on 16.11.2015

Accepted on 21.11.2015           © RJPT All right reserved

Research J. Pharm. and Tech. 8(12): Dec., 2015; Page 1679-1684

DOI: 10.5958/0974-360X.2015.00303.0