CC BY
16
https://doi.org/10.29013/ELBLS-20-4-72-78
William Hou, High School Student Ivy Mind Academy, Skillman, New Jersey E-mail: william.hou9@gmail.com Betty Wang, Employment: IvyMind Academy E-mail: betty.wang@ivymind.org
IDENTIFYING GENE MUTATIONS MOST COMMONLY ASSOCIATED WITH GLIOBLASTOMA
Abstract. Certain gene mutations are often associated with the development and presence of glioblastoma (GBM). The most commonly associated gene mutants with GBM are found to be those of tumor protein p53 (TP53), phosphatase and tensin homolog (PTEN), and epidermal growth factor receptor (EGFR), all ofwhich have a major role in regulating cell proliferation. TP53 had a mutation frequency per kilo base (M/kb) of 28.8, PTEN of 10.85, and EGFR of 7.25. Somatic mutations in specific regions of these genes have been correlated with lower survival rates in GBM patients often due to irregularities in the functions of the genes' proteins. These regions include the DNA-binding domain of TP53, the C2 domain of PTEN, and amino acid 289 in EGFR. Mutations occurring in such regions can alter the genes' proteins; altered proteins with inhibited functions cannot regulate cell proliferation to the same extent, boosting the oncogenesis of GBM and leading to poorer prognoses. EGFR and PTEN are also associated with focal adhesion shown through an enrichment analysis, further suggesting that mutations in these genes lead to inhibited abilities to trigger cell apoptosis, a function that is critical for regulating cell proliferation. The majority of deleterious mutations are missense and truncating, suggesting that these types of mutations have the largest impact on GBM survival rates. Further research into TP53, PTEN, and EGFR may bring new treatments targeting these genes, allowing for a better prognosis in GBM patients and potentially other cancers.
Keywords: glioblastoma (GBM), gene mutation, bioinformatics analysis.
Introduction monly associated with the development of GBM and
GBM is a lethal, aggressive brain tumor, mak- encourage further research into these select genes for ing up a large portion of all brain cancers [1]. While a better understanding of the tumor. Additionally, treatments are available for GBM, the prognosis the methodology used may be applied to data sets for is poor, and the cancer occurs more frequently in other diseases to determine genes most commonly older patients at a median age of 64 [2]. Treatments associated with those diseases. include surgery, radiation, and chemotherapy, but The data used for this analysis were taken from there is currently no cure for glioblastoma [1]. The XenaBrowser and were originally from The Cancer exact cause of GBM is unknown, as most cases occur Genome Atlas (TCGA), a program dedicated to irregularly and are not inherited [1]. This study is cataloguing genetic mutations associated with vari-designated to find genes whose mutations are com- ous cancers. The data set consists of 314 samples
from GBM patients. Out of the 314 samples taken, only genes with the highest M/kb are used in the analysis because of their higher correlation with GBM.
Methods
The data taken from XenaBrowser are analyzed to find genes with the most frequent mutations [3], and is sorted using Linux commands to find the most frequently appearing genes in the data. These genes and their frequencies are tabulated in Microsoft Excel, with their respective gene lengths and full names. The gene lengths are taken from GenCodeGenes [4], and the genes' full names are matched with the gene symbols. Once the gene lengths and full names are matched, the frequency of each gene is then normalized with its gene length, and the genes are sorted by the highest frequency of mutations per base length (M/kb). Genes with less than 1 M/kb were removed from the analysis due to their relative insignificance and for conciseness, and only genes with an M/kb of 4 or more are tabulated. Data of only genes with more than 1 M/kb are extracted when analyzing the distribution of mutation types. The extracted data are sorted using Linux commands to determine the frequency of each mutation type, and the results are tabulated in Microsoft Excel.
A TCGA Glioblastoma (GBM) study is selected in CBioPortal [5], and is matched with gene symbols TP53, PTEN, and EGFR, which have some of the highest M/kb. A mutation chart is created from the gene symbols inputted, which marks locations of mutations in GBM patients onto each gene.
Detailed analyses of the top genes are conducted using past studies that associate the genes with GBM. An enrichment analysis of genes with the most M/kb is done through DAVID Bioinformatics Resources in order to find associated terms with the genes, linking the genes' mutations to the impairment of their functions [6; 7]. The top genes are submitted as a gene list into DAVID, and a functional annotation chart is created from the list, giving the most associated terms with the genes.
Results
Three of the genes with the most mutations per kilo base (M/kb) are found to be TP53, PTEN, and EGFR. After normalizing the frequency of mutations for every kilo base in each gene, 329 out of the 15,350 genes found in the data set had one or more mutations per kilo base. For these 329 genes, the vast majority of mutations are missense, which changes a single amino acid in the coded protein [8], and makes up 80.2% of all mutations (Table 1).
Table 1. - Frequency of Mutation Types in Top 329 Normalized Genes
Mutation Type Frequency
Missense 3927
Nonsense 337
3'UTR 115
Splice Site 112
Frame Shift Del 98
RNA 87
Intron 87
5'UTR 55
In Frame Del 28
Frame Shift Ins 24
3'Flank 11
In Frame Ins 5
5'Flank 5
Translation Start Site 3
Nonstop Mutation 2
However, though only 92 genes were found to have an M/kb of 2 or higher, genes with a lower frequency may still influence GBM. The top 92 genes are sorted, and closer analysis is done on genes TP53, PTEN, and EGFR because of their high value of M/kb (Table 2). Despite that POM121L12 has the third highest M/kb (Table 2), the gene is not analyzed because of its relatively low frequency compared with TP53, PTEN, and EGFR. TP53 had an M/kb of roughly 28.8, the highest from the data provided by TCGA (Table 2), while PTEN and EGFR had an M/kb of 10.85 and 7.25 respectively (Table 2). The majority of TP53 mutations are point mutations that occur in the DNA-binding domain (Fig-
ure 1). A number of PTEN mutations occur in the C2 domain (Figure 2), and a significant amount of EGFR mutations occur on amino acid 289, with 18 mutations out of 67 samples (Figure 3). Additionally, an enrichment analysis through DAVID Bioin-
formatics Resources found that the top 329 genes with the highest frequency per base length are highly associated with focal adhesion and ECM-receptor interaction, with p-values of 5.75E-10 and 1.08E-06 respectively (Table 3).
Table 2. - Frequency of Gene Mutations per Kilo Base Length
Gene Symbol Gene Name Frequency Length Frequency per Kilo Base Length (M/kb)
TP53 tumor protein p53 113 3924 28.797145770
PTEN phosphatase and tensin homolog 109 10048 10.847929936
POM121L12 POM121 transmembrane nucleoporin-like 12 10 1269 7.880220646
EGFR epidermal growth factor receptor 94 12961 7.252526811
TRIM51 tripartite motif-containing 51 12 1832 6.550218341
TUBA3C tubulin, alpha 3c 9 1551 5.802707930
KIF2B kinesin family member 2B 13 2335 5.567451820
DCAF12L2 DDB1 and CUL4 associated factor 12-like 2 9 1673 5.379557681
RB1 retinoblastoma 1 30 6169 4.863024801
TPTE2 transmembrane phosphoinositide 3-phos-phatase and tensin homolog 2 11 2288 4.807692308
OR8K3 In multiple Geneids 9 1878 4.792332268
NLRP5 NLR family, pyrin domain containing 5 19 4023 4.722843649
ZFP42 ZFP42 zinc finger protein 12 2651 4.526593738
MAGEC2 melanoma antigen family C2 9 1991 4.520341537
UGT2B28 UDP glucuronosyltransferase 2 family, polypeptide B28 8 1833 4.364429896
PIK3CA phosphatidylinositol-4,5-bisphosphate 3-kinase, catalytic subunit alpha 40 9411 4.250345341
Category Term Count % PValue
1 2 3 4 5
KEGG PATHWAY hsa04510:Focal adhesion 22 0.04931077 5.75E-10
KEGG PATHWAY hsa04512:ECM-receptor interaction 12 0.026896784 1.08E-06
KEG G_PATHWAY hsa04151:PI3K-Akt signaling pathway 23 0.051552169 1.14E-06
KEGG PATHWAY hsa05214:Glioma 10 0.022413986 5.00E-06
KEGG PATHWAY hsa05222:Small cell lung cancer 11 0.024655385 6.62E-06
KEGG PATHWAY hsa05146: Amoebiasis 12 0.026896784 7.77E-06
KEGG PATHWAY hsa05218:Melanoma 10 0.022413986 1.05E-05
GOTERM_BP_DI-RECT GO :0030198~extracellular matrix organization 16 0.035862378 1.87E-06
Table 3. - Associated Terms with Top 329 Normalized Genes for Glioblastoma
1 2 3 4 5
GOTERM_BP_DI-RECT G0:0006936~muscle contraction 12 0.026896784 2.66E-06
GOTERM_BP_DI-RECT G0:0007156~homophilic cell adhesion via plasma membrane adhesion molecules 14 0.031379581 4.12E-06
GOTERM_BP_DI-RECT G0:0051209~release of sequestered calcium ion into cytosol 8 0.017931189 6.20E-06
R282W/Gfs*63
-
......
H P53_.. P53 I P53 tetramer^^^^^^^^n
Figure 1. Locations of TP53 Mutations in Glioblastoma Patients (taken from CBioPortal)
11 ....... y .....
I DSPc PTEN C2
Figure 2. Locations of PTEN Mutations in Glioblastoma Patients (taken from CBioPortal)
A289V/T/I and 1 more
• •■ - '" ■ • * " . V - . .
Recep_L_.. Furin-like Recep_L_.. H GF_recep_IV Pkinase_Tyr
1 1 1 1 1 1 1 1 1 1 1 1 II
Figure 3. Locations of EGFR Mutations in Glioblastoma Patients (taken from CBioPortal)
TP53 Association with Glioblastoma
The main function of TP53 is to regulate cell division by creating tumor protein 53 (p53), which prevents uncontrollable cell reproduction and is done by triggering cell apoptosis or other genes to repair a cell's DNA. Half of all cancers are found to have somatic mutations in TP53, most of which changes p53 so that the protein is no longer able to regulate cell division, leading to tumor developments. Because of a reduced ability to control cell proliferation, modified p53 proteins are often linked not only to
GBM, but a wide variety of other cancers; according to Wang et al. (2013), they may also increase GBM resistance to temozolomide and therefore result in ineffectiveness of certain treatments involving such a drug (Wang et al, 2013).
Out of the 314 GBM samples provided by Xen-aBrowser [3], 113 TP53 mutations were found (Table 2). Végran et al. (2013) found that 90% of TP53 mutations occur in the TP53 DNA binding to domain, but only missense mutations usually significantly hinder p53 from binding DNA [10].
0
0
100
200
300
393aa
5
0
100
200
300
403aa
0
Similarly, roughly 93% of TP53 mutations provided by CBioPortal occur in the DNA binding domain with all mutations being either missense or truncating (Figure 1). Proper DNA binding is essential for protein p53 to perform its functions, including triggering apoptosis, and impaired DNA binding of p53 was linked to higher cancer rates in mice [11].
PTEN Association with Glioblastoma
PTEN is responsible for a phosphatase enzyme that controls cell division and contributes to the triggering of apoptosis; consequently, the PTEN phosphatase acts as a tumor suppressor. As such, mutations of PTEN often modify the PTEN phosphatase so that the enzyme can no longer perform its main function to the same degree. PTEN inhibits focal adhesions, which in turn controls cell proliferation [12], so mutations of PTEN can change focal adhesions and thus aid tumor progression [13]. Most often, PTEN mutations are missense or truncating (Figure 2). Subsequently, PTEN mutations can lead to shorter survival rates in GBM patients. A meta-analysis found that GBM patients with PTEN mutations generally had a poorer prognosis, though the results are tentative and could have been biased to some extent [14]. Another study, though focused on gliomas in general as opposed to specifically GBM, yielded similar results, adding that PTEN mutations occurred later in glioma progression [15].
In GBM samples, a large amount of truncating and missense PTEN mutations occur in the C2 domain (Figure 2). Since the C2 domain is used to inhibit cell migration, the domain may have a role in tumor suppression [16]. Mutations occurring in the C2 domain can likely affect the ability of PTEN to perform its functions and allow faster GBM progression.
EGFR Association with GBM
EGFR is involved in the production of the protein epidermal growth factor receptor, which binds to ligands in order dimerize with other epidermal
growth factor receptors, leading to the triggering of cell proliferation. Brennan et al. (2013) found that about 57% of GBMs in one study [17], while Xu et al. (2017) found EGFR overexpression in 60% of primary glioblastomas and 10% of secondary glioblastomas [18]. Although EGFR mutations are highly correlated with the development of GBM, attempts at targeting the gene have not yet been successful [19]. An et al. (2018) yielded similar results, as it found that while EGFR amplification and EGFRvIII appear commonly in GBM, immunotherapies using EGFR inhibitors have been ineffective [20]. Of the GBM data from CBioPortal [5], 18 out of 67 mutations from 543 samples occurred on alanine 289 (Figure 3), and patients generally have a worse prognosis if these mutations occur [21], most likely due to the mutations' negative impacts on protein EGFR's functions. Conclusion
The purpose of this bioinformatics analysis is to identify several genes that are most frequently associated with the development and occurrence of glioblastoma (GBM). After the analysis based on GBM data from TCGA was conducted, genes TP53, PTEN, and EGFR are found to be highly associated with the oncogenesis of GBM, and may tentatively be applied to other types of cancers, especially gliomas. Mutations in these genes, specifically missense mutations, may result in altered proteins with inhibited functions. PTEN and EGFR are also associated with focal adhesions; a loss of ability to control focal adhesions likely leads to uncontrolled cell proliferation and tumor progression. Successfully targeting these genes for therapeutic treatments may be able to prolong survival in GBM patients and improve the prognosis for this cancer; however, such current attempts have proven to be unsuccessful. The methodology used in this bioinformatics analysis may be applied to other diseases with available data pertaining to gene mutations found in patients, which may assist in identifying key genes related to such diseases.
References:
1. Davis M. E. Glioblastoma: Overview of Disease and Treatment. Clinical journal of oncology nursing, 20(5 Suppl), 2016. - P. 2-8. URL: https://doi.org/10.1188/16.CJON.S1.2-8
2. Thakkar J. P., Dolecek T. A., Horbinski C., et al. Epidemiologic and molecular prognostic review of glioblastoma. Cancer Epidemiol Biomarkers Prev. 2014; 23(10): 1985-1996. Doi:10.1158/1055-9965. EPI-14-0275
3. Goldman M. J., Craft B., Hastie M. et al. Visualizing and interpreting cancer genomics data via the Xena platform. Nat Biotechnol. 2020.
4. Frankish A., et. al. GENCODE reference annotation for the human and mouse genomes". 2018.
5. Cerami et al., Cancer Discov. 2012 and Gao et al., Sci. Signal. 2013.
6. Huang D. W., Sherman B. T., Lempicki R. A. Systematic and integrative analysis of large gene lists using DAVID Bioinformatics Resources. Nature Protoc. 2009; 4(1): 44-57.
7. Huang D. W., Sherman B. T., Lempicki R. A. Bioinformatics enrichment tools: paths toward the comprehensive functional analysis of large gene lists. Nucleic Acids Res. 2009; 37(1): 1-13.
8. Zhou X., Iversen E. S.Jr. & Parmigiani G. Classification of Missense Mutations of Disease Genes. Journal of the American StatisticalAssociation, 2005; 100(469), 51-60. URL: https://doi. org/10.1198/016214504000001817
9. Wang X., Chen J. X., Liu J. P., You C., Liu Y. H., Mao Q. Gain of function of mutant TP53 in glioblastoma: prognosis and response to temozolomide. Ann Surg Oncol. 2014; 21(4): 1337-1344. Doi:10.1245/ s10434-013-3380-0
10. Vegran F., Rebucci M., Chevrier S., Cadouot M., Boidot R. & Lizard-Nacol S. Only missense mutations affecting the DNA binding domain of p53 influence outcomes in patients with breast carcinoma. PloS one, 8(1), e55103. 2013. URL: https://doi.org/10.1371/journal.pone.0055103
11. Timofeev O., Schlereth K., Wanzel M., et al. p 53 DNA binding cooperativity is essential for apoptosis and tumor suppression in vivo. Cell Rep. 2013; 3(5): 1512-1525. Doi:10.1016/j.celrep.2013.04.008
12. Tamura M., Gu J., Matsumoto K., Aota S., Parsons R., Yamada K. M. Inhibition of cell migration, spreading, and focal adhesions by tumor suppressorPTEN. Science. 1998; 280(5369): 1614-1617. Doi:10.1126/ science.280.5369.1614
13. Tamura M., Gu J., Takino T., Yamada K. M. Tumor suppressor PTEN inhibition of cell invasion, migration, and growth: differential involvement of focal adhesion kinase and p130Cas. Cancer Res. 1999; 59(2): 442-449.
14. Han F., Hu R., Yang H., Liu J., Sui J., Xiang X., Wang F., Chu L. & Song S. PTEN gene mutations correlate to poor prognosis in glioma patients: a meta-analysis. OncoTargets and therapy, 2016; 9, 3485-3492. URL: https://doi.org/10.2147/OTT.S99942
15. Yang Y., Shao N., Luo G., et al. Mutations of PTEN gene in gliomas correlate to tumor differentiation and short-term survival rate. Anticancer Res. 2010; 30(3): 981-985.
16. Raftopoulou M., Etienne-Manneville S., Self A., Nicholls S., Hall A. Regulation of cell migration by the C2 domain of the tumor suppressor PTEN. Science. 2004; 303(5661): 1179-1181. Doi:10.1126/sci-ence.1092089
17. Brennan C. W., Verhaak R. G., Mc Kenna A., Campos B., Noushmehr H., Salama S. R., Zheng S., Chakravar-ty D., Sanborn J. Z., Berman S. H., Beroukhim R., Bernard B., Wu C.J., Genovese G., Shmulevich I., Barnholtz-
Sloan J., Zou L., Vegesna R., Shukla S. A., Ciriello G., ... TCGA Research Network. The somatic genomic landscape of glioblastoma. Cell, 2013; 155(2), 462-477. URL: https://doi.org/10.10167j.cell.2013.09.034
18. Xu H., Zong H., Ma C., Ming X., Shang M., Li K., He X., Du H. & Cao L. Epidermal growth factor receptor in glioblastoma. Oncologyletters, 2017; 14(1), 512-516. URL: https://doi.org/10.3892/ol.2017.6221
19. Hatanpaa K. J., Burma S., Zhao D. & Habib A. A. Epidermal growth factor receptor in glioma: signal transduction, neuropathology, imaging, and radioresistance. Neoplasia (New York, N.Y.), 2010; 12(9), 675-684. URL: https://doi.org/10.1593/neo.10688
20. An Z., Aksoy O., Zheng T., Fan Q. W. & Weiss W. A. Epidermal growth factor receptor and EGFRvIII in glioblastoma: signaling pathways and targeted therapies. Oncogene, 2018; 37(12), 1561-1575. URL: https://doi.org/10.1038/s41388-017-0045-7
21. Binder Z. A., Thorne A. H., Bakas S., et al. Epidermal Growth Factor Receptor Extracellular Domain Mutations in Glioblastoma Present Opportunities for Clinical Imaging and Therapeutic Development. Can-cerCell. 2018; 34(1): 163-177.e7. Doi:10.1016/j.ccell.2018.06.006
