�������� ����� ��

Sequence and expression variations in 23 genes involved in mitochondrial andnon-mitochondrial apoptotic pathways and risk of oral leukoplakia and cancer

Sayantan Datta, Anindita Ray, Richa Singh, Pinaki Mondal, AnalabhaBasu, Navonil De Sarkar, Mousumi Majumder, Guruparasad Maiti, AradhitaBaral, Ganga Nath Jha, Indranil Mukhopadhyay, Chinmay Panda, ShantanuChowdhury, Saurabh Ghosh, Susanta Roychoudhury, Bidyut Roy

PII: S1567-7249(15)30022-2DOI: doi: 10.1016/j.mito.2015.09.001Reference: MITOCH 1033

To appear in: Mitochondrion

Received date: 25 February 2015Revised date: 15 August 2015Accepted date: 18 September 2015

Please cite this article as: Datta, Sayantan, Ray, Anindita, Singh, Richa, Mondal, Pinaki,Basu, Analabha, De Sarkar, Navonil, Majumder, Mousumi, Maiti, Guruparasad, Baral,Aradhita, Jha, Ganga Nath, Mukhopadhyay, Indranil, Panda, Chinmay, Chowdhury,Shantanu, Ghosh, Saurabh, Roychoudhury, Susanta, Roy, Bidyut, Sequence and ex-pression variations in 23 genes involved in mitochondrial and non-mitochondrial apop-totic pathways and risk of oral leukoplakia and cancer, Mitochondrion (2015), doi:10.1016/j.mito.2015.09.001

This is a PDF file of an unedited manuscript that has been accepted for publication.As a service to our customers we are providing this early version of the manuscript.The manuscript will undergo copyediting, typesetting, and review of the resulting proofbefore it is published in its final form. Please note that during the production processerrors may be discovered which could affect the content, and all legal disclaimers thatapply to the journal pertain.

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

1

Title: Sequence and expression variations in 23 genes involved in mitochondrial and non-

mitochondrial apoptotic pathways and risk of oral leukoplakia and cancer

Authors: Sayantan Datta1,2

, Anindita Ray1, Richa Singh

1, Pinaki Mondal

3, Analabha Basu

4,

Navonil De Sarkar1,5

, Mousumi Majumder1,6

, Guruparasad Maiti7, Aradhita Baral

8, Ganga Nath

Jha9, Indranil Mukhopadhyay

1, Chinmay Panda

7, Shantanu Chowdhury

8, Saurabh Ghosh

1,

Susanta Roychoudhury3 and Bidyut Roy

1

Address: 1

Human Genetics Unit, Indian Statistical Institute, Kolkata, India

2, present address: John Hopkins School of Medicine, Baltimore, USA

3Molecular and Human

Genetics Division, Indian Institute of Chemical Biology, CSIR,

Kolkata, India

4National Institute of Biomedical Genomics, Kalyani, West Bengal, India

5present address: Fred Hutchinson Cancer Research center, University of Washington, Seattle, USA

6Present address: University of Western Ontario, Departments of Anatomy and Cell Biology, Oncology, and

Pathology, London, Ontario, Canada

7Department of Oncogene Regulation, Chittaranjan National Cancer Institute, Kolkata, India

8Proteomics and Structural Biology Unit, Institute of Genomics and Integrative Biology, CSIR, Delhi, India

9Department of Anthropology, Vinoba Bhave University, Hazaribagh, Jharkhand

Running Title: Apoptotic pathway genes and risk of oral cancer

Keywords: apoptotic pathway genes, SNPs, expression, risk, oral cancer, leukoplakia

Abbreviations Used: MAF: minor allele frequency; SNP: single nucleotide polymorphism;

UTR: untranslated region.

Financial supports: Indian Statistical Institute, Kolkata, India and Department of

Biotechnology, Government of India.

Corresponding author: Bidyut Roy, Human Genetics, Unit, Indian Statistical Institute, 203 BT

Road, Kolkata 700108, India. +91-33-25753213, broy@isical.ac.in

Acknowledgements: Authors thank Prof. R. R. Paul and Prof. Jay Gopal Ray, oral pathologists,

for his active support during sample collection and diagnosis of patients. Grant Supports: Indian

Statistical Institute and Department of Biotechnology, India

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

2

Abstract

Oral cancer is usually preceded by pre-cancerous lesion and related to tobacco abuse. Tobacco

carcinogens damage DNA and cells harboring such damaged DNA normally undergo apoptotic

death, but cancer cells are exceptionally resistant to apoptosis. Here we studied association

between sequence and expression variations in apoptotic pathway genes and risk of oral cancer

and precancer. Ninety nine tagSNPs in 23 genes, involved in mitochondrial and non-

mitochondrial apoptotic pathways, were genotyped in 525 cancer and 253 leukoplakia patients

and 538 healthy controls using Illumina Golden Gate assay. Six SNPs (rs1473418 at BCL2;

rs1950252 at BCL2L2; rs8190315 at BID; rs511044 at CASP1; rs2227310 at CASP7 and

rs13010627 at CASP10) significantly modified risk of oral cancer but SNPs only at BCL2,

CASP1and CASP10 modulated risk of leukoplakia. Combination of SNPs showed a steep

increase in risk of cancer with increase in “effective” number of risk alleles. In silico analysis of

published data set and our unpublished RNAseq data suggest that change in expression of BID

and CASP7 may have affected risk of cancer. In conclusion, three SNPs, rs1473418 in BCL2,

rs1950252 in BCL2L2 and rs511044 in CASP1, are being implicated for the first time in oral

cancer. Since SNPs at BCL2, CASP1 and CASP10 modulated risk of both leukoplakia and

cancer, so, they should be studied in more details for possible biomarkers in transition of

leukoplakia to cancer. This study also implies importance of mitochondrial apoptotic pathway

gene (such as BCL2) in progression of leukoplakia to oral cancer.

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

3

1. Introduction:

Oral cancer affects more than 263,000 people and kills about 127,000 of them worldwide

annually. In India alone, almost 70,000 people develop oral cancer and 50,000 die yearly (Ferlay

et al., 2010). Oral cancer almost always involves a two-step process – pre-cancerous lesion

followed by carcinoma (Reibel, 2003). A survey on Indian populations found that about 80% of

oral cancers were preceded by pre-cancerous lesions and oral leukoplakia being the most

common among them (Gupta et al., 1989). The relationship between tobacco abuse (both

smoking and smokeless forms) and damage to mucosal cells and DNA, leading to pre-cancerous

and cancerous lesions, is now well-established (Proia et al., 2006). Cells harboring severe DNA

damage normally enter a programmed cell death pathway known as apoptosis. However, cancer

cells are known to be exceptionally resistant to apoptotic death (Hanahan and Weinberg, 2000;

Zhivotovsky and Orrenius, 2006). Apoptosis, an intricately complex mechanism involving a

large number of gene products, is mediated by two distinct (mitochondrial i.e. intrinsic and non-

mitochondrial i.e. extrinsic) pathways leading to the activation of a set of effecter caspases. DNA

sequence variations in the apoptotic pathway genes have been reported to be associated with the

risk of several kinds of cancer and most of these polymorphisms were located in the functionally

important gene regions viz. exons, promoters, and 3‟ and 5‟ UTRs (Enjuanes et al., 2008; Frank

et al., 2006; Ghavami et al., 2009; Kim et al., 2009; Sun et al., 2007; Yang et al., 2008). Though

many studies have reported association of oral cancer with polymorphisms in the genes of the

carcinogen metabolism and DNA repair pathways, research elucidating the association of

apoptotic gene polymorphisms with oral cancer and pre-cancer has been few (Chen et al., 2007;

Hopkins et al., 2008; Scully and Bagan, 2009).

Here, we report results of a case-control association study based on genotyping of tag

SNPs from functionally important regions of 23 genes involved in mitochondrial and non-

mitochondrial apoptotic pathways in oral leukoplakia and cancer patients and healthy controls.

2. Materials and Methods:

2.1 Study population:

Procedures for blood sample collection and written informed consent were reviewed and

approved by the Institutional Ethical Committee. Cancer and leukoplakia patients with history of

tobacco use were recruited from Dr. R Ahmed Dental College and Hospital, a tertiary referral

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

4

center in Kolkata, India, with active help of dentists. All patients were confirmed for leukoplakia

or oral squamous cell carcinoma (OSCC) histopathologically. Disease-free healthy individuals

with regular tobacco habit were recruited as controls (Datta et al., 2007). Cancer (n=525) and

leukoplakia patients (n=253) and healthy controls (n=538) voluntarily donated blood for this

study. All patients and controls were personally interviewed to get information on age, sex,

occupation, alcohol consumption, type of tobacco habits, daily tobacco use frequency and

duration of habits.

Since about 90% of smokers used both cigarettes and bidis, data on bidi and cigarette

smokers were not analyzed separately. In both patient groups, only a few (<3%) individuals

consumed alcohol occasionally. Therefore, alcohol consumption was also not considered in

analysis. Individuals who smoke tobacco are termed „smokers‟ and those who dip/chew tobacco

are termed „chewers‟. Dose of tobacco smoking was measured as pack-years (PY) and 1PY is

defined as smoking 1 pack cigarette/bidi per day for one year. Lifetime smokeless tobacco

exposure was measured in terms of frequency of chewing/dipping per day multiplied by the

duration of habit. This is termed as chewing-year (CY) and 1CY is defined as chewing tobacco

once a day for one year. Some of the patients and controls had both tobacco smoking and

chewing habits (Datta et al., 2007).

2.2 SNP selection:

Candidate tag SNPs were selected in a set of 23 genes (Table 1) that were obtained from

a systematic Medline search for genes involved in apoptosis. Tag SNPs are defined as variants

with r2 ≥ 0.8 with other SNPs in HapMap individuals with European ancestry (CEU) (HapMap

project, 2003) and these tagSNPs were chosen using Tagger software. Selection of candidate

SNPs was done in three ways, mainly focused to include SNPs with putative functional effect in

protein structure and gene expression. Firstly, tag non-synonymous SNPs were selected from the

exonic region of the gene. Secondly, tag SNPs from putatively functional regions of the gene

viz.5‟UTR, 3‟UTR and 2kb regions flanking the gene were selected. Thirdly, SNPs previously

reported to be associated with cancer or other diseases were included. We selected only SNPs

with a minor allele frequency ≥0.01 in Caucasian population of the HapMap (CEU). All the

selected SNPs were then filtered by Illumina technology criteria (score ≥0.6 or Golden Gate

validated status, Illumina, Inc.).

A total of 99 SNP loci from 23 genes satisfied the above mentioned criteria and were

selected for final genotyping (Table 1). These 99 SNPs were distributed in 5‟ UTR and near gene

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

5

regions (n=26), coding regions (non-synonymous changes, n=28), and in 3‟ UTR and near gene

regions (n=45).

2.3 Genotyping:

Illumina Platform:

Genomic DNA, isolated from peripheral blood lymphocytes (Datta et al., 2007), was

quantified using Pico Green and diluted to a final concentration of 50ng/L. Genotyping was

carried out on Illumina Golden Gate genotyping platform and genotype clustering was performed

using Illumina Bead Studio (Version 3). Illumina established an assay score based on the

nucleotide composition of the DNA region and on the presence of duplicated or highly repetitive

sequences, palindromes, and neighboring polymorphisms. Polymorphisms showing an Illumina

quality score <0.6 were rejected for the final pool of genotyped SNPs. Eight DNA intra-assay

duplicates were included in each of 96-well plate assay system.

TaqMan platform:

Seven SNPs chosen randomly were re-genotyped in 300 individuals on TaqMan platform

(Applied Biosystems Inc.) for confirmation. These 300 individuals consisted of 100 randomly

chosen individuals from each of the control, leukoplakia and cancer groups. The assay was done

at the default annealing and extension temperature, recommended by the manufacturer, (a

denaturing step of 15 s at 95oC, followed by annealing and extension for 1 min at 60

oC for 40

cycles).

2.4 Data analysis:

Quality control of data and tests for association were performed using PLINK v1.07

(Purcell et al., 2007) and SPSS version 16.0 (SPSS, Chicago, IL). An initial pruning of the data

based on the following criteria was performed: (i) Individuals who had genotype data at less than

90% of the loci, (ii) SNP loci that had genotype data in less than 90% of the individuals, (iii)

SNP loci that had a minor allele frequency (MAF) less than 0.01 in cases and controls taken

together; and (iv) SNP loci that deviated significantly from HWE (Fisher‟s exact test) at level of

significance 0.001 in controls. Genotypic test of association (at 2 degrees of freedom) and allelic

test of association (at 1 degree of freedom) were carried out using logistic regression, adjusting

for age, sex and tobacco doses. In order to correct for multiple testing, False Discovery Rate

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

6

(FDR) procedure with an overall significance level of 0.05 (Benjamini and Hochberg, 1995) was

used.

The combined effect of six SNPs was determined through allele dosage analysis by

categorizing the subjects based on the number of “effective” risk alleles. The analysis included

only those individuals in whom genotypes at all six SNP loci were available. Since the effect

sizes (defined by the SNP-specific ORs) were not uniform, the allele dosage score of an

individual was computed by the weighted mean of the number of risk alleles at six SNPs (i.e., 2

for two risk alleles, 1 for one risk allele, and 0 for no risk allele in an individual) with weights as

the relative log ORs of different SNPs. The “effective” number of risk alleles harbored by each

individual was calculated using a weighted score:

k k

k Wi Ri / Wi

i = 1 i =1

where, k is the number of SNPs (k=6 in this study), Ri is the number of risk alleles at the ith

locus (0, 1or 2) and Wi is the natural logarithm of Odds Ratio (OR) of the risk allele at the ith

locus, i=1,2,…, k. Since the ORs of the different SNPs are not equal, the allele dosage scores

weigh the different SNPs proportionately. While the number of risk alleles that an individual

harbors is a discrete variable, the number of “effective” risk alleles is a continuous variable and

can range between 0 and 12. Therefore the “effective” number of risk alleles was divided into 4

classes (0-3, 3-4, 4-5 and 5-12) and risk of cancer in successive classes was calculated.

Since the number of individuals with fewer than two “effective” risk alleles was not

large, we pooled all such individuals with the group harboring three “effective” risk alleles and

considered this pooled set as a reference group. ORs and P values, for every unit increase in the

number of “effective” risk alleles, were calculated after adjusting for sex, age and tobacco dose

using logistic regression.

To explore whether genotypes of SNPs influence expression, in silico analysis was

performed using expression profile of data set series GSE6536 (Gene Expression Omnibus,

(http://www.ncbi.nlm.nih.gov/geo, (Stranger et al., 2007), 1000 genomes data base (Abecasis et

al., 2012) and Geuvadis RNA sequencing data base (Lappalainen et al., 2013). P-values were

calculated comparing mean of expression values of genes having different genotypes by

independent T- and Anova tests.

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

7

3. Results:

Cases and controls differed significantly with respect to demographic attributes such as

age, sex, smoking and chewing doses (Table 2); therefore, these were included as covariates in

the regression analysis. Thirty-five individuals and seven SNPs had call rates less than 90%, 12

SNPs failed HWE test in controls and 18 SNPs failed MAF test. These SNPs and individuals

were excluded from further analyses and data from 1281 individuals and 72 SNPs were included

for analysis. Genotypes obtained for blinded duplicate DNA samples in Illumina Golden Gate

genotyping platform were concordant for all SNPs. Similarly, genotypes determined by TaqMan

and Illumina platforms were completely concordant.

Six SNP loci in 6 different genes (rs1473418 at BCL2; rs1950252 at BCL2L2; rs8190315

at BID; rs511044 at CASP1; rs2227310 at CASP7 and rs13010627 at CASP10) were found to be

significantly associated with risk of cancer at both genotypic (data not shown) and allelic tests

after covariate adjustments and multiple testing corrections (Table 3). Four of these 6 SNPs, at

BCL2, BID, CASP1 and CASP7, increased risk of cancer whereas remaining 2 SNPs, at BCL2L2

and CASP10, decreased risk of cancer with respect to controls. In case of leukoplakia, two SNPs

(at BCL2 and CASP1) increased but SNP at CASP10 decreased risk of leukoplakia compared to

controls (Table 3). SNPs at BCL2, CASP1 and CASP7 (mentioned above) and rs3181179 at

CASP5 increased risk of cancer when compared to leukoplakia (data not shown). Results of

logistic regression for allelic test of association for all the 99 SNPs have been shown separately

(Supplementary Table 1).

SNPs at BCL2, BCL2L2 and CASP1 are present in the regulatory regions but expression

of these genes was not modulated by SNPs when checked in published data set. It is important to

mention that variant genotypes were few in data set of 4 populations (i.e. Caucasian, Japanese,

Han Chinese and Yoruba), so expression could not be compared properly. SNPs at BID, CASP7

and CASP10 would cause non-synonymous amino acid changes (i.e. replace of serine by glycine,

Aspartic acid by glutamic acid and valine by isoleucine, respectively). Interestingly, using

published expression data set, significant change in expression of BID and CASP7 was observed

(p=0.02 and p=0.01, respectively) when expression of these two genes in individuals with wild

genotype was compared with those in individuals with variant genotypes.

To study combined effect of risk alleles at six associated loci, “effective” number of risk

alleles was computed from number of risk alleles harboured by an individual and, then, odds

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

8

ratio at respective loci was estimated. Compared to individuals with three or less “effective” risk

alleles, which was taken as the reference, there was significant increment in risk of cancer with

an increase in number of “effective” risk alleles (Figure 1). There was a modest increase (OR-

6.02, and 95%CI:2.72-13.32) in risk of cancer among individuals who harboured 3 to 4

“effective” number of risk alleles but the risk increased steeply to 39.29 folds (95%CI: 12.56-

122.92) when the “effective” risk alleles were 4 to 5. Interestingly, if the “effective” number of

risk alleles was 5 or more, then risk of cancer was increased to 136.53 times (95%CI: 54.51-

341.94).

4. Discussion:

This study identified six SNP loci in six different genes of intrinsic and extrinsic

apoptotic pathways that were found to modify risk of oral cancer but only three of them could

modify risk of leukoplakia at both genotypic (data not shown) and allelic levels (Table 3). Of

these six SNPs, three (rs8190315 in BID; rs2227310 in CASP7 and rs13010627 in CASP10) have

been previously reported to be associated with several diseases including cancers (Chae et al.,

2011; Enjuanes et al., 2008; Ghavami et al., 2009; Kim et al., 2009; Lee et al., 2009; Yoo et al.,

2009). In this study, G allele at rs2227310 in CASP7 was found to increase risk of oral cancer

compared to both leukoplakia (data not shown) and controls (Table 3). This allele has previously

been found to increase risk of lung cancer (Yoo et al., 2009) and also confer poor prognosis in

gastric and colorectal cancers (Lee et al., 2009). Other SNPs at CASP7 were also found to

modulate risk of cervical cancer in Chinese population (Shi et al., 2015). CASP7 has several

isoforms with distinctly different roles. Non-functional isoform of CASP7 is thought to

regulate negatively the functional non- isoforms and rs2227310 produces an Asp to Glu change

in functional non- isoform of CASP7 causing damage in the isoform. Observed association in

this study may also possibly be explained by linkage disequilibrium (LD) of this allele with other

functional variants. For example, the rs2227309 in CASP7, completely linked with the

rs2227310 in several HapMap populations, was suggested to influence apoptotic capacity (Chae

et al., 2011; Garcia-Lozano et al., 2007). In the present study, rs13010627 at CASP10 was also

found to decrease risk of both leukoplakia and cancer compared to the controls. This SNP has

been previously associated with increased risk of CML and breast cancer (Frank et al., 2006;

Ghavami et al., 2009; Yang et al., 2008). Two separate European case-control studies reported

association of this SNP with breast cancer but a meta-analysis carried out by Breast Cancer

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

9

Consortium on over 30 case-control studies failed to corroborate above mentioned findings of

the two European studies. Same meta-analysis study also did not find any significant association

of rs8190315 in BID with breast cancer, though the Sheffield Breast Cancer study (a participant

in the meta-analysis) reported significant association between this SNP and risk of breast cancer

(Gaudet et al., 2009). In our study, rs8190315 in BID was found to significantly increase risk of

cancer compared to controls. This might suggest a population specific risk conferment by this

SNP. Others also studied SNPs at few of the apoptosis genes (say, BCL2, BID, CASP10 and

CASP7) in other cancers (Dorjgochoo et al., 2013; Guan et al., 2013; Meyer et al., 2013; Wu et

al., 2011) but the results were not significant or marginally significant. This will be first study to

report association of SNPs at BCL2, CASP1 and CASP10 with the risk of both leukoplakia and

oral cancer (Table 3) with respect to control. It is also interesting to note that SNPs at BCL2,

CASP1, CASP7 and CASP5 increased risk of cancer with respect to leukoplakia (data not

shown). So, our results showed that SNPs at BCL2 and CASP1 increased risk of cancer with

respect to both control and leukoplakia. It suggests that deregulation of mitochondrial apoptosis,

like the death receptor mediated apoptosis, is initiated at precancer stage through BCL2.

Six SNP loci that independently predicted risk of oral cancer also significantly increased

risk of cancer when their combined effect was considered. There was a steep increase in risk

with increase in “effective” number of risk alleles, and harboring of 5 or more “effective” risk

alleles elevated disease risk more than 130 folds (Figure 1). This suggests a multiplicative

interaction among risk alleles.

To fully understand role of these SNPs in the development of cancer, functional studies

need to be carried out, because SNPs chosen for this study were based on their functional

relevance to apoptosis. Expression of BID and CASP7 was observed to alter as genotypes of

individuals changed from wild to variant types as evidenced from published data set on HapMap

population. So, SNPs at BID and CASP7, that are present in coding region and displayed

significant association, might be surrogate markers for other “actual” causative loci since the

present study was based on tagSNPs. Moreover, our unpublished RNAseq data from 12 cancer

samples also revealed that expression of BID and CASP7 was significantly deregulated

(p=0.00006 and p=0.00003, respectively) in oral cancer tissues compared to adjacent normal

tissue.

In mitochondrial apoptotic pathway, different stimuli activate BH3-only family members

(i.e. initiators such as BID) and inhibit pro-survival BCL2 like proteins (i.e. guardians) and,

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

10

thereby, enabling activation of pro-apoptotic effectors (such as BAX and BAK), which disrupt

mitochondrial outer membrane to release cytochrome c. Thereafter, different caspases (such as

caspase 7) is activated by cytochrome c leading to apoptosis (Czabotar et al., 2014). In non-

mitochondrial (or death receptor mediated) apoptotic pathway, death receptor activates different

caspases to lead to apoptosis. So, our results imply that sequence variations of apoptotic genes

working within as well as outside mitochondria might have played important role to increase risk

of oral leukoplakia and cancer. Since, SNP at anti-apoptotic BCL2 modulated risk of both

leukoplakia and cancer and controls release of cytochrome c, so, this study suggests that

deregulation of mitochondrial apoptotic pathway, through BCL2, play important role in

progression of precancerous lesion to oral cancer.

5. Conclusion:

Overarching strengths of the study came from stringent selection criteria and rigorous

testing that have been employed throughout the study to minimize reporting of false positive

results. Analysis of combined effect of risk alleles at different loci strongly suggests joint effects

of multiple loci that can dynamically magnify disease risk. This study has provided valuable

insights that DNA sequence and expression variations in mitochondrial apoptotic pathway genes

may have played important roles to increase risk of oral pre-cancer and cancer.

Acknowledgements: Authors thank Prof. R. R. Paul and Prof. Jay Gopal Ray, oral

pathologists, for his active support during sample collection and diagnosis of patients. Grant

Supports: Indian Statistical Institute and Department of Biotechnology, India

Conflict of Interest: We declare that we have no conflicts of interest

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

11

REFERENCES

The International HapMap Project. 2003, Nature 426, 789-796. Abecasis, G.R., Auton, A., Brooks, L.D., DePristo, M.A., Durbin, R.M., Handsaker, R.E., Kang, H.M., Marth, G.T., McVean, G.A., 2012. An integrated map of genetic variation from 1,092 human genomes. Nature 491, 56-65. Benjamini, Y., Hochberg, Y., 1995. Controlling the False Discovery Rate: A Practical and Powerful Approach to Multiple Testing. Journal of the Royal Statistical Society. Series B (Methodological) 57, 289-300. Chae, Y.S., Kim, J.G., Sohn, S.K., Lee, S.J., Kang, B.W., Moon, J.H., Park, J.Y., Jeon, S.W., Bae, H.I., Choi, G.S., Jun, S.H., 2011. RIPK1 and CASP7 polymorphism as prognostic markers for survival in patients with colorectal cancer after complete resection. J Cancer Res Clin Oncol 137, 705-713. Chen, K., Hu, Z., Wang, L.E., Sturgis, E.M., El-Naggar, A.K., Zhang, W., Wei, Q., 2007. Single-nucleotide polymorphisms at the TP53-binding or responsive promoter regions of BAX and BCL2 genes and risk of squamous cell carcinoma of the head and neck. Carcinogenesis 28, 2008-2012. Czabotar, P.E., Lessene, G., Strasser, A., Adams, J.M., 2014. Control of apoptosis by the BCL-2 protein family: implications for physiology and therapy. Nat Rev Mol Cell Biol 15, 49-63. Datta, S., Majumder, M., Biswas, N.K., Sikdar, N., Roy, B., 2007. Increased risk of oral cancer in relation to common Indian mitochondrial polymorphisms and Autosomal GSTP1 locus. Cancer 110, 1991-1999. Dorjgochoo, T., Xiang, Y.B., Long, J., Shi, J., Deming, S., Xu, W.H., Cai, H., Cheng, J., Cai, Q., Zheng, W., Shu, X.O., 2013. Association of genetic markers in the BCL-2 family of apoptosis-related genes with endometrial cancer risk in a Chinese population. PLoS One 8, e60915. Enjuanes, A., Benavente, Y., Bosch, F., Martin-Guerrero, I., Colomer, D., Perez-Alvarez, S., Reina, O., Ardanaz, M.T., Jares, P., Garcia-Orad, A., Pujana, M.A., Montserrat, E., de Sanjose, S., Campo, E., 2008. Genetic variants in apoptosis and immunoregulation-related genes are associated with risk of chronic lymphocytic leukemia. Cancer Res 68, 10178-10186. Ferlay, J., Shin, H.-R., Bray, F., Forman, D., Mathers, C., Parkin, D.M., 2010. Estimates of worldwide burden of cancer in 2008: GLOBOCAN 2008. International Journal of Cancer 127, 2893-2917. Frank, B., Hemminki, K., Wappenschmidt, B., Meindl, A., Klaes, R., Schmutzler, R.K., Bugert, P., Untch, M., Bartram, C.R., Burwinkel, B., 2006. Association of the CASP10 V410I variant with reduced familial breast cancer risk and interaction with the CASP8 D302H variant. Carcinogenesis 27, 606-609. Garcia-Lozano, J.R., Torres, B., Fernandez, O., Orozco, G., Alvarez-Marquez, A., Garcia, A., Gonzalez-Gay, M.A., Nunez-Roldan, A., Martin, J., Gonzalez-Escribano, M.F., 2007. Caspase 7 influences susceptibility to rheumatoid arthritis. Rheumatology (Oxford) 46, 1243-1247. Gaudet, M.M., Milne, R.L., Cox, A., Camp, N.J., Goode, E.L., Humphreys, M.K., Dunning, A.M., Morrison, J., Giles, G.G., Severi, G., Baglietto, L., English, D.R., Couch, F.J., Olson, J.E., Wang, X., Chang-Claude, J., Flesch-Janys, D., Abbas, S., Salazar, R., Mannermaa, A., Kataja, V., Kosma, V.M., Lindblom, A., Margolin, S., Heikkinen, T., Kampjarvi, K., Aaltonen, K., Nevanlinna, H., Bogdanova, N., Coinac, I., Schurmann, P.,

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

12

Dork, T., Bartram, C.R., Schmutzler, R.K., Tchatchou, S., Burwinkel, B., Brauch, H., Torres, D., Hamann, U., Justenhoven, C., Ribas, G., Arias, J.I., Benitez, J., Bojesen, S.E., Nordestgaard, B.G., Flyger, H.L., Peto, J., Fletcher, O., Johnson, N., Dos Santos Silva, I., Fasching, P.A., Beckmann, M.W., Strick, R., Ekici, A.B., Broeks, A., Schmidt, M.K., van Leeuwen, F.E., Van't Veer, L.J., Southey, M.C., Hopper, J.L., Apicella, C., Haiman, C.A., Henderson, B.E., Le Marchand, L., Kolonel, L.N., Kristensen, V., Grenaker Alnaes, G., Hunter, D.J., Kraft, P., Cox, D.G., Hankinson, S.E., Seynaeve, C., Vreeswijk, M.P., Tollenaar, R.A., Devilee, P., Chanock, S., Lissowska, J., Brinton, L., Peplonska, B., Czene, K., Hall, P., Li, Y., Liu, J., Balasubramanian, S., Rafii, S., Reed, M.W., Pooley, K.A., Conroy, D., Baynes, C., Kang, D., Yoo, K.Y., Noh, D.Y., Ahn, S.H., Shen, C.Y., Wang, H.C., Yu, J.C., Wu, P.E., Anton-Culver, H., Ziogoas, A., Egan, K., Newcomb, P., Titus-Ernstoff, L., Trentham Dietz, A., Sigurdson, A.J., Alexander, B.H., Bhatti, P., Allen-Brady, K., Cannon-Albright, L.A., Wong, J., Chenevix-Trench, G., Spurdle, A.B., Beesley, J., Pharoah, P.D., Easton, D.F., Garcia-Closas, M., 2009. Five polymorphisms and breast cancer risk: results from the Breast Cancer Association Consortium. Cancer Epidemiol Biomarkers Prev 18, 1610-1616. Ghavami, S., Hashemi, M., Ande, S.R., Yeganeh, B., Xiao, W., Eshraghi, M., Bus, C.J., Kadkhoda, K., Wiechec, E., Halayko, A.J., Los, M., 2009. Apoptosis and cancer: mutations within caspase genes. J Med Genet 46, 497-510. Guan, X., Liu, Z., Liu, H., Yu, H., Wang, L.E., Sturgis, E.M., Li, G., Wei, Q., 2013. A functional variant at the miR-885-5p binding site of CASP3 confers risk of both index and second primary malignancies in patients with head and neck cancer. FASEB J 27, 1404-1412. Gupta, P.C., Bhonsle, R.B., Murti, P.R., Daftary, D.K., Mehta, F.S., Pindborg, J.J., 1989. An epidemiologic assessment of cancer risk in oral precancerous lesions in India with special reference to nodular leukoplakia. Cancer 63, 2247-2252. Hanahan, D., Weinberg, R.A., 2000. The hallmarks of cancer. Cell 100, 57-70. Hopkins, J., Cescon, D.W., Tse, D., Bradbury, P., Xu, W., Ma, C., Wheatley-Price, P., Waldron, J., Goldstein, D., Meyer, F., Bairati, I., Liu, G., 2008. Genetic polymorphisms and head and neck cancer outcomes: a review. Cancer Epidemiol Biomarkers Prev 17, 490-499. Kim, D.H., Xu, W., Ma, C., Liu, X., Siminovitch, K., Messner, H.A., Lipton, J.H., 2009. Genetic variants in the candidate genes of the apoptosis pathway and susceptibility to chronic myeloid leukemia. Blood 113, 2517-2525. Lappalainen, T., Sammeth, M., Friedlander, M.R., t Hoen, P.A., Monlong, J., Rivas, M.A., Gonzalez-Porta, M., Kurbatova, N., Griebel, T., Ferreira, P.G., Barann, M., Wieland, T., Greger, L., van Iterson, M., Almlof, J., Ribeca, P., Pulyakhina, I., Esser, D., Giger, T., Tikhonov, A., Sultan, M., Bertier, G., MacArthur, D.G., Lek, M., Lizano, E., Buermans, H.P., Padioleau, I., Schwarzmayr, T., Karlberg, O., Ongen, H., Kilpinen, H., Beltran, S., Gut, M., Kahlem, K., Amstislavskiy, V., Stegle, O., Pirinen, M., Montgomery, S.B., Donnelly, P., McCarthy, M.I., Flicek, P., Strom, T.M., Lehrach, H., Schreiber, S., Sudbrak, R., Carracedo, A., Antonarakis, S.E., Hasler, R., Syvanen, A.C., van Ommen, G.J., Brazma, A., Meitinger, T., Rosenstiel, P., Guigo, R., Gut, I.G., Estivill, X., Dermitzakis, E.T., 2013. Transcriptome and genome sequencing uncovers functional variation in humans. Nature 501, 506-511. Lee, W.K., Kim, J.S., Kang, H.G., Cha, S.I., Kim, D.S., Hyun, D.S., Kam, S., Kim, C.H., Jung, T.H., Park, J.Y., 2009. Polymorphisms in the Caspase7 gene and the risk of lung cancer. Lung Cancer 65, 19-24.

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

13

Meyer, A., Coinac, I., Bogdanova, N., Dubrowinskaja, N., Turmanov, N., Haubold, S., Schurmann, P., Imkamp, F., von Klot, C., Merseburger, A.S., Machtens, S., Bremer, M., Hillemanns, P., Kuczyk, M.A., Karstens, J.H., Serth, J., Dork, T., 2013. Apoptosis gene polymorphisms and risk of prostate cancer: a hospital-based study of German patients treated with brachytherapy. Urol Oncol 31, 74-81. Proia, N.K., Paszkiewicz, G.M., Nasca, M.A., Franke, G.E., Pauly, J.L., 2006. Smoking and smokeless tobacco-associated human buccal cell mutations and their association with oral cancer--a review. Cancer Epidemiol Biomarkers Prev 15, 1061-1077. Purcell, S., Neale, B., Todd-Brown, K., Thomas, L., Ferreira, M.A., Bender, D., Maller, J., Sklar, P., de Bakker, P.I., Daly, M.J., Sham, P.C., 2007. PLINK: a tool set for whole-genome association and population-based linkage analyses. Am J Hum Genet 81, 559-575. Reibel, J., 2003. Prognosis of oral pre-malignant lesions: significance of clinical, histopathological, and molecular biological characteristics. Crit Rev Oral Biol Med 14, 47-62. Scully, C., Bagan, J., 2009. Oral squamous cell carcinoma: overview of current understanding of aetiopathogenesis and clinical implications. Oral Dis 15, 388-399. Shi, T.Y., He, J., Wang, M.Y., Zhu, M.L., Yu, K.D., Shao, Z.M., Sun, M.H., Wu, X., Cheng, X., Wei, Q., 2015. CASP7 variants modify susceptibility to cervical cancer in Chinese women. Sci Rep 5, 9225. Stranger, B.E., Forrest, M.S., Dunning, M., Ingle, C.E., Beazley, C., Thorne, N., Redon, R., Bird, C.P., de Grassi, A., Lee, C., Tyler-Smith, C., Carter, N., Scherer, S.W., Tavare, S., Deloukas, P., Hurles, M.E., Dermitzakis, E.T., 2007. Relative impact of nucleotide and copy number variation on gene expression phenotypes. Science 315, 848-853. Sun, T., Gao, Y., Tan, W., Ma, S., Shi, Y., Yao, J., Guo, Y., Yang, M., Zhang, X., Zhang, Q., Zeng, C., Lin, D., 2007. A six-nucleotide insertion-deletion polymorphism in the CASP8 promoter is associated with susceptibility to multiple cancers. Nat Genet 39, 605-613. Wu, I.C., Zhao, Y., Zhai, R., Liu, C.Y., Chen, F., Ter-Minassian, M., Asomaning, K., Su, L., Heist, R.S., Kulke, M.H., Liu, G., Christiani, D.C., 2011. Interactions between genetic polymorphisms in the apoptotic pathway and environmental factors on esophageal adenocarcinoma risk. Carcinogenesis 32, 502-506. Yang, M., Sun, T., Wang, L., Yu, D., Zhang, X., Miao, X., Liu, J., Zhao, D., Li, H., Tan, W., Lin, D., 2008. Functional variants in cell death pathway genes and risk of pancreatic cancer. Clin Cancer Res 14, 3230-3236. Yoo, S.S., Choi, J.E., Lee, W.K., Choi, Y.Y., Kam, S., Kim, M.J., Jeon, H.S., Lee, E.B., Kim, D.S., Lee, M.H., Kim, I.S., Jheon, S., Park, J.Y., 2009. Polymorphisms in the CASPASE genes and survival in patients with early-stage non-small-cell lung cancer. J Clin Oncol 27, 5823-5829. Zhivotovsky, B., Orrenius, S., 2006. Carcinogenesis and apoptosis: paradigms and paradoxes. Carcinogenesis 27, 1939-1945.

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

14

Table 1- Genes included in the study and distribution of SNPs in different

functional regions Gene

(n=23)

Chromosome No. of

SNPs

(n=99)

Exonic non-

synonymous

(n=28)

5' UTR/near

gene

(n=26)

3' UTR/near

gene

(n=45)

BAX 19 2 1 1

BCL10 1 4 1 3

BCL2 18 6 2 4

BCL2L2 14 4 1 3

BID 22 5 1 4

BIRC4 X 2 1 1

BIRC5 17 8 1 2 5

CASP1 11 4 1 1 2

CASP3 4 3 1 1 1

CASP10 2 5 3 1 1

CASP5 11 8 4 2 2

CASP6 4 3 2 1

CASP8 2 6 1 2 3

CASP7 10 6 2 3 1

CASP9 1 1 1

FAS 10 8 3 5

FASLG 1 4 1 2 1

HTRA2 2 1 1

NFKB1 4 3 2 1

PARP1 1 4 2 2

TNF 6 6 1 1 4

TNFRSF1A 12 4 1 2 1

TNFRSF10A 8 2 2

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

15

Table 2: Description of patients and controls

a Exclusive smokers have tobacco smoking habit only;

b Pack Year

c Exclusive chewers have tobacco chewing/dipping habit only,

d Chew Year

e Mixed habitués have both tobacco smoking and chewing habits.

Demography and

Tobacco Habits

Control

[N=538] (%)

Leukoplakia

[N=253] (%)

p-value

Cases

[N=525] (%)

p-value

Male

405(75) 221(87)

<0.001

345 (66)

0.001 Female

133 (25) 32(13) 180 (34)

Age (Mean±SD) 48.6±12.1 47.8±10.5 0.36 54.1±11 <0.001

aExclusive smokers

193(36)

143(57)

<0.001

107(20)

<0.001

b

Mean PY±SD 22.8±20.8 19.5±16 0.11 25.7±18 0.28

cExclusive chewers

223(41) 38(15) <0.001 284(54) <0.001

d

Mean CY±SD 197±192 168±163 0.38 222±256 0.22

e Mixed habitués

122(23) 72(28) 0.11 134(26) 0.36

Mean PY±SD 19±22 22±17 0.32 21±25 0..49

Mean CY±SD 141±186 141±263 1.0 125±129 0.42

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

16

Table 3- Comparison of allele frequencies among cases and controls and risk of the disease

*P-values are age, sex, tobacco-dose adjusted and FDR corrected from allele-based logistic regression tests

**MAF- Minor Allele Frequency, NS: non-significant, NA: genotypes were not in HWE

Initially, 538 controls, 253 leukoplakia and 525 cancer patients were genotyped. After data pruning, 532 controls, 250 leukoplakia, and 499

cancer patients passed the inclusion criteria and were included in the association analysis

Genotypes at BCL2, BCL2L2, BID, CASP1, CASP7 and CASP10 are G>C, G>A, T>C, T>C, C>G and G>A, respectively

SNP ID

(Minor allele/

Other allele)

Gene Chromosome Functional

Location

Risk

allele

MAF **

Control

MAF **

Leukoplakia

MAF**

Cancer

Control-

Leukoplakia

(P-Value;

OR, 95%CI)*

Control-

Cancer

(P-Value;

OR, 95%CI)*

Previous

reports of

association

rs1473418

(C/G) BCL2 18 5'UTR C 0.004 0.062 0.018

1.39E-05;

20.55,

(6.59-64.02)

0.04; 5.06,

(1.59-16.15)

Novel

rs1950252

(A/G) BCL2L2 14 3'UTR A 0.033 0.019 0.008 NS

0.03;

0.29,

(0.12-0.68)

Novel

rs8190315

(C/T) BID 22 Ser9Gly C 0.011 0.014 0.042 NS

0.001;

3.95,

(2.02-7.76)

Reported

rs511044

(C/T) CASP1 11

Flanking

3'UTR C 0.0009 0.106 0.024

0.0001;

291.5,

(26.65-318.8)

0.009;

35.74,

(4.61-277.1)

Novel

rs2227310

(G/C) CASP7 10 Asp254Glu G 0.018 0.24 0.240 NA

6E-13;

8.85,

(5.11-15.34)

Reported

rs13010627

(A/G) CASP10 2 Val366Ile A 0.085 0.006 0.011

0.0007;

0.05,

(0.01-0.2)

1.9E-07;

0.12,

(0.06-0.24)

Reported

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

17

Figure1: Increase in risk of oral cancer with increase in "effective" number of risk

alleles

Legend to Figure 1: “Effective” number of risk alleles and the effect size (i.e. OR) of the

respective SNPs was calculated based on the number of risk alleles harbored by an individual

(see Methods). The parentheses in boxes represent Odds Ratios (OR) and 95%CI. Genotypes

at 6 SNPs (i.e. BCL2, BCL2L2, BID, CASP1, CASP7 and CASP10) are G>C, G>A, T>C, T>C,

C>G and G>A respectively and minor alleles are risk alleles.

0

20

40

60

80

100

120

140

0 to3 (reference) 3 to4 4 to 5 5 and above

Od

ds

Rat

io (

OR

)

"Effective" number of risk alleles

OR= 136.53,

(54.51-341.94)

OR= 39.29,

(12.56-122.92)

OR= 6.02,

(2.72, 13.32)

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

18

Highlights:

1. DNA sequence variation in apoptotic pathway genes

2. SNPs at these genes modify risk of oral precancer and cancer

3. SNPs at mitochondrial apoptotic pathway genes may be important for progression of leukoplakia to

cancer

4. Combination of risk alleles magnify risk of cancer many folds.