Tumor Biology

, Volume 34, Issue 1, pp 521–529

Lack of mutational events of RAS genes in sporadic thyroid cancer but high risk associated with HRAS T81C single nucleotide polymorphism (case–control study)

Authors

  • Mosin S. Khan
    • Department of Clinical BiochemistrySher-I-Kashmir Institute of Medical Sciences
  • Arshad A. Pandith
    • Advanced Centre for Human GeneticsSher-I-Kashmir Institute of Medical Sciences
  • Mahboob ul Hussain
    • Department of BiotechnologyUniversity of Kashmir
  • Mohammad Iqbal
    • Department of Clinical BiochemistrySher-I-Kashmir Institute of Medical Sciences
  • Nighat P. Khan
    • Department of Clinical BiochemistrySher-I-Kashmir Institute of Medical Sciences
  • Khurshid A. Wani
    • Department of General SurgerySher-I-Kashmir Institute of Medical Sciences
  • Shariq R. Masoodi
    • Department of EndocrinologySher-I-Kashmir Institute of Medical Sciences
    • Department of Clinical BiochemistrySher-I-Kashmir Institute of Medical Sciences
Research Article

DOI: 10.1007/s13277-012-0577-y

Cite this article as:
Khan, M.S., Pandith, A.A., ul Hussain, M. et al. Tumor Biol. (2013) 34: 521. doi:10.1007/s13277-012-0577-y

Abstract

High incidence of thyroid cancer worldwide indicates the importance of studying genetic alterations that lead to its carcinogenesis. Specific acquired RAS mutations have been found to predominate in different cancers, and HRAS T81C polymorphism has been determined to contribute the risk of various cancers, including thyroid cancer. We screened the exons 1 and 2 of RAS genes (HRAS, KRAS, and NRAS) in 60 consecutive thyroid tissue (tumor and adjacent normal) samples, and a case–control study was also conducted for HRAS T81C polymorphism in HRAS codon 27 using the polymerase chain reaction-restriction fragment length polymorphism to test the genotype distribution of 140 thyroid cancer patients in comparison with 170 cancer-free controls from a Kashmiri population. No mutation was found in any of the thyroid tumor tissue samples, but we frequently detected polymorphism at nucleotide 81 (T > C) in exon 1 of HRAS gene. In HRAS T81C SNP, frequencies of TT, TC, and CC genotypes among cases were 41.4, 38.6, and 20.0 %, while in controls genotype frequencies were 84.1, 11.7, and 4.2 %, respectively. A significant difference was observed in variant allele frequencies (TC + CC) between the cases and controls (58.6 vs. 16 %) with odds ratio = 7.4; confidence interval (CI) = 4.3–12.7 (P < 0.05). Interestingly, combined TC and CC genotype abundantly presented in follicular thyroid tumor (P < 0.05). Moreover, a significant association of the variant allele (TC + CC) was found with nonsmokers (P < 0.05). This study shows that although thyroid cancer is highly prevalent in this region, the mutational events for RAS genes do not seem to be involved. Contrary to this HRAS T81C SNP of HRAS gene moderately increases thyroid cancer risk with rare allele as a predictive marker for follicular tumors.

Keywords

Papillary thyroid cancerKashmiri populationRestriction fragment length polymorphismSingle strand conformational polymorphismBenign thyroid disease

Introduction

Thyroid carcinoma is the most prevalent endocrine malignancy. Since the 1990s, thyroid cancer has become the fastest increasing cancer in women [1]. Thyroid cancer remains the most common endocrine malignancy and accounts for 2.5 % of all human cancers [2]. From 1973 to 2006, there has been a 2.6-fold increase in the incidence of thyroid carcinomas [3]. Thyroid nodules can be hyperplastic benign adenomas or malignant lesions and can be derived from thyroid follicular epithelial cells or C cells [4]. Approximately 90 % of thyroid malignancies are well-differentiated thyroid carcinomas, which are classified as papillary or follicular based on histopathological criteria. Even though differentiated thyroid carcinomas are usually curable by the combination of surgery, radioiodine ablation, and thyroid-stimulating hormone (TSH) suppressive therapy, recurrence occurs in 20–40 % of patients [5, 6]. Four mutation types, that is, BRAF, RAS point mutation, RET/PTC, and PAX8/peroxisome proliferator-activated receptor g (PPARg) rearrangements constitute the majority of mutations known to occur in the two most common types of thyroid cancer, papillary and follicular carcinoma [710]. Point mutations of the RAS gene are not restricted to a particular type of thyroid tumors and are found in follicular carcinomas, papillary carcinomas, and follicular adenomas. Nevertheless, RAS mutations have also been found in MTC, which arise from thyroid parafollicular C-cells, though with discordant results. In particular, point mutations in the RAS genes have been previously studied in small series of sporadic MTC and were either not detected [1113] or reported in only a few cases [14, 15]. Ciampi et al. [16] did not show any mutations in any of the RAS genes in germline and somatic RET-positive MTC cases, although they reported 17.6 % mutations in the RAS genes (HRAS 12.1 %, KRAS 3.7 %, NRAS 1.8 %) in RET-negative MTC. Margarida et al. [17] reported 2.5 % RAS mutations in RET-positive sporadic MTC, but no RAS gene mutation was found in hereditary MTC.

The three human RAS genes (HRAS, KRAS, and NRAS) encode highly related G-proteins that propagate signals arising from cell membrane receptors to various intracellular targets. Point mutations in the specific domains of the RAS gene either increase its affinity for GTP (mutations in codons 12 and 13) or inactivate its autocatalytic GTPase function (mutation in codon 61), resulting in permanent RAS activation and chronic stimulation of its downstream targets along the MAPK and PI3K/AKT signaling pathways [18]. Frequent RAS mutations have been reported in a number of human cancers, including adenocarcinoma of the pancreas (90 %), colon (50 %), and lung (30 %) [19]. In thyroid cancer, frequency of RAS gene mutations is reported to be around 50 % [19]. The HRAS gene is commonly activated in human urinary tract tumors [20]. A number of mutations were described in this gene that is implicated in the development of various type of cancer. HRAS gene can participate in carcinogenesis in two ways: first if the protein function is modified and second if its expression is enhanced [19, 21, 22]. Taparowsky et al. [23] described a point mutation that is thought to be involved in generating cancer, a single nucleotide polymorphism in codon 27 in the first exon (SNP 81T>C, rs12628) of HRAS gene. Silent mutation at position 81 in exon 1 has no effect on p21 structure and function [23, 24] and therefore is not significant for carcinogenesis [25]. As opposed to this conclusion, it was reported that HRAS T81C SNP can indicate an increased risk of skin [26], oral [27], bladder [28], and gastric cancer [29]. Castro et al. [30] has first time investigated HRAS T81C polymorphism in thyroid tumors and has associated it with aneuploidy in follicular tumors of the thyroid. Although HRAS T81C polymorphism does not alter the amino acid sequence of the protein, it may affect the expression of the gene inducing overexpression [30]. It is possible that it is linked to another polymorphic locus inside regulatory intronic region. However, the number of studies conducted to examine the HRAS T81C polymorphism is not sufficient; moreover, the results of them are controversial yet [29].

Keeping in view the role played by RAS gene alterations in various cancers, the aim of our study was to assess the contribution of mutations of this gene in thyroid cancer. While sequencing of RAS gene family, there was high frequency of HRAS 81T>C variation in tumor tissues so it was considered as an informative SNP in a gene with a potential link to carcinogenesis. Thus, we hypothesize that the HRAS T81C polymorphism may play a role in HRAS functional activity, which may play a role in modulating the susceptibility to thyroid cancer. In order to verify our hypothesis, a population-based case–control study was also conducted to investigate the association between the HRAS T81C genotypes and the risk of thyroid cancer in Kashmiri population.

Methodology

Study subjects

Sixty patients who underwent total thyroidectomy/hemi-thyroidectomy or lobectomy in the Department of General Surgery, Sher-I-Kashmir Institute of Medical Sciences (SKIMS) between 2010 and 2012 were included in this study for sequence analysis of RAS gene family. Among the cases 48 (80 %) were females and 12 (20 %) were males. Also, 36 (60 %) of them were <45 years of age and 24 (40 %) of ≥45. The number of smokers among cases was six (10 %) and non-smokers 54 (90 %). TSH levels were elevated in 25 (41.6 %) patients and normal in 35 (58.4 %) patients. Benign thyroid diseases were found in 48 of 60 (80 %) patients. Nodal metastasis and vascular/capsular invasion was positive in 15 (25 %) and 26 (43 %) thyroid cancer patients, respectively. Well-differentiated cancer grade was present in 57 of 60 (95 %) thyroid cancer patients The patients <45 years of age having stage I and stage II disease were 34 (56.6 %) and two (3.4 %), respectively; similarly, patients who were ≥45 years of age having stage I/II and stage III/above were 15 (25 %) and nine (15 %), respectively. Almost all the samples were histologically confirmed as differentiated thyroid carcinomas (papillary—42 (70 %) and follicular—eight (13.4 %), respectively) except few cases of medullary thyroid carcinoma—four (6.6 %) and Hurthle cell carcinoma—six (13.4 %).

A total of 140 peripheral blood samples from thyroid cancer patients were collected from the department of Nuclear Medicine SKIMS between October 2010 and August 2011. Also 170 blood samples were collected from control subjects who were not having any sort of malignancy from the same hospital and belonging to the same geographical area, ethnic background. The controls were healthy, free from any cancer. A written informed consent from each recruited subject was sought, and the study was approved by the institutional ethical committee. Sample size for this study was calculated by statistical software (G POWER 3.0.10). The power of study (1 − β) was 90 % (0.9), and allocation ratio (controls/cases) was taken as 1.2 for this case–control study.

DNA extraction

Tissue samples (both tumor and adjacent normal tissue) collected after thyroid surgery were snap-frozen immediately and stored at −80 °C. DNA from tissues was extracted using Quick gDNA MiniPrep (Zymo Research Corporation, 17062 Murphy Ave, Irvine, CA 92614, USA) according to given protocol for sequencing analysis of RAS gene family (NRAS, HRAS, KRAS). For HRAS T81C genotyping, the genomic DNA was extracted from peripheral blood samples (n ± 300, controls and cases) using standard proteinase-K digestion, salting out by ammonium acetate and ethanol precipitation method.

Mutational analysis of RAS gene family

Exons 1 and 2 of each of the NRAS, HRAS, and KRAS were amplified using the specific primers [3135], given in Table 1. Polymerase chain reaction (PCR) amplification was carried out in a 50-μL volume containing 50 ng of genomic DNA; 1× PCR buffer containing 2 mM MgCl2; 200 μM each of dATP, dGTP, dTTP, and dCTP; 1.5 U of Taq DNA polymerase (Biotools, Valle de Tobalina, 52-Nave 39 28021, Madrid, Spain); and 0.2 pM of forward and reverse primers (Sigma-Aldrich 3050 Spruce St., St. Louis, MO 63103, USA). After an initial denaturation at 95 °C for 7 min, 35 cycles of 94 °C for 35 s, a specific annealing temperature (HRAS and NRAS exon 1, 60 °C; KRAS exons 1 and 2, NRAS exon 2, and HRAS exon 2, 55 °C) for 35 s, and 72 °C for 35 s were performed. Final elongation was given at 72 °C for 7 min. PCR products were run on 2 % agarose gel and analyzed under a UV illuminator. For quality control, distilled water was used instead of DNA as a negative control. Single-strand conformation polymorphism (SSCP) analysis of the amplicons of exons 1 and 2 of NRAS, HRAS, and KRAS was performed on 6 % nondenaturing polyacrylamide gel utilizing a non-radioactive silver staining procedure [36]. The purified PCR amplicons of the tumor samples showing mobility shift on SSCP analysis were confirmed by direct sequencing. In order to rule out germ line mutation, DNA from the blood samples of the patients showing mutations were subjected to direct DNA sequencing, using the automated DNA sequencer ABI prism 310.
Table 1

Primer sequences, annealing temperatures, and product size of exons for direct sequencing and RFLP

Gene

Primer sequence

Tm (°C)

Product size (bp)

NRAS1

1F 5′-AGTACTGTAGATGTGGCTCGCC-3′

60

185

1R 5′-CCTCACCTCTATGGTGGGATC-3′

NRAS2

2F 5′-CCCCTTACCCTCCACAC-3′

55

196

2R 5′-AGGTTAATATCCGCAAATGAC-3′

HRAS1

1F 5′-CAGGAGACCCTGTAGGAGGA-3′

60

186

1R 5′-GGCACCTGGACGGCGGCGCTAG-3′

HRAS2

2F 5′-TCCTGCAGGATTCCTACCGG-3′

55

194

2R 5′-GGTTCACCTGTACTGGTGGA-3′

KRAS1

1F 5′-GTACTGGTGGAGTATTIGAT-3′

55

285

1R 5′-TGAAAATGGTCAGAGAAACC-3′

KRAS2

1F 5′-CCTTCTCAGGATTCCTACAG-3′

55

155

1R 5′-TTATTTATGGCAAATACACAAATA-3′

H-RAS T81C genotyping

PCR-restriction fragment length polymorphism (RFLP) assay was performed for genotyping of HRAS T81C. PCR was performed to amplify a 186-bp DNA segment encompassing HRAS T81C SNP using above-mentioned forward and reverse primers. PCR was carried out in a final volume of 25 μL containing 50 ng genomic DNA template, 1× PCR buffer with 2 mM MgCl2, 200 μM dNTPs (Biotools, Valle de Tobalina, 52-Nave 39 28021, Madrid, Spain), 0.2 pM of each primer, and 0.75 U Taq polymerase (Sigma-Aldrich 3050 Spruce St., St. Louis, MO 63103, USA). For PCR amplification, the standard protocol was used as discussed previously for exon 1 of HRAS gene. For RFLP, the PCR products (186 bp) were digested with DraIII (Thermo Scientific, 3747 N. Meridian Rd., Rockford, IL 61101, USA) (3 U at 37 °C for 16 h). For codon 81 of HRAS, the CC homozygote (variant) with DraIII restriction site was cut into fragments of 128 and 58 bp, while the TT homozygote (wild) presented a single fragment of 186 bp and heterozygous (T/C) form displayed 186, 128, and 58 bp fragments (Fig. 1). DNA fragments were subjected to electrophoresis on a 3 % agarose gel for resolution, and samples were also run on 8 % polyacrylamide gel to confirm results.
https://static-content.springer.com/image/art%3A10.1007%2Fs13277-012-0577-y/MediaObjects/13277_2012_577_Fig1_HTML.gif
Fig. 1

Digestion of PCR product by Dra III. TT allele (186 bp) shown in lane 1; the TC heterozygous (186 and 128 bp) in lanes 4 and 6; and homozygous CC mutant allele (128 and 58 bp) in lanes 2, 3, 5, 7, 8, 9 and M; 100 bp DNA ladder

Statistical analysis

χ2 test was used to compare cases and controls for categorical variables like age, sex, type, TSH levels, smoking status, etc. A goodness-of-fit χ2 test was used to determine whether the polymorphisms were in Hardy–Weinberg equilibrium between cases and controls. Odds ratios (ORs) with 95 % confidence intervals (CIs) were used as estimates of the relative risk or degree of association between certain genotypes or other related risk factors of thyroid cancer. P value was calculated by Pearson’s method. A P value of less than 0.05 indicates that the SNP is real or statistically significant. P value of greater than 0.05 indicates that the SNP is mere by chance or statistically insignificant. Univariate analysis was performed by Minitab software. Multivariate approach using binary logistic regression analysis between the statistically significant parameters reported in Table 3 was used. The genotypes were included as an independent variable and analyzed between the cases and the controls together with the others to really assess which parameters really influence the cancer risk. The multivariate analysis was performed by SPSS 16.0 software.

Results

Exons 1 and 2 each of NRAS, HRAS, and KRAS genes were screened for mutations in 60 tissue samples of thyroid cancer cases. Total of six exons of RAS gene family were screened for mutations especially in codons 12, 13, and 61. No mutations were observed in any of the six exons studied, particularly in codons 12, 13, and 61. However, DNA sequencing of HRAS exon 1 showed frequent T to C substitution in codon 27 of exon 1 at cDNA position 81, which is located in a wobble base position. The substitution (T81C) in codon 27 was found in 16 of 60 (26.6 %) tumor tissue samples. HRAS 81T>C substitution was found in 12 of 42 (28.5 %) papillary thyroid cancer tissues and four of eight (50 %) follicular thyroid cancer tissues. For polymorphic analysis of HRAS 81T>C SNP, a total of 140 thyroid cancer cases and 170 cancer-free healthy controls were taken. The cases included 26 (19 %) males and 114 (81 %) female patients (1:4.4), and the controls consisted of 140 (82.4 %) males and 30 (17.6 %) females. Of the total number of cases, 124 (89 %) were non-smokers and 16 (11 %) were smokers; 40 (29 %) patients were above 45 years of age, and 100 (71 %) patients were below 45 years of age (Table 2). The subjects were considered never smokers only if until the day of sample collection they had not consumed tobacco, and subjects were considered smokers if they are smoking presently or had quit smoking since last 6 months or less before sample collection. Table 2 shows demographic information and other parameter of cases and controls whereas the distribution of HRAS T81C genotypes, and its allele frequency in cases and controls are shown in Tables 3 and 4. Due to the very low frequency of the “CC” genotype and an increased risk associated with TC and CC genotypes, TC + CC was compared against TT. Frequencies of TT, TC, and CC genotypes among cases were 41.4, 38.6, and 20 %, while in controls were 84.1, 11.7, and 4.2 %, respectively, with OR of 7.4; 95 % CI = 4.3–12.7. The cases had a higher frequency of the rare allele (TC + CC) (58.6 %) than the controls (15.9 %), and this pattern of distribution of rare alleles among two groups showed statistical significance (P < 0.05). This finding shows an increased risk with TC + CC combination of genotypes against TT genotype (Table 3). The frequency of mutant C allele was 39.3 % in cases and 10 % in controls. This observation showed a highly statistical significance of rare allele (C) between cases and controls (P < 0.05) with an odds ratio of 5.8 (CI = 3.7–8.7). When classified further into groups, our study interestingly found higher number of rare allele (TC + CC) in rural dwellers (64 %) than urban dwellers (36 %) having P < 0.05. Non-smokers were more significantly associated with combined TC and CC against TT in cases than controls (adjusted OR 6.4; 95 % CI = 2.4–16.4; P < 0.05) while as ever smokers showed insignificant association (P > 0.05). Males have high percentage of rare allele (TC + CC) in cases as compared to controls (adjusted OR 11.5; 95 % CI = 3.6–36.9), and the association is statistically significant (P < 0.05). Association of variant allele with other clinicopathological characteristics is given in Table 3. While age, dwelling, gender, smoking status (except ever smokers, P > 0.05), and genotype (TC + CC) were associated with thyroid cancer in odds adjusted univariate analysis, the same parameters were associated with this disease in multivariate logistic regression analysis (Table 5).
Table 2

Frequency distribution analysis of selected demographic and risk factors in thyroid cancer cases and controls

Characteristics

Cases

Controls

χ2 value

P value

n = 140 (%)

n = 170 (%)

Age group

 <45

100 (71)

60 (35)

40.14

<0.0001

 ≥45

40 (29)

110 (65)

Sex

 Female

114 (81)

30 (17.6)

125.56

<0.0001

 Male

26 (19)

140 (82.4)

Dwelling

 Rural

112 (80)

50 (29.4)

78.75

<0.0001

 Urban

28 (20)

120 (70.6)

Smoking

 Never

124 (89)

50 (29.4)

109.12

<0.0001

 Ever

16 (11)

120 (70.6)

Benign thyroid disease

 Yes

84 (60)

   

 No

56 (40)

TSH levels

 Elevated

100 (71)

   

 Normal

40 (29)

Histological types

 Papillary

118 (84)

   

 Follicular

22 (16)

Tumor Grade

 WD

134 (96)

   

 PD

6 (4)

Stage, <45 years

 Stage I

94 (67)

   

 Stage II

6 (4.3)

Stage, ≥45 years

 Stages I and II

36 (25.7)

   

 Stage III and above

4 (3)

Vascular/capsular invasion

 Yes

66 (47)

   

 No

74 (53)

Lymph node metastasis

 Yes

52 (37)

   

 No

88 (63)

TSH thyroid-stimulating hormone, WD well differentiated, PD poorly differentiated

Table 3

Association between HRAS T81C phenotypes and clinicopathologic characteristics

 

Cases

TT

TC+CC

Controls

TT

TC + CC

OR (95% CI)

Adjusted OR (95 % CI)

P value

Overall genotype

n = 140

58

82

n = 170

143

27

7.4 (4.3–12.7)

7.4 (4.3–12.7)

<0.05

Age group

 <45

100 (71)

40

60

60 (35)

49

11

6.7 (3–14.4)

3.9 (1.7–9.2)

<0.05

 ≥45

40 (29)

18

22

110 (65)

94

16

7.1 (3.1–15.6)

6.9 (2.6–17.7)

<0.05

Sex

 Female

114 (81)

50

64

30 (17.6)

26

4

8.3 (2.6–25.3)

7.6 (2.0–28.8)

<0.05

 Male

26 (19)

8

18

140 (82.4)

117

23

11.4 (4.3–29.2)

11.5 (3.6–36.9)

<0.05

Dwelling

 Rural

112 (80)

40

72

50 (29.4)

34

16

3.8 (1.8–7.7)

3.7 (1.5–9.1)

<0.05

 Urban

28 (20)

18

10

120 (70.6)

109

11

5.5 (2.0–14.8)

5.2 (1.4–9.1)

<0.05

Smoking

 Never

124 (89)

48

76

50 (29.4)

33

17

3 (1.5–5.9)

3.1 (1.3–7.4)

<0.05

 Ever

16 (11)

10

6

120 (70.6)

110

10

6.6 (1.98–21.7)

7.2 (1.2–42.0)

<0.05

Benign thyroid disease

 Yes

84 (60)

34

50

   

1.1 (0.5–2.42)

 

>0.05

 No

56 (40)

24

32

TSH levels

 Elevated

100 (71)

44

56

   

0.7 (0.3–1.6)

 

>0.05

 Normal

40 (29)

14

26

Histological types

 Papillary

118 (84)

54

64

   

0.26 (0.06–1.0)

 

<0.05

 Follicular

22 (16)

4

18

Tumor

Grade

 WD

134 (96)

56

78

   

0.7 (0.05–8.6)

 

>0.05

 PD

6 (4)

2

4

Stage, <45 years

 Stage I

94 (67)

38

56

   

0.7 (0.06–8.7)

 

>0.05

 Stage II

6 (4.3)

2

4

Stage, ≥45 years

 Stages I and II

36 (25.7)

16

20

   

1.25 (0.15–9.8)

 

>0.05

 Stage III and above

4 (3)

2

2

TSH thyroid-stimulating hormone, WD well differentiated, PD poorly differentiated

Table 4

Distribution of HRAS T81C genotypes and its allele frequency in cases and controls

 

Cases

Controls

OR (95 % CI)

P value

(n = 140)

(n = 170)

Genotype

 TT

58 (41.4)

143 (84.1)

  

 TC

54 (38.6)

20 (11.7)

6.6 (3.6–12.0)

<0.05

 CC

28 (20)

7 (4.2)

9.8 (4.0–23.6)

<0.05

Allele type

 T

170 (60.7)

306 (90)

  

 C

110 (39.3)

34 (10)

5.8 (3.7–8.7)

<0.05

Table 5

Multivariate analysis of selected demographic and risk factors in thyroid cancer cases and controls

Parameters

Frequency

Coding (reference)

Sig. (P value)

Exp. (OR)

95.0 % CI for Exp.

Lower

Upper

Genotype

TT

201

0.000

    

TC + CC

109

1.000

0.000

4.489

1.963

10.266

Gender

Female

144

1.000

    

Male

166

0.000

0.000

16.355

7.223

37.031

Smoking

Nonsmoker

174

1.000

    

Smoker

136

0.000

0.000

7.382

3.357

16.235

Dwelling

Rural

162

1.000

    

Urban

148

0.000

0.00

5.542

2.488

12.346

Age

<45

159

1.000

    

>45

151

0.000

0.015

2.528

1.200

5.328

Discussion

Studies on a variety of tumors have demonstrated some “hot spots” in RAS gene family that are susceptible to point mutations. Many studies have detected different types of RAS mutations in human thyroid tumors [24, 3741], but RAS gene family members have not been screened for mutation in the same sample series in thyroid tumors in Kashmiri patients. Thus, aim of this study was to carry out comprehensive mutational screening of all three RAS genes HRAS, NRAS, and KRAS (exons 1 and 2) in a series of 60 confirmed thyroid cancer samples. During our study for mutational analysis of HRAS in thyroid cancer, HRAS T81C was frequently observed and was considered to be an informative SNP. Since this polymorphism has been reported only once in thyroid cancer; further, evaluation was imperative, to elucidate the conformity of the results in the backdrop of different ethnic backgrounds; thus, we conducted a case–control polymorphic study of HRAS T81C to assess the role of this SNP in thyroid cancer from in Kashmiri population (north India).

Activating RAS mutations occur in ∼30 % of human cancers. Activating mutations of all three RAS oncogenes (HRAS, KRAS, and NRAS) in thyroid tumors were first reported in 1988 [41, 42]. Also earlier studies demonstrated that RAS mutations were more frequent in follicular than in papillary tumors and that the two types of carcinoma had different mutation patterns [43]. The presence of mutations in up to 50 % of micro-follicular adenomas supported the contention that RAS oncogene activation was an early event in follicular thyroid tumorigenesis [3844]. In subsequent studies on RAS oncogenes in thyroid tumors, various laboratories reported disparate results regarding the incidence of mutations, isoform pattern (HRAS, KRAS, and NRAS), and correlation of mutations with histology. Our study was limited to screening of two hot spot exons of each RAS family of genes but in contrast to most of the studies showed no activating mutations in the thyroid tumors. Despite this, our report corroborates with a few studies that did not observe any mutational event of RAS genes in thyroid tumors. The incidence of RAS oncogene mutations ranged from 0 to 50 % in papillary cancer [4547], 0–85 % in adenomas [48], 14–62 % in follicular carcinomas [48, 49], and 0–60 % in anaplastic carcinomas [45, 48]. Although some investigators found no correlation between the mutated RAS oncogenes isoform and tumor pathology [39], others reported a higher frequency of mutations in the codon 61 of HRAS and NRAS in follicular tumors and poorly differentiated carcinomas [39, 49]. All these reports though clearly show the involvement of RAS genes in the development of thyroid tumors, but there has been discrepancies regarding the frequencies of the mutations in RAS genes.

However, during our study for mutational analysis of HRAS in thyroid cancer, HRAS T81C was frequently observed and was considered to be an informative SNP. Several molecular alternations in RAS gene have been identified such as minisatellite and acquired mutations in various cancers. Particularly, HRAS has been predominantly studied and shown to be involved in different cancers [19, 20, 22].

The frequency of three genotypes in our study revealed genotypic frequency of TT, TC, and CC as 41.4, 38.6, and 20 % in cases and 84.1, 11.7, and 4.2 % in controls, respectively. This pattern of distribution of allele frequency in control population group was compatible with few studies conducted by Zhang et al. and Arshad et al. [29, 31]. In thyroid cases, however, we found higher frequency of variant genotypes as compared to other studies conducted on various cancers [29, 31, 50, 51]. In our study, the distribution of genotypes in cases is in accordance with the only study conducted on thyroid cancer by Castro et al. [30] but shows marked difference in control population which is possibly due to ethnic differences between populations that were the subject of the studies. Besides, the frequency of genotypes found in cases in this study was found to be in accordance with Andreas Johne et al. [28] conducted on bladder cancer (CC = 20 % vs. CC = 13.5 %). The frequency of C allele observed in thyroid cases was 0.39 compared with 0.10 in controls, and this difference was highly significant (P < 0.05) and was in tune with a study conducted on oral cancer, bladder, and gastric cancer [29, 31] in addition to the study on thyroid cancer [30]. Our study revealed a 7-fold increased risk of thyroid cancer in carriers of the variant genotype (TC + CC) in cases with aggregate frequency of 58.6 % in comparison to 16 % in controls (TC + CC) (P < 0.05). Variant allele was observed higher in the few other studies conducted by Castrol et al. [30], Johne et al. [28], Zhang et al. [29], and Arshad et al. [31], compared with our study but the frequency as well as fold increase in the risk was found higher in our report (62 vs. 47 vs. 47 vs. 36 vs. 58 %—our study). This study, therefore, reveals a statistically significant increased risk for thyroid cancer, both either when stratified with C allele or in combination of the variant genotypes TC + CC compared with the TT genotype. Thus, our report shows that HRAS T81C SNP is a strong susceptibility factor for the development of thyroid cancer. Consistent with the tissue specificity hypothesis and various studies that had confirmed that the HRAS gene plays a more important role in bladder cancer acquired amino acid mutations in the hotspot codons 12, 13, and 61, which prolong the GTP-bound activated state of the HRAS product [52], this polymorphism HRAS T81C SNP does not lead to the alternation of ras protein structure, and it affects the cancer susceptibility possibly through linkage disequilibrium with other potential functional variants of HRAS. One of the linkage candidates is a region of variable tandem repeats about 1 kb downstream exon 4, with a possible transcriptional enhancer activity [53]. Another associated polymorphic site is hexanucleotide repeat located about 80 bp upstream of the 5′-end of exon1 [54]. Yet another report has shown that HRAS T81C might be serving as a marker of other polymorphisms in intron D2 of HRAS that would act as regulators of IDX inclusion [30].

We found TC and CC genotypes frequently in follicular tumors as against papillary tumors (21 vs. 82 %), and this difference showed statistically significant relation (P < 0.05). This observation in our report is in contrast with the only study conducted on HRAS T81C polymorphism in thyroid cancer [30]. Furthermore, interestingly we found an inverse relation in never smokers in cases with higher frequency of combined TC and CC genotypes (61 %) (P < 0.05) while as ever smokers presented with lesser frequency of variant alleles possibly showing reduced risk for thyroid cancer (adjusted OR = 3.6, P > 0.05) (Table 3). A mechanism by which cigarette smoking might reduce the risk of thyroid cancer is by lowering the endogenous levels of TSH [5558]. Although not all studies have reported this effect [59], it has been suggested that increased levels of TSH are associated with an increased risk of thyroid cancer [60]. Thus, an agent which lowers TSH would then potentially protect against the disease. An additional explanation for the effect of smoking on thyroid cancer among women is suggested by studies which indicate that smoking may have an anti-estrogenic effect [61].

Conclusion

It is evident from our study that although thyroid cancer is highly prevalent in this region, the mutational events for RAS genes do not seem to be involved in the thyroid carcinogenesis. Contrary to this HRAS T81C SNP has been found to moderately increase thyroid cancer risk with variant alleles implicated more in follicular thyroid tumors. Interestingly, we also observed that HRAS T81C variant allele shows reduced risk for the smokers with thyroid cancer. These correlations need to be authenticated in a large sample study in the future to help better discern racial and ethnic differences to determine the aggressiveness of thyroid cancer.

Acknowledgments

The authors acknowledge the technical staff of operation theater of Department of General Surgery especially Mr. Abdul Ahad who helped us in procuring the tissue samples. Our thanks are also due to the Head and faculty members of Department of Nuclear Medicine who helped us in procuring the blood samples. The authors also acknowledge the timely and precious help of Dr. Rayes Ahmad Dar of Department of Statistics, Sher-I-Kashmir Institute of Medical Sciences.

Conflicts of interest

None

Copyright information

© International Society of Oncology and BioMarkers (ISOBM) 2012