HSPB1 rs2070804 polymorphism is associated with the depth of primary tumor
Abstract
Background
Colorectal cancer (CRC) is the third most common cancer in the world. Genome-wide association studies are a powerful method to analyze the status of single-nucleotide polymorphisms (SNPs) in specific genes. Heat shock proteins (HSPs) were found to be involved in the cancer progression and chemoresistance. However, there is still no further study about polymorphisms of HSP beta-1 (HSPB1) in colorectal cancer. We proposed the SNP of HSPB1 may be correlated with the progression and metastasis in colon cancer.
Methods
We recruited 379 colorectal cancer patients and categorized as four stages following the UICC TNM system. Then, we selected tagging SNPs of HSPB1 by 10% minimum allelic frequency in Han Chinese population from the HapMap database and analyze with the Chi-square test.
Results
We demonstrated the association of HSPB1 genetic polymorphisms rs2070804 with tumor depth with colorectal cancer. But, there is a lack of association between HSPB1 genetic polymorphisms and colorectal cancer invasion, recurrence or metastasis.
Conclusions
The polymorphisms of HSPB1 seemed to change the tumor behavior of colorectal cancer. HSPB1 rs2070804 polymorphism is associated with the depth of the primary tumor. But, there is no further correlation with other to the clinical parameters such as cancer invasiveness, local recurrence, or distant metastasis.
1 BACKGROUND
Colorectal cancer (CRC) is the third most common cancer in males and the second in females, accounting for nearly 10% of new cancer cases and 8% of cancer-related deaths worldwide.1 Although the mortality rate of CRC in United States, Japan, and Singapore has declined gradually due to screening program, the incidence and mortality of CRC in most countries of the world have increased rapidly in the past decades.2 From numerous research, scientists conclude that both genetic disposition and environmental factors contribute to carcinogenesis of CRC. Several risk factors have also been established, including positive family history, obesity, red meat intake, smoking, and heavy alcohol intake.
There are two well-known forms of hereditary CRC: familial adenomatous polyposis and Lynch syndrome (also called hereditary nonpolyposis CRC), which is due to germline mutations in the APC gene3 and germline mutations in DNA MMR genes,4 respectively. However, the majority of patients with CRC (over 80%) have sporadic disease without evidence of inherited associated disorder. Only 5% to 6% of patients with CRC are associated with germline mutations that confer as an inherited predisposition to CRC.5 Patients with a positive family history of CRC were suggested to contain hereditary predisposition, sharing common risk factors, or both. Recently, new studies have failed to discover other high-penetrance genes, suggesting that multiple small low-penetrance alleles may explain the remaining genetic risk for CRC.
Genome-wide association studies (GWAS) is a high throughput approach to determine single-nucleotide polymorphisms (SNPs). GWAS can be conducted using sets of SNPs that tag the most widely known common variants in the genome and scan for associations independent of a gene's function or chromosomal position. The first susceptibility locus for CRC identified in these studies was 8q24. This genomic region was noted for prostate cancer, through a linkage study followed-up by an association study and independently through an admixture scan in African Americans.6
Heat shock protein (HSP) is a highly conserved molecular chaperone protein responding for stress response. Evidence show HSPs are also involved in the cancer progression, metastasis, prognosis, and drug resistance.7 HSP 27, about 27 kDa in size, is encoded by the HSP beta-1 (HSPB1) gene. HSP27 can interact with Bid or cytochrome c to play an antiapoptotic role or as a chaperone to prevent misfolded proteins.8-12 So far, close relationship of HSP27 with cancers has been found in breast cancer,13, 14 esophageal cancer,15, 16 CRC,17 head and neck cancer,18 and other solid tumors. A study in the Chinese population showed that the C allele of the rs2868371 functional promoter polymorphism of HSPB1 was associated with an increased risk of lung cancer, but better survival in patients with advanced NSCLC compared with the G allele.19Another study demonstrated that HSPB1 polymorphisms (rs2868371) were associated with poorer overall survival in US population with non-small-cell lung cancer after radio(chemo)therapy.20 But no studies have shown the role of HSPB1 in CRC. Therefore, we propose that SNPs of HSPB1 could be a progression marker in colon cancer.
To investigate the hypothesis of different patterns of HSPB1 SNP on CRC susceptibility across populations and phenotypic subgroups, we aimed to assess whether inherited risk variants at rs2868371, rs2009836, rs2070804, rs7459185, and rs2961047 exerted a differential effect on CRC. rs2868371 and rs2009836 locate on the near 5′ UTR of HSPB1, rs2070804 of HSPB1 locates on intron, rs7459185 and rs2961047 locate on the near 3′ UTR of HSPB1.
2 METHODS
2.1 Study subjects
This study was approved by the Institutional Review Board of Kaohsiung Medical University Hospital. We recruited 379 patients with CRC after obtaining their informed consent form. UICC TNM staging system was applied to classify the tumors by anatomic disease extent. UICC TNM staging system records the primary(T) and regional nodal extent(N) of the tumor and the absence or presence of metastases(M) and is categorized into four stages (stage I-IV). All the patients were diagnosed by colonoscopy and pathological report. Local recurrence and distant metastasis events were also recorded for analysis by chart review.
2.2 DNA extraction
DNA was extracted from buffy coat by treating with 0.5% SDS lysis buffer at first. For digestion of nuclear protein, protease K (1 mg/mL) was added and then incubated at 60°C for 4 hours. After using the Gentra (Qiagen, Hilden, Germany) extraction kit and followed by 70% alcohol precipitation, the total DNA was acquired.
2.3 Genotyping
We selected tagging SNPs of HSPB1 by 10% minimum allelic frequency (MAF) in Han Chinese population from the HapMap database (http://hapmap.ncbi.nlm.nih.gov/). A total of five tagging SNP of HSPB1 were genotyped by using the TaqMan Allelic Discrimination Assay (Applied Biosystems, Foster City). rs2868371 and rs2009836 locate on the near 5′ UTR of HSPB1, rs2070804 of HSPB1 locates on intron, rs7459185 and rs2961047 locate on the near 3′ UTR of HSPB1. PCR was applied by using ABI StepOne Plus with the following condition: denaturation at 95°C for 10 minutes, followed by 40 cycles of denaturation at 92°C for 15 seconds and annealing and elongation at 60°C for 1 minute, then hold at 60°C for fluorescence detection. System SDS software version 1.2.3 was used to measure and analyze the fluorescence of probes.
2.4 Statistical analysis
R 3.2.0 was applied in this study. All the correlation between genotypes and pathologic states were examined by the Chi-square test. P values less than 0.05 were considered statistically significant.
3 RESULTS
3.1 Basal characteristic of study subjects
Among 379 colorectal patients, 207 (54.6%) patients were male individuals, and mean age of all patients was 64.3 ± 12.1. In this study, 22.1% of patients with CRC were stage I, 36.1% of patients with CRC were stage II, 33.2% of patients with CRC were stage III, and 8.6% of patients with CRC were stage IV (Table 1).
| Characteristics | Patients with colorectal cancer |
|---|---|
| Number of subjects | 379 |
| Gender: male, No (%) | 207 (54.6%) |
| Age, y | 64.3 ± 12.1 |
| Range | 20-93 |
| TNM stagea | |
| I | 77 (22.1%) |
| II | 126 (36.1%) |
| III | 116 (33.2%) |
| IV | 30 (8.6%) |
- a There were 30 subjects without TNM stage information.
3.2 Association of HSPB1 genetic polymorphisms for the tumor depth of patients with colorectal cancer
As shown in Table 2, the correlation between HSPB1 genetic polymorphisms and stages of tumor depth, lymph node, and TNM were examined. rs2070804 of HSPB1 showed significant association with advanced cancer depth status. GG and GT genotypes of rs2070804 showed more prevalent in T3 and T4 stages compared with T1 and T2 stages (P = 0.0281). There was no evidence supporting the association with the other four SNPs (rs2868371, rs2009836, rs7459185, and rs2961047) of HSPB1. In addition, we did not find any correlation between the genetic polymorphisms of HSPB1 and the status of lymph node metastasis or cancer stages.
| SNP | Genotype | Depth | P value | LN | P value | TNM | P value | |||
|---|---|---|---|---|---|---|---|---|---|---|
| T1+T2 | T3+T4 | N0 | N1+N2 | I+II | III+IV | |||||
| rs2868371 | CC | 20 (17.4) | 30 (13.7) | 0.6590 | 26 (13.8) | 24 (16.0) | 0.8337 | 26 (13.3) | 24 (17.3) | 0.5925 |
| CG | 58 (50.4) | 112 (51.4) | 98 (51.8) | 77 (51.3) | 102 (52.0) | 70 (50.3) | ||||
| GG | 37 (32.2) | 76 (34.9) | 65 (34.4) | 49 (32.7) | 68 (34.7) | 45 (32.4) | ||||
| rs2009836 | CC | 6 (5.6) | 11 (5.3) | 0.9229 | 8 (4.5) | 8 (5.6) | 0.8658 | 10 (5.4) | 7 (5.3) | 0.9990 |
| CT | 37 (34.6) | 77 (36.8) | 66 (37.1) | 50 (35.2) | 67 (36.2) | 48 (36.4) | ||||
| TT | 64 (59.8) | 121 (57.9) | 104 (58.4) | 84 (59.2) | 108 (58.4) | 77 (58.3) | ||||
| rs2070804 | GG | 5 (4.4) | 16 (7.4) | 0.0281* | 10 (5.3) | 11 (7.4) | 0.7446 | 11 (5.7) | 10 (7.2) | 0.7899 |
| GT | 30 (26.3) | 83 (38.4) | 66 (35.3) | 51 (34.2) | 65 (33.7) | 49 (35.3) | ||||
| TT | 79 (69.3) | 117 (54.2) | 111 (59.4) | 87 (58.4) | 117 (60.6) | 80 (57.5) | ||||
| rs7459185 | CC | 17 (14.6) | 35 (15.6) | 0.4280 | 30 (15.5) | 22 (14.2) | 0.9333 | 28 (14.1) | 24 (16.7) | 0.7965 |
| CG | 52 (44.4) | 113 (50.5) | 92 (47.7) | 76 (49.0) | 97 (48.7) | 69 (47.9) | ||||
| GG | 48 (41.0) | 76 (33.9) | 71 (36.8) | 57 (36.8) | 74 (37.2) | 51 (35.4) | ||||
| rs2961047 | AA | 0 (0.0) | 0 (0.0) | 0.5462 | 0 (0.0) | 0 (0.0) | 0.5365 | 0 (0.0) | 0 (0.0) | 0.3912 |
| AG | 12 (10.6) | 27 (12.9) | 21(11.8) | 21 (14.1) | 21 (11.3) | 20 (14.5) | ||||
| GG | 101 (89.4) | 182 (87.1) | 157 (88.2) | 128 (85.9) | 165 (88.7) | 118 (85.5) | ||||
- Abbreviations: HSPB1, heat shock protein; LN, lymph node; SNP, single-nucleotide polymorphism.
- * Significant (P < 0.05) values are in bold.
3.3 Lack of association between HSPB1 genetic polymorphisms and colorectal cancer invasion
Tumor depth is affected by the ability of tumor invasion, and the correlation between tumor depth status and rs2070804 of HSPB1 was identified in this study. Accordingly, the association between invasion and HSPB1 also needed to be considered. Tumor invasion was grouped into the vascular invasion and perineural invasion and analyzed the association with genetic polymorphisms of HSPB1. However, as shown in Table 3, no statistically significant correlation was found between the HSPB1 gene and invasion of CRC. Besides, there is also no difference in local recurrence or distant metastasis in these five SNPs, which is shown in Table 4.
| SNP | Genotype | Vascular invasion (%) | P value | Perineural invasion (%) | P value | ||
|---|---|---|---|---|---|---|---|
| Positive | Negative | Positive | Negative | ||||
| rs2868371 | CC | 11 (11.6) | 35 (14.8) | 0.4816 | 12 (13.8) | 35 (14.3) | 0.9936 |
| CG | 47 (49.5) | 125 (52.7) | 45 (51.7) | 126 (51.4) | |||
| GG | 37 (38.9) | 77 (32.5) | 30 (34.5) | 84 (34.3) | |||
| rs2009836 | CC | 3 (3.3) | 14 (6.3) | 0.3704 | 7 (8.1) | 10 (4.4) | 0.3898 |
| CT | 30 (32.6) | 81 (36.7) | 28 (32.6) | 83 (36.6) | |||
| TT | 59 (64.1) | 126 (57.0) | 51 (59.3) | 134 (59.0) | |||
| rs2070804 | GG | 7 (7.4) | 13 (5.5) | 0.1591 | 4 (4.5) | 16 (6.7) | 0.4447 |
| GT | 39 (41.5) | 75 (31.9) | 35 (39.3) | 78 (32.5) | |||
| TT | 48 (51.1) | 147 (62.6) | 50 (56.2) | 146 (60.8) | |||
| rs7459185 | CC | 15 (15.3) | 34 (14.1) | 0.9319 | 10 (11.0) | 38 (15.3) | 0.4271 |
| CG | 46 (46.9) | 118 (49.0) | 49 (53.8) | 116 (46.8) | |||
| GG | 37 (37.8) | 89 (36.9) | 32 (35.2) | 94 (37.9) | |||
| rs2961047 | AA | 0 (0.0) | 0 (0.0) | 0.3014 | 0 (0.0) | 0 (0.0) | 0.8012 |
| AG | 9 (9.6) | 31 (13.8) | 10 (11.8) | 30 (12.8) | |||
| GG | 85 (90.4) | 194 (86.2) | 75 (88.2) | 204 (87.2) | |||
- Abbreviations: HSPB1, heat shock protein; SNP, single-nucleotide polymorphism.
| SNP | Genotype | Recurrence (%) | P value | Metastasis (%) | P value | ||
|---|---|---|---|---|---|---|---|
| Positive | Negative | Positive | Negative | ||||
| rs2868371 | CC | 15 (14.7) | 37 (15.2) | 0.5472 | 15 (15.3) | 37 (15.0) | 0.3276 |
| CG | 57 (55.9) | 121 (49.8) | 56 (57.1) | 122 (49.4) | |||
| GG | 30 (29.4) | 85 (35.0) | 27 (27.6) | 88 (35.6) | |||
| rs2009836 | CC | 7 (7.0) | 10 (4.4) | 0.6045 | 7 (7.3) | 10 (4.4) | 0.5029 |
| CT | 38 (38.0) | 83 (36.9) | 37 (38.5) | 84 (36.7) | |||
| TT | 55 (55.0) | 132 (58.7) | 52 (54.2) | 135 (59.0) | |||
| rs2070804 | GG | 7 (6.7) | 14 (5.9) | 0.9034 | 6 (6.0) | 15 (6.2) | 0.9912 |
| GT | 37 (35.6) | 81 (34.0) | 35 (35.0) | 83 (34.3) | |||
| TT | 60 (57.7) | 143 (60.1) | 59 (59.0) | 144 (59.5) | |||
| rs7459185 | CC | 19 (17.9) | 34 (13.7) | 0.2382 | 10 (11.0) | 38 (15.3) | 0.2886 |
| CG | 55 (51.9) | 117 (47.2) | 49 (53.8) | 116 (46.8) | |||
| GG | 32 (30.2) | 97 (39.1) | 32 (35.2) | 94 (37.9) | |||
| rs2961047 | AA | 0 (0.0) | 0 (0.0) | 0.1878 | 0 (0.0) | 0 (0.0) | 0.2371 |
| AG | 9 (9.3) | 34 (14.5) | 10 (11.8) | 30 (12.8) | |||
| GG | 88 (90.7) | 201 (85.5) | 75 (88.2) | 204 (87.2) | |||
- Abbreviations: HSPB1, heat shock protein; SNP, single-nucleotide polymorphism.
4 DISCUSSION
Numerous SNPs have been reported to be associated with CRC.21 Some SNPs are in connection with increasing risk of developing CRC,22, 23 some SNPs appear to lower the metastatic risk of CRC and others have been associated with the efficacy of various treatments for CRC.24
Increasing GWAS have brought new hope to identify susceptibility loci of CRC in the past decade. Several previous large-scale meta-analysis studies of GWAS on colorectal cancer identified novel CRC risk loci at 14q22 (BMP4), 16q22 (CDH1), 19q13 (RHPN2), and 20p12,25 genetic loci at 1q41, 3q26 (MYNN), 12q13 (ATF1/DIP2B), and 20q13 (LAMA5),26 6p21 (CDKN1A), 11q13 (POLD3), and Xp22 (SHROOM2),27 and 5q31, 12p13 (CCND2), and 20p12 (HAO1/PLCB1).28 So far, evidence-based medicine has not concluded determinative foci.
The heat shock 27 kDa protein beta 1 (also called HSP 27, Hsp27) made by gene HSPB1, located on 7q11.23, is a member of the small HSP family and is constitutively expressed at low levels in normal cells. Aberrant elevated expression of HSP27 has been reported in colorectal cancer.17 Furthermore, Hsp27 expression also plays a role in circumventing TP53 mediated oncogene-induced senescence in CRC29 and involved in the chemoresistance of colorectal cancer cells possibly by reducing caspase-3 activity.30 Hsp27 interacts with actin microfilaments and tubulin/ microtubules to stabilize the structure of the cytoskeleton to maintain cell survival.19, 31
SNPs in HSPs were commonly mentioned in recent years. Several HSPs are key mediator in protein folding to establish the correct protein confirmation.32, 33 So, SNPs may result in the function change in the management of misfolded proteins and could also change the invasiveness of cancer. Savas et al34 reported that DDX20-rs197412, PTGS2-rs5275, and HSPA5-rs391957 were associated with disease-free survival in colorectal cancer in 102 previously reported common genetic polymorphisms. A meta-analysis also showed the polymorphism of HSP70-2 in the three main HSP 70 was associated with the risk of cancer (HSP70-2; AG versus AA: OR = 1.38; 95% CI: 1.03-1.84; GG versus AA: OR = 2.34, 95% CI: 1.21-4.54).35 Therefore, we try to find if SNPs of Hsp27 will influence the behavior of colorectal cancer.
Our study was designed to investigate the relationship between SNPs of HSPB1 and cancer stage in CRC. The data revealed rs2070804 SNP is associated with the depth of the primary tumor. Other five SNPs were found at other clinical studies, but they fail to demonstrate their relationship with CRC.
The correlation between HSPB1 rs2070804 polymorphism and the depth of the primary tumor was supposed to be connected to the invasiveness of colorectal cancer which could result in more metastasis or recurrence. However, further analysis cannot demonstrate this outcome and no clinical significance in cancer metastasis or recurrence. The same situation was also noted in some clinical studies. The depth of tumor invasion is not correlated to cancer recurrence or survival.36-38 The presence of vascular invasion, perineural invasion, and postoperative CEA level was more commonly mentioned in correlating to the poor cancer survival.37 Although some HSPs showed their SNPs correlated with cancer progression, the SNPs in HSP 27 seemed to have no correlation to cancer progression but only to the tumor depth.
There are still some limitations and delimitations to this study such as variations in different races and limited case numbers. Further validation by larger prospective studies and expelling the influence of chemotherapy or radiotherapy may further clarify the relationship between SNPs of Hsp27 and colorectal cancer. Besides, a new approach may be more helpful in analyzing the influence of SNPs of Hsp27.
5 CONCLUSION
In summary, we have recruited 379 patients with CRC of Han Chinese population and identified the correlation of HSPB1 genetic polymorphisms rs2070804 with the tumor depth, but not with colorectal cancer invasion, recurrence or metastasis.
FUNDING INFORMATION
This work was supported by grants through funding from the Ministry of Science and Technology (MOST107-2321-B-037-003, MOST107-2314-B-037-116, MOST107-2314-B-037-022-MY2, MOST107-2314-B-037-023-MY2) and the Ministry of Health and Welfare (MOHW106-TDU-B-212-113006, MOHW107-TDU-B-212-123006, MOHW107-TDU-B-212-114026B) funded by Health and welfare surcharge of tobacco products), by the Grant of Biosignature in Colorectal Cancers, Academia Sinica, Taiwan, R.O.C.
CONFLICTS OF INTEREST
The authors declare that they have no conflicts of interest.
AUTHOR CONTRIBUTIONS
CSH, CYH, YWH, PTM, WCC, YJC, JYW, and PLW designed and performed the experiments. CSH, CYH, YWH, PTM, WCC, YJC, JYW, and PLW performed the data analysis and presentation. CSH, CYH, YWH, PTM, WCC, YJC, JYW, and PLW prepared the manuscript.
