Non-melanoma skin cancer (NMSC) is the most common cancer in the world with a lifetime risk for development as high as 2 in 3 in Queensland, Australia. Mortality is quite low, representing an approximate 360 deaths in Australia annually but cost of treatment is extremely high, estimated at $232 million each year. Squamous cell carcinoma (SCC) and basal cell carcinoma (BCC) are the two most common forms of NMSC. Although BCC generally do not have the propensity to metastasise, they are highly invasive and can be locally destructive. SCC on the other hand is invasive and has metastatic potential. SCC is generally derived from a precursor lesion, solar keratosis (SK), which is also considered to be a biomarker of BCC, SCC and malignant melanoma. According to one theory, SKs actually represent the first recognisable stage of SCC development and therefore may be indicative of the earliest stage of NMSC development. In addition to these common forms of NMSC, rarer forms such as keratoacanthoma (KA), which spontaneously regress, and SCC in situ, which rarely become invasive, may provide clues into protective mechanisms associated with prevention of development. Like all other cancers, NMSC arises from an accumulation of genetic abnormalities that result in severe cellular dysfunction. A number of genes have been proposed in the development of NMSC, including p53, CDKN2a, Bcl-2 and the Ras family of genes, which are typically associated with proliferative and differentiation processes. Also, a number of genetic disorders that predispose individuals to NMSC have also been identified. Genetic abnormalities in these genes may be a result of somatic mutations that may be promoted by environmental carcinogens. For NMSC, ultraviolet (UV) radiation is the primary environmental stimulus that acts upon skin to generate mutations. UV effects are 2-fold; the first being direct damage produced by UVB radiation and the second being indirect damage as a result of UVA-induced oxidative stress. In addition to mutations of genes that directly result in carcinogenesis, polymorphic variants of genes may also play a role in susceptibility to NMSC. These susceptibility genes may have immunogenic, detoxifying or transcriptional roles that could be involved in increased mutagenesis or activation of cancer causing genes. The purpose of this study was ultimately to identify further molecular based mechanisms associated with the development of non-melanoma skin cancer. Initially, this study aimed to examine the effects of aberrant chromosomal regions on NMSC development and also to identify candidate genes within these regions that may be implicated in the development and progression of NMSC. Also, based on chromosomal and functional implications, a number of candidate genes were assessed using association analysis to determine their involvement in susceptibility to the earliest stages of NMSC development. Implicated susceptibility genes were then further investigated to determine their response to UV radiation. Therefore the methodological approach of these studies was based on three broad technical applications of cytogenetic, association and expression analyses. Previous comparative genomic hybridisation (CGH) studies implicated the 18q chromosomal region in progression of SK to SCC and this region was therefore suspected of harbouring one or more tumour suppressor genes that were associated with a more malignant phenotype. Following on from this analysis, loss of heterozygosity (LOH) analysis was used for further delineation of this region and possibly to implicate candidate genes involved in progression. Additionally, CGH was used to investigate keratoacanthoma to determine aberrant regions that might be involved in progression and also regression of this NMSC. Genes that had potential functional roles in NMSC development and that were located in or near regions implicated by these cytogenetic analyses were further investigated using association analysis. Association analysis was performed using polymerase chain reaction and subsequent restriction enzyme digestion or GeneScan analysis to determine genotype and allele frequencies in an SK affected versus control population for polymorphisms within a number of candidate genes. This population was carefully phenotyped so that not only genotypic factors could be analysed but also their interaction with a number of phenotypic and environmental risk factors. Genes with polymorphisms that did show association with solar keratosis development were then examined functionally. Specifically, gene expression analysis was undertaken to investigate their response to UV radiation. Both UVA only and combined UVA/UVB treatments were used for short term irradiation and also for long term irradiation with recovery to determine differential effects of UV range and dose in human skin. Relative mRNA expression analysis of these genes was performed using quantitative real time reverse transcription polymerase chain reaction to determine if UV radiation imposed gene expression changes in the skin. A combination of these methodologies provided a wide basis for investigation of NMSC. Cytogenetic, association and expression analyses all allow for the identification of molecular risk factors that cause or are associated with NMSC development and progression. These analyses provided diverse results that implicated various molecular mechanisms in the development of NMSC. Cytogenetic analysis is a powerful technique, especially for the identification of a broad range of aberrations throughout the genome. This study employed LOH analysis to investigate an implicated region involved in progression to SCC and to attempt identification of candidate genes that may be involved in this process. LOH analysis was successfully performed on 9 SCCs, 5 SCCs in situ and 2 SKs using 8 microsatellite markers within the 18q region. Polymerase chain reaction (PCR) was used to amplify polymorphic regions of these markers and genotypic composition was determined for normal and cancerous tissue within the specimen. In heterozygote individuals, determined by analysis of normal tissue, the cancerous tissue was examined to determine if alleles within the implicated region had been lost. However, after analysis of multiple different samples, there was no LOH detected in any of the samples examined for this analysis. This does not necessarily reject a role for 18q, or genes within this region, as the localisation of candidate tumour suppressor genes within a small region may indicate a tighter region of involvement than was expected. As such, a more targeted study may further delineate this region and implicate candidate genes in progression of SK to the more malignant phenotype of SCC. Further CGH analysis of keratoacanthoma was also undertaken to identify aberrations associated with development and also regression of this skin cancer. CGH was performed using universal amplification and nick translation to incorporate a fluorescent dye. Differentially labelled normal and tumour DNA were then competitively hybridised to a normal metaphase spread and fluorescence emission indicated either amplification or deletion of specific chromosomal regions. In total, 6 KA samples were analysed, with 2 samples each from evolving, matured and regressing stages of KA development. In general, regressing KAs appeared to be more highly associated with deleted regions than evolving and matured KAs. Specifically, the 15q chromosomal region that was deleted in regressing KAs but amplified in evolving or matured KAs, may be significantly involved in the process of KA regression. Also various candidate genes that were being considered for analysis were located in or near some of these implicated regions, including GSTM1, GSTP1 and SSTR2. As such, these candidate genes were targeted for further investigation. A number of susceptibility genes that were located in or near aberrant regions implicated in NMSC development were investigated using association analysis. These genes included members of the somatostatin receptor family (SSTR1 and SSTR2), members of the glutathione-S-transferase (GST) family (GSTM1, GSTT1, GSTP1 and GSTZ1) and the vitamin D receptor (VDR). Studies detected a number of interesting interactions between genetic, environmental and phenotypic factors in the development of the early stages of non-melanoma skin cancer. Additionally, genes implicated in NMSC development were further investigated using expression analysis to determine response to UV radiation. Association analysis was initially performed on members of the somatostatin receptor family. Somatostatin is a growth inhibiting factor, amongst other things, that mediates its actions through the somatostatin receptors (SSTRs). The presence of these receptors (SSTR1-5) in tumour cells indicates a potential for somatostatin to bind and suppress growth, as well as allowing for therapeutic treatment with somatostatin analogues. Additionally, expression of these receptors in normal tissue, including skin, should allow for potential protection against tumour growth. The genes for SSTR1 and SSTR2 have been shown to contain dinucleotide repeat polymorphisms, and although these polymorphisms may not directly result in altered expression or binding potential, they may be linked to another functional polymorphism that does. Using association analysis the SSTR1 and SSTR2 genes were investigated to determine whether they play a role in the development of solar keratosis. Results showed that there were no significant differences between SSTR1 and SSTR2 polymorphism frequencies in the tested solar keratosis population (P = 0.10 and P = 0.883, respectively) as compared to an unaffected population. Hence, these studies do not support a role for the SSTR1 or SSTR2 genes in solar keratosis development. Further association analysis and subsequent expression analysis was also performed on members of the glutathione-S-transferase family. The GST enzymes play a role in the detoxification of a number of carcinogens and mutagens, including those produced by UV-induced oxidative stress. This study examined the role of GSTM1, GSTT1, GSTP1 and GSTZ1 gene polymorphisms in susceptibility to SK development. Association analysis was performed to detect allele and genotype frequency differences in SK affected and control populations using PCR and restriction enzyme digestion. No significant differences were detected in GSTP1 and GSTZ1 allele or genotype frequencies, however polymorphisms within both genes were found to be in linkage disequilibrium, as previously reported, and a new allelic variant of the GSTZ1 gene was identified. Significant associations between GSTM1 (P = 0.003) and GSTT1 (P = 0.039) genotypes and SK development were detected, with the null variants of both genes conferring an approximate 2-fold increase in risk for solar keratosis development (OR: 2.1; CI: 1.3-3.5 and OR: 2.3; CI: 1.0-5.0 for GSTM1 and GSTT1, respectively). For the GSTM1 gene, this risk was significantly higher in conjunction with high outdoor exposure (OR: 3.4; CI: 1.9-6.3) and although the GSTT1 gene showed a similar trend (OR: 2.9; CI: 1.1-7.7), this did not reach significance. The increased risk of SK development associated with these genes is likely due to a decreased ability of the skin to detoxify mutagenic compounds produced by UV-induced oxidative stress, and hence a decreased ability to protect against carcinogenesis. Implication of the GSTM1 and GSTT1 null variants in solar keratosis development prompted interest in analysis of gene expression changes in response to UV radiation. Due to the high homology of the GSTM1 gene with other GSTM genes, and therefore potential issues with primer specificity, the GSTT1 gene was focussed on for the expression studies. Real time reverse transcription PCR, incorporating SYBR green fluorescence and 18S as a comparative gene, was used to study GSTT1 gene expression changes in response to both UVA and combined UVA/UVB radiation. It was found that only short term UV radiation had an effect on GSTT1 expression changes, whereas no alteration of gene expression was seen after 4 and 12 hours of recovery from long term irradiation between irradiated and matched non-irradiated skin samples. This indicated that changes in gene expression for the GSTT1 gene apparently occur relatively quickly after exposure to UV radiation. Analysis of both UVA only and combined UVA/UVB short term irradiation indicated that an initial decrease in expression, followed by an increase was likely to represent translation into protein and subsequent transcription of mRNA, and in some cases a second decrease indicated further translation. Hence, it appears as though UV radiation does have a significant effect on the expression of at least one GST gene, and that UV radiation in combination with genetic variation of these genes may play a role in the development of NMSC. Finally, association and subsequent expression analysis was also performed on the vitamin D receptor. The hormonal form of vitamin D, 1a25 dihydroxyvitamin D3, has been shown to have numerous cancer-related effects, including antiproliferative, differentiation, proapoptotic and antiangiogenic effects. These effects are mediated through the binding of 1a25 dihydroxyvitamin D3 to the vitamin D receptor and subsequent transcriptional pathways. Polymorphisms within the VDR are known to regulate its transcription and therefore expression, which is linked to the ability of 1a25 dihydroxyvitamin D3 to bind. Association analysis of a 5 initiation codon variant (Fok I) and two 3 variants (Apa I and Taq I) was performed in SK affected and control populations. Although the Fok I variant showed no association with SK development, both the Apa I and Taq I variants were found to be associated with SK development (P = 0.043 and P = 0.012, respectively). In particular, the Aa and Tt genotypes were associated with increased risk of SK. These results were however more complicated, as shown by further analysis. This showed that genotypes containing at least one allele that conferred decreased VDR transcription (ie. AA/Aa and Tt/tt) increased risk of SK development by 2-fold in fair skinned individuals (OR: 2.1; CI: 1.2-3.7 and OR: 1.7; CI: 1.1-2.7 for Apa I and Taq I variants, respectively) but also found to decrease the risk of SK development by 2-fold in medium skinned individuals (OR: 0.5; CI: 0.3-1.0 for Apa I variants). Additionally, genotypes containing 2 alleles conferring decreased transcription of the VDR gene were found to further increase the risk for SK development in fair skinned individuals (OR: 2.5; CI: 1.4-4.5 and OR: 2.4; CI: 1.2-5.0 for Apa I and Taq I variants, respectively), indicating a possible additive effect for the alleles. The highly differential association of the VDR gene polymorphisms amongst phenotypes may reflect a combination between the ability of an individual to synthesise 1a25 dihydroxyvitamin D3 with the binding availability of the VDR. To further investigate the role of VDR in NMSC, expression analysis of the VDR gene was undertaken using real time reverse transcription PCR, with SYBR green fluorescence and 18S as a comparative gene, to examine expression pattern changes associated with UV radiation. It was found that short term irradiation, as well as long term irradiation and recovery were associated with gene expression changes. Short term irradiation resulted in patterns indicative of translation and subsequent transcription, whereas long term irradiated samples resulted in reduction of VDR expression that was recovered after an extended period of time. Thus, VDR expression is clearly influenced by UV exposure. It would be very interesting to see more specifically if particular VDR genotypes, which appear to play a role in NMSC risk, also are affected differentially by UV exposure. It is possible that VDR expression is reduced to limit excessive binding of 1a25 dihydroxyvitamin D3, although since both UVA and UVB radiation affect VDR expression, this may not be mediated the effect of 1a25 dihydroxyvitamin D3 but rather a different pathway resulting from a general UV response. In summary, the detection of a number of susceptibility genes involved in SK development and their subsequent expression analysis in response to UV radiation has given further insight into the molecular changes associated with NMSC. In fact, both detoxification genes (GSTM1 and GSTT1) and a transcription related gene (VDR), were found to confer susceptibility to solar keratosis, an early stage skin lesion with tumourigenic potential. This suggests that even the earliest stages of skin cancer are mediated through a wide range of effects. Additionally, expression changes related to these genes indicate that they are associated with the well known environmental carcinogen of UV radiation and that their effects may be mediated through a wide range of pathways. Although implication of the 18q region in SCC progression was not confirmed in this study, it is still likely to play a role in malignant transformation. The implication of this region, as well as the implication of susceptibility genes has vastly increased knowledge into processes associated with NMSC. Although additional analysis can confirm and further implicate these molecular alterations, this study has resulted in a more comprehensive understanding of NMSC that may ultimately be of benefit in terms of prognosis and treatment.
| Identifer | oai:union.ndltd.org:ADTP/194946 |
| Date | January 2004 |
| Creators | Carless, Melanie, n/a |
| Publisher | Griffith University. School of Health Science |
| Source Sets | Australiasian Digital Theses Program |
| Language | English |
| Detected Language | English |
| Rights | http://www.gu.edu.au/disclaimer.html), Copyright Melanie Carless |
Page generated in 0.003 seconds