Mutagenesis vol. 19 no. 1 pp. 3-11,
January 2004
© 2004 UK Environmental Mutagen Society/Oxford University Press
REVIEW The restriction site mutation (RSM) method: clinical applications
Swansea Clinical School, University of Wales Swansea, Singleton Park, Swansea SA28PP, UK
 |
Abstract |
|---|
 Top Abstract IntroductionCollection of tissue samplesDNA extractionRSM procedureHelicobacter pylori detection in...TP53 mutations in pre-malignant...TP53 mutations in pre-malignant...TP53 mutations in pre-malignant...TP53 mutations in pre-malignant...Genetic stability of...Types and prevalence of...TP53 mutation as a...ConclusionsReferences |
|---|
The restriction site mutation (RSM) method has been developed over the past 13 years as a sensitive mutation test which can detect mutations in restriction sites in any gene. Due to the fact that 5/8 of the main mutation hotspots in the TP53 gene fall within restriction sites, RSM can analyse them for the presence of rare mutations (1 mutation in 10 000 non-mutated copies). After validating the usefulness of RSM in detecting mutagen-induced mutations, we recently turned our attention to looking for TP53 mutations in pre-malignant tissue. We show here that RSM can detect early TP53 mutations in pre-malignant tissue of the oesophagus, stomach, colon and bladder. We can also use these clinical mutation data to speculate as to the causative mutagens involved in these cancer conditions. We here use an example of mutations detected in gastric tissue and those induced in vitro by hydrogen peroxide.
|
Introduction |
|---|
|
TopAbstractIntroductionCollection of tissue samplesDNA extractionRSM procedureHelicobacter pylori detection in...TP53 mutations in pre-malignant...TP53 mutations in pre-malignant...TP53 mutations in pre-malignant...TP53 mutations in pre-malignant...Genetic stability of...Types and prevalence of...TP53 mutation as a...ConclusionsReferences |
|---|
Somatic mutations are induced continually in human cells. These mutations have the potential to cause disease, including cancer. In order to induce cancer, the mutation must provide the cell (and hence clone) with a selective growth advantage. This growth advantage is exploited during tumour evolution and is accompanied by the accumulation of further mutations which aid its further selection. Mutations are accumulated in such cells due to spontaneous processes as well as by exposure to exogenous mutagens. Spontaneous mutations arise in every cell of every animal at a very low level, at a frequency of
10–9 mutations/base/division (Wabl, 1982
). Substantially more important in mutagenesis and carcinogenesis are induced mutations, which can be introduced into the DNA of cells at levels 103 times higher than spontaneous mutations (10–6). These induced mutations are believed to be responsible for cancer development in most cases. Examples of paired culprit mutagen and tumour type include cigarette smoke and lung cancer and sunlight and skin cancer. Whether mutations arise by spontaneous or induced processes, their frequency within affected tissues is very important in terms of carcinogenesis. Very rare cancer-related mutations (before clonal expansion) are probably present in all tissues of all adult individuals (Frank and Nowak, 2003). It is only when clonal expansion occurs that phenotypic effects are noted. In the case of carcinogenesis, these phenotypic effects are life threatening.
Not all mutations are created equal and the differences in mutations (types and positions) present in clinical tissues can provide important information on the culprit mutagens. Mutagens are known to induce characteristic mutation patterns (mutation fingerprints) in DNA, hence allowing their subsequent identification by DNA sequence methods. These characteristic mutation fingerprints are a consequence of preferential DNA adduct sites and the influence of DNA repair and, in the case of carcinogenesis, the mutation fingerprint is influenced by the selective nature of specific mutations. The tissue-specific nature of point mutations as a consequence of tissue-specific mutagens (e.g. skin and UV) has been demonstrated, indeed ‘concordance’ between mutation patterns can be seen in tissues with similar mutagen exposure potentials (Lutz et al., 1998).
Methodologies for the detection of mutations in tumour samples where the mutations have been clonally amplified and are often present in every cell are numerous and pose no challenge. However, detecting mutations in pre-malignant tissues is somewhat difficult. Hence, the paucity of mutation data in pre-malignant tissues. The reason for the lack of early mutation data is the lack of methodologies capable of detecting low frequency mutations hidden amongst an excess of non-mutated sequences. Another aspect of mutation detection involves the gene in which mutations are studied. In mutagen-exposed model systems there are many selection-based mutation tests which can detect rare mutations in target genes and, hence, identify mutagens; examples of these systems include HPRT (Albertini et al., 1982), Lac I (Kohler et al., 1991), Lac Z (Gossen et al., 1989) and Sup F (Kraemer and Seidman, 1989). However, these systems are of little use in detecting mutations in clinical material, particularly in cancer-related genes such as TP53 (see below).
Restriction site mutation (RSM) is a methodology which has been developed by ourselves and others for the detection of low frequency mutations in any gene (Parry et al., 1990; Jenkins et al., 1997, 1998, 1999a,b, 2000, 2001). The development of this methodology has now progressed to the point that we can detect mutations induced in vitro by exposure to specific mutagens. Importantly, we can now also apply RSM to look for the same mutations in the clinical material of patients who may have been exposed to this specific mutagen. This may prove a powerful approach in establishing the origin of clinical mutational events and may help in providing direct evidence for mutagen-induced carcinogenesis. Importantly, if mutations can be detected in cancer-related genes (such as TP53) early in cancer development, it may be possible to use such early mutations as prognostic markers of cancer development. This is particularly true for mutations which drive cancer progression and hence are selected for during tumour evolution. It is well known that early detection is the key to better survival in carcinogenesis (Etzioni et al., 2003).
RSM, as the name suggests, relies on the sequence specificity of bacterial restriction enzymes. These enzymes cleave DNA at 4–6 base sequences (Pingoud and Geltsch, 2001), but fail to recognize these sequences if a single base is changed. Hence, if a mutation falls within a restriction enzyme (RE) site, it will alter the ability of the enzyme to cleave it. By digesting the DNA with a particular RE, followed by PCR with primers flanking the RE site, only undigested (mutated) DNA will escape digestion and produce a PCR product. This is the basis of RSM (see Figure 1). In practice, a second digestion step removes any non-mutated DNA which may have escaped the initial digestion (for a review see Jenkins et al., 1999b). Hence, any PCR products generated can be sequenced to reveal the RE site mutation. Using a large number of RE sites within a particular gene allows mutation to be assessed across large tracts of sequences. RSM can be tailored to suit particular mutagens/mutagen-exposed tissues, by choosing RE sites that contain bases prone to specific mutations (e.g. RE sites rich in GC bases or AT bases or containing CpG sites, etc.). Obviously, the main drawback of the RSM approach is that mutations can only be studied in DNA sequences covered by a RE site. However, most genomes have RE sites covering 50% of the DNA (Cotton, 1989). Hence, despite the fact that a lot of REs are not suitable for RSM, RE site coverage is still reasonably good in human genomic DNA.
|
Initial work in developing RSM focused on determining its sensitivity by using known mutagenic exposures in vitro and in vivo. These studies demonstrated that the RSM approach could detect mutations induced by dimethylhydrazine (Jenkins et al., 1997
), N-methyl-N-nitrosourea (Suzen et al., 1998), 4-nitroquinoline-1-oxide (Jenkins et al., 1998), benzo[a]pyrene (Sueiro et al., 2000), 2-acetylaminofluorene (Jenkins et al., 2000), N-ethyl-N-nitrosourea (Jenkins et al., 1999a; Song et al., 2001) and hydrogen peroxide (Jenkins et al., 2001). These mutation studies were carried out in a variety of species (human, mouse, rat and flounder). This allowed optimization of the methodology and allowed us to gain experience in choosing suitable REs for RSM analyses. RSM has been shown to be capable of detecting mutational events even when the mutation is present amongst a >10 000-fold excess of non-mutated DNA (Jenkins et al., 2001). Figure 2 illustrates the sensitivity of RSM by using tumour DNA bearing a codon 175 mutation in TP53. A similar methodology developed by the Cerutti group produced a more sensitive version of the method (1:1 000 000) by undertaking preparative digestions to increase the gene copy numbers analysed (Aguilar et al., 1993). However, this approach was extremely time consuming and complex, hence we have used a simpler, less time consuming version in our studies. After extensive development of RSM, we became aware that RSM was not going to be accepted for standard genotoxicity testing, which had been our initial aim. This was a consequence of other established mutation tests being just as (if not more) effective, e.g. the Ames test (Ames et al., 1973), HPRT, Lac I and Lac Z. Hence, we made the decision in 1999 to exploit the sensitivity of RSM in examining clinically important mutations. Due to the fact that RSM can detect rare mutations in genes such as TP53, it was reasonable to assume that it could contribute substantially to the understanding of early mutagenesis in clinical tissues. For the clinical studies, the quantitative element of RSM, i.e. the determination of mutation frequencies, was not employed, as we were instead more interested in the presence or absence of a TP53 mutation. However, the strength of the mutated band is noted, as strong bands reflect prominent mutations whereas weak bands reflect rarer mutations. In this review, we focus on the use of RSM in detecting clinically important mutations in pre-malignant tissues. The RSM method has been employed by other groups for this purpose and there are published data on its use in detecting TP53 mutations in inflammatory colonic tissue (Ambs et al., 1999; Perwez Hussain et al., 2000) and in oral tissues (Mollaoglu et al., 2001).
|
This review focuses on mutations of the TP53 gene in particular. Although we have used RSM to analyse mutations in other cancer-related genes, most of our studies have concentrated on the TP53 gene. The main reason for this is the unique role of p53 in cancer development. TP53 is known to be the most frequently mutated gene in most (if not all) tumour types (Hainaut and Hollstein, 2000
). The TP53 gene is mutated most frequently in cancer due to its central role in the control of the cell cycle and apoptosis (Cadwell and Zambetti, 2001). The anti-proliferative properties of p53 are a huge hurdle to tumour evolution and this is the reason why so many tumours exist with defective p53 proteins. Indeed, tumours that lose p53 are known to become genetically unstable, allowing the accumulation of further cancer-promoting genetic alterations. Specifically, p53 loss is linked to increased spontaneous mutation rates (Havre et al., 1995), increased chromosome instability (Bouffler et al., 1995) and aneuploidy (Fukasawa et al., 1996). There has been substantial investigation of TP53 mutation induction in tumours such that databases (Hernandez-Boussard et al., 1999) of the types and positions of these mutations are now available (www.iarc.fr). This is a huge bonus for mutation analysis as it indicates the common positions of induced mutations (so-called hotspots). The mutation hotspots arise as a consequence of the most effective inactivating amino acid substitutions which remove p53 function. Figure 3 illustrates the location of TP53 mutations in all tumour types present in the IARC database, demonstrating their non-random distribution. These data allow the design of RSM targets (RE sites) that match these hotspots most closely. The fact that TP53 mutations are not concentrated at one key codon, but are spread over five or six codons provides the possibility of determining mutation fingerprints in clinical tissues for comparison with mutagen-induced fingerprints in vitro. However, with RSM this can only be partially achieved, as only a small number of RE sites and hence DNA bases are studied. In addition, the TP53 hotspots tend to centre on those containing CpG sites, thus further affecting the chance of obtaining an exact mutation fingerprint.
|
In this review we describe the accumulation of TP53 mutations in four pre-malignant tissues. Importantly, these tissues were histologically evaluated and found to be free from cancer. The four tissues are oesophageal, gastric, colon and bladder tissues. We have detected TP53 mutations in these tissues for two main reasons. Firstly, to determine how early in cancer progression TP53 mutations occur and, secondly, to identify what type of TP53 mutations occur with a view to comparing these with putative mutagenic exposures.
|
Collection of tissue samples |
|---|
|
TopAbstractIntroductionCollection of tissue samplesDNA extractionRSM procedureHelicobacter pylori detection in...TP53 mutations in pre-malignant...TP53 mutations in pre-malignant...TP53 mutations in pre-malignant...TP53 mutations in pre-malignant...Genetic stability of...Types and prevalence of...TP53 mutation as a...ConclusionsReferences |
|---|
Tissue samples were collected from patients attending endoscopy/colonoscopy/cystoscopy clinics at Morriston Hospital Swansea, Singleton Hospital Swansea and the University Hospital of Wales Cardiff, respectively. The samples were taken from consenting patients only after ethical approval was obtained from the local ethics committee. In the case of oesophageal and gastric biopsies, these were obtained by endoscopy in Professor J.N. Baxter’s clinics at Morriston Hospital Swansea. As well as biopsies from the affected area (Barrett’s oesophagus, gastritis, etc.), normal biopsies were also taken from non-affected areas as an internal control. Biopsies were also taken for concurrent histological analysis. In the case of the colorectal samples, tissue was obtained during colonoscopy or surgery for colorectal cancer during Mr J. Beynon’s surgical lists at Singleton Hospital Swansea. Adjacent normal tissue was also obtained. In the case of the clam ileocystoplasties, tissue samples were taken during cystoscopy in Mr T. Stephenson’s surveillance clinics at the University Hospital of Wales Cardiff, for patients who had previously received a clam cystoplasty. As well as tissue from the clam cystoplasty, adjacent tissue from the normal bladder remnant was also taken. The numbers of tissues analysed in these studies were as follows: 93 oesophageal samples from 42 patients; 52 gastric samples from 52 patients; 75 colon samples from 55 patients; 76 bladder samples from 38 patients.
|
DNA extraction |
|---|
|
TopAbstractIntroductionCollection of tissue samplesDNA extractionRSM procedureHelicobacter pylori detection in...TP53 mutations in pre-malignant...TP53 mutations in pre-malignant...TP53 mutations in pre-malignant...TP53 mutations in pre-malignant...Genetic stability of...Types and prevalence of...TP53 mutation as a...ConclusionsReferences |
|---|
DNA was extracted from the biopsies using a high salt method (Stratagene, Cambridge, UK). The DNA was checked for quantity and quality by spectrophotometry at 260/280 nm. The DNA concentration was adjusted to 100 ng/µl and stored at –20°C.
|
RSM procedure |
|---|
|
TopAbstractIntroductionCollection of tissue samplesDNA extractionRSM procedureHelicobacter pylori detection in...TP53 mutations in pre-malignant...TP53 mutations in pre-malignant...TP53 mutations in pre-malignant...TP53 mutations in pre-malignant...Genetic stability of...Types and prevalence of...TP53 mutation as a...ConclusionsReferences |
|---|
Mutation analysis was performed at codons 175, 213, 248, 249 and 282 of TP53. Codon 273, which is a major TP53 mutation hotspot, was not available for mutation analysis as there is no restriction enzyme coverage. RSM analysis was performed as previously described (Jenkins et al., 2002, 2003
). Mutated RSM products were sequenced using a CEQ2000 automated DNA sequencer (Beckman Coulter, High Wycombe, UK). Both strands of the PCR products were sequenced and mutations were only accepted if present on both strands. |
Helicobacter pylori detection in gastric biopsies |
|---|
|
TopAbstractIntroductionCollection of tissue samplesDNA extractionRSM procedureHelicobacter pylori detection in...TP53 mutations in pre-malignant...TP53 mutations in pre-malignant...TP53 mutations in pre-malignant...TP53 mutations in pre-malignant...Genetic stability of...Types and prevalence of...TP53 mutation as a...ConclusionsReferences |
|---|
DNA extracted from gastric biopsies with gastritis histology was subject to PCR determination of H.pylori presence and subtype. PCR primers designed for the flagellin gene were used to identify H.pylori presence and primers for the CagA virulence factor were used to assign CagA positivity. Details of the PCR conditions were as described elsewhere (Morgan et al., 2003
). |
TP53 mutations in pre-malignant oesophageal tissue |
|---|
|
TopAbstractIntroductionCollection of tissue samplesDNA extractionRSM procedureHelicobacter pylori detection in...TP53 mutations in pre-malignant...TP53 mutations in pre-malignant...TP53 mutations in pre-malignant...TP53 mutations in pre-malignant...Genetic stability of...Types and prevalence of...TP53 mutation as a...ConclusionsReferences |
|---|
Barrett’s oesophagus is a pre-malignant condition linked to chronic reflux (heartburn) (Jankowski et al., 1999
). Patients who reflux bile and stomach acid for many years (or decades) may eventually succumb to oesophageal adenocarcinoma, a highly aggressive tumour with very low survival rates (Shahin and Murray, 1999). Patients progress to cancer through a number of distinct histological stages, beginning with inflammation (oesophagitis), followed by metaplasia, low grade dysplasia, high grade dysplasia and, finally, adenocarcinoma. TP53 abnormalities are known to be involved in the progression of Barrett’s tissues to cancer (Jenkins et al., 2002), indeed, tumours carrying TP53 mutations are thought to be more aggressive (Schneider et al., 2000). Hence, we studied the histological progression of this condition for the presence of early TP53 mutations (Jenkins et al., 2003) using RSM. Figure 4a shows the results of this investigation and demonstrates the progressive accumulation of TP53 mutations during the histological progression to cancer (Jenkins et al., 2003), suggesting that these mutations may be involved in driving cancer by providing a selective advantage to cells. We are now aiming to see if TP53-positive patients progress along the histological path to cancer faster than those without a TP53 mutation. If so, then early TP53 mutation, as detected here by RSM, may represent a useful prognostic marker in Barrett’s patients. Table I contains the details of the PCR and digest conditions, whilst Tables II and III contain the TP53 mutation data from this study. Table II confirms that these mutations were detectable early in cancer progression, being present at the metaplasia stage. The mutation types detected in this study are discussed below in terms of potential mutagens known to produce such mutation patterns.
|
|
|
|
|
TP53 mutations in pre-malignant gastric tissue |
|---|
|
TopAbstractIntroductionCollection of tissue samplesDNA extractionRSM procedureHelicobacter pylori detection in...TP53 mutations in pre-malignant...TP53 mutations in pre-malignant...TP53 mutations in pre-malignant...TP53 mutations in pre-malignant...Genetic stability of...Types and prevalence of...TP53 mutation as a...ConclusionsReferences |
|---|
Gastric cancer has been linked to infection of the stomach with the bacterium H.pylori (International Agency for Research on Cancer, 1994
). Infection with H.pylori leads to a histological progression starting with gastritis, leading through intestinal metaplasia and dysplasia to, finally, cancer. As with Barrett’s oesophagus, TP53 mutations are known to be involved in gastric cancer (Uchino et al., 1993) and hence we sought TP53 mutations in clinical biopsies of gastric tissue with a range of histologies. We employed RSM to exploit this technique’s sensitivity to determine the histological stage harbouring the first TP53 mutations. We also aimed to link the TP53 mutations found in the gastritis samples to concurrent infection with H.pylori, in order to investigate the role that H.pylori plays in gastric cancer progression. For the H.pylori association only gastritis samples were analysed, as it is known that H.pylori infection is reduced as the tissue histology progresses to metaplasia and beyond. Figure 4B shows the accumulation of TP53 mutations during histological progression in gastric cancer (Morgan et al., 2003). Tables II and III again contain the data on the TP53 mutations found and show that TP53 mutations were detectable early (gastritis stage). There was no clear link between TP53 mutation and H.pylori infection. This may have been due to the fact that most of the gastritis patients were H.pylori-positive (16/20), hence it may have been difficult to assign a link due to the lack of H.pylori-negative patients (6/7 TP53-mutated patients were positive for H.pylori). When the H.pylori-positive patients were broken down into those carrying the virulence factor CagA (Morgan et al., 2003) it was found that four of the six patients positive for H.pylori who contained TP53 mutations also carried this virulence factor, whereas 2/6 did not; this association was not statistically significant (P = 0.5) and the samples were too small to make any firm associations. As with the Barrett’s oesophagus patients, we are aiming to determine if the presence of an early TP53 mutation is a prognostic marker by following up these sets of patients to monitor their histological progression. |
TP53 mutations in pre-malignant colon tissue |
|---|
|
TopAbstractIntroductionCollection of tissue samplesDNA extractionRSM procedureHelicobacter pylori detection in...TP53 mutations in pre-malignant...TP53 mutations in pre-malignant...TP53 mutations in pre-malignant...TP53 mutations in pre-malignant...Genetic stability of...Types and prevalence of...TP53 mutation as a...ConclusionsReferences |
|---|
Colorectal cancer is often used as a model of cancer formation, as its adenoma–carcinoma sequence is well established. In addition, from Vogelstein’s well-known genetic pathway of colorectal cancer (Fearon and Vogelstein, 1990
), it is accepted that TP53 mutation is a late event in carcinogenesis of the colon, occurring at the carcinoma stage. However, we aimed to see if by using a sensitive mutation test we could detect TP53 mutations in a subset of pre-cancerous colon cells prior to carcinoma formation. Previous studies of TP53 mutation in colorectal cancer development have used relatively insensitive techniques to detect mutations, such as DNA sequencing. Hence, we analysed colon polyps of varying sizes in order to examine the stage at which TP53 mutations first arose. Our results demonstrate that TP53 mutations were indeed detectable in polyps, albeit at a low level (7%), so perhaps TP53 mutations are earlier events than previously thought (Williams et al., 2002). There was no correlation in this study between the size or histology of the polyp and the presence of a TP53 mutation. Figure 4C shows the distribution of TP53 mutations in the colonic tissues analysed. Again, the types and number of mutations detected are shown in Tables II and III. It should be noted that this colorectal study was only performed with one of the codons of TP53, codon 248. This was due to time limitations in this particular study and due to the fact that in all other studies this codon accumulated most mutations. Hence, this study may underestimate the proportion of tissues with a TP53 mutation by not studying codons 175, 213, 249 and 282 (20–75% of TP53 mutations in the three other tissues were outside codon 248). Furthermore, a comparison of the codon preference for TP53 mutations in colon tissue is obviously not possible. |
TP53 mutations in pre-malignant bladder tissue |
|---|
|
TopAbstractIntroductionCollection of tissue samplesDNA extractionRSM procedureHelicobacter pylori detection in...TP53 mutations in pre-malignant...TP53 mutations in pre-malignant...TP53 mutations in pre-malignant...TP53 mutations in pre-malignant...Genetic stability of...Types and prevalence of...TP53 mutation as a...ConclusionsReferences |
|---|
Clam ileocystoplasty is an operation that has been traditionally used to increase bladder volume and reduce bladder pressure in groups of patients who suffer from incontinence (Bramble, 1982
). The operation involves taking a piece of small bowel and inserting it into the centre of the bladder (hence the coining of the term ‘clam’). This operation, whilst reasonably successful at treating incontinence, has led to increased cancer rates in the inserted bowel region or the anastomosis (stitch line) of clam cystoplasties (Filmer and Spencer, 1990). The increased bladder cancer rates in these patients has an unknown basis, but in order to assess whether this is due to increased genetic instability in the transplanted tissue, TP53 mutations were sought in the bowel tissue and bladder remnant of patients who had undergone clam ileocystoplasties. The results of this study showed that the inserted bowel/stitch line was indeed unstable and a number of patients possessed TP53 mutations (K.D.Ivil, G.J.S.Jenkins, E.M.Parry, S.A.Jenkins, J.M.Parry and T.P.Stephenson, in preparation). Figure 4D illustrates the accumulation of TP53 mutations in the clam ileocystoplasty. The types and locations are again shown in Tables II and III. As mentioned earlier, the finding of TP53 mutations in pre-malignant bladder tissues raises the possibility of using these data to predict cancer progression in such patients. |
Genetic stability of histologically normal tissue |
|---|
|
TopAbstractIntroductionCollection of tissue samplesDNA extractionRSM procedureHelicobacter pylori detection in...TP53 mutations in pre-malignant...TP53 mutations in pre-malignant...TP53 mutations in pre-malignant...TP53 mutations in pre-malignant...Genetic stability of...Types and prevalence of...TP53 mutation as a...ConclusionsReferences |
|---|
Interestingly, in all four tissue types studied, there were no detectable TP53 mutations in the histologically normal tissue (101 normal tissue samples analysed). Despite the sensitivity of RSM (1 mutant sequence in >10 000 non-mutated sequences), no such mutations were detectable. This suggests that TP53 mutations, if present at all, must be present at frequencies beyond the scope of RSM analysis (<10–5) before clonal expansion has occurred. This suggests that normal tissues do not accumulate TP53 mutations frequently and hence should be genetically stable.
|
Types and prevalence of TP53 mutations in clinical specimens |
|---|
|
TopAbstractIntroductionCollection of tissue samplesDNA extractionRSM procedureHelicobacter pylori detection in...TP53 mutations in pre-malignant...TP53 mutations in pre-malignant...TP53 mutations in pre-malignant...TP53 mutations in pre-malignant...Genetic stability of...Types and prevalence of...TP53 mutation as a...ConclusionsReferences |
|---|
The TP53 mutations identified in these studies show some similarities as well as some interesting differences. As alluded to earlier, analysis of the mutation patterns in clinical material offers the possibility of identifying the causative mutagenic exposure. Most of the mutations induced in these four tissues were GC
AT transitions (Table III). The highest level of transversions was seen in the bladder tissues (14% GCCG), which may represent exposure to a mutagen capable of causing bulky DNA adducts which are more prone to transversion events (Denissenko et al., 1998). Nitrosamines have been implicated in bladder carcinogenesis (Mirvish, 1995), in particular in those receiving a clam cystoplasty (Nurse and Mundy, 1989). Hence, they may play a role in inducing the mutations described here, but they tend to induce more GCAT mutations rather than transversions (Mirvish, 1995). The involvement of an exogenous mutagen in the bladder mutations is supported by the fact that only 25% of the bladder GCAT mutations were at CpG sites, as opposed to 60–100% in the other three tissues (Table III).
Due to the high levels of CpG mutations in the three tissues of the gastrointestinal tract studied here, it is possible that these mutations may have been induced spontaneously as a consequence of increased proliferation of the pre-malignant tissue (even normal gastrointestinal tracts are hyperproliferative in order to replace the gut lining regularly). The high level of GCAT mutations at CpG sites in the gastrointestinal tract tissues is particularly evident in the oesophageal and colon tissues (Table III). CpG sites are known to be genetically unstable (when methylated) and readily lead to GCAT mutations, especially in proliferating tissues, where there is less time for DNA repair. However, it should be borne in mind that CpG sites are also preferential binding sites for some mutagenic chemicals (Denissenko et al., 1997; Chen et al., 1998), hence, these CpG site mutations could be produced by mutagen exposure. Indeed, it has been suggested that CpG sites are preferential targets for reactive oxygen species (ROS) (Harris, 1998), known to be produced during inflammation and inflammation-mediated carcinogenesis (Jackson and Loeb, 2001). Indeed, nitric oxide (NO), produced in inflamed tissues, has been implicated in CpG mutagenesis in the TP53 gene (Ambs et al., 1999). However, in a previous study of ours, only 20% of H2O2-induced mutations were found at CpG sites (Jenkins et al., 2001), therefore, we believe that these CpG mutations are probably spontaneous in origin.
Determining the association between mutagen exposure and clinical mutation is complicated due to several factors, including the overlapping specificities of spontaneous mutations and mutagen-specific mutations (often GCAT mutations). However, the profiles of the mutations (type plus position) may be informative in suggesting links between mutagens and clinically manifested mutations. As an example, Figure 5 shows a comparison between the TP53 mutations detected in the oesophageal and gastric tissue to those induced by H2O2 in vitro. This comparison was made due to the inflammatory origin of the two cancer-prone conditions (Perwez Hussain and Harris, 2000), hence, the possible role for inflammation-mediated ROS such as H2O2 and NO. From Figure 5 it is possible to see that the types of mutations induced by H2O2 and those seen in gastric and oesophageal tissue are somewhat similar (mainly GCAT transitions). However, when the actual mutation profiles are compared (Figure 6), distinct differences appear between the two tissues and the in vitro study. Whilst in theory it is possible to link mutations in clinical material to those induced by suspect carcinogens in vitro, as mentioned earlier, selection may cloud the issue. The issue of mutation selection has previously been well discussed (Cooper and Krawczak, 1993; Krawczak and Cooper, 1996). Clinically important TP53 mutations will represent those that alter p53 function. However, mutations induced in vitro represent the mutation pattern minus selection. An example of this can be seen in Figure 6; a major hotspot for H2O2-induced mutations is the third base of codon 247. However, this will not normally cause amino acid substitutions and, hence, will be neutral; this is presumably the reason why this mutation is not detected in the clinical samples. In Figure 6, once the codon 247 hotspot for H2O2 is removed, there appears to be a possible similarity between gastric TP53 mutations and those induced by H2O2. The oesophageal mutations differ somewhat from the other two due to the lack of mutations at codon 249. Further study of the clinical mutations and those induced by ROS other than H2O2 would aid in understanding the contribution of inflammation-mediated ROS to upper gastrointestinal tract carcinogenesis.
|
|
|
TP53 mutation as a prognostic marker of cancer development |
|---|
|
TopAbstractIntroductionCollection of tissue samplesDNA extractionRSM procedureHelicobacter pylori detection in...TP53 mutations in pre-malignant...TP53 mutations in pre-malignant...TP53 mutations in pre-malignant...TP53 mutations in pre-malignant...Genetic stability of...Types and prevalence of...TP53 mutation as a...ConclusionsReferences |
|---|
It has been demonstrated here that TP53 mutations arise early in cancer development in these tissues. From Figure 4 it is possible to rank the tissues for TP53 mutation accumulation. This shows that gastric tissue accumulates more TP53 mutations than oesophageal, whilst oesophageal tissue accumulates more mutations than bladder and colon, respectively. In addition, the mutation data show that gastric tissue samples display the earliest TP53 mutations, with 35% of gastritis tissues (i.e. merely inflamed tissue) carrying mutated TP53 genes in at least 1 in 10 000 cells (the detection limit of RSM). Hence, clonal expansion of the TP53-mutated clone must have already occurred in this inflamed tissue in order to be detectable by RSM. The fact that TP53 mutations are present in inflamed gastric tissue confirms previous cytogenetic data from our laboratory, showing the unstable nature of gastric tissue in general (Williams et al., 2003
). This also begs the question, do patients with TP53-mutated cells progress to cancer faster than patients devoid of TP53 mutations? Only close follow-up of these cohorts of patients over 5 years will answer this question. This follow-up takes the form of monitoring the histological progression correlated with TP53 mutation data. Hence, a patient with gastritis who progresses to intestinal metaplasia (the next histological step) will be assessed for the presence of a TP53 mutation. If a TP53 mutation is shown to be linked to cancer progression in any of these tissues studied here, it may become useful to monitor the presence of TP53 mutations in pre-malignant tissues of a wide variety of tissue types. RSM is certainly a convenient method of assessing the presence of such mutations. It may prove to be that the cut-off point for mutation detection with RSM (1 in 10 000) may adequately separate high risk (post-expansion) cancer patients from low risk (pre-expansion) patients. |
Conclusions |
|---|
|
TopAbstractIntroductionCollection of tissue samplesDNA extractionRSM procedureHelicobacter pylori detection in...TP53 mutations in pre-malignant...TP53 mutations in pre-malignant...TP53 mutations in pre-malignant...TP53 mutations in pre-malignant...Genetic stability of...Types and prevalence of...TP53 mutation as a...ConclusionsReferences |
|---|
The studies described in this review have demonstrated that RSM can detect some of the hotspot TP53 mutations in pre-malignant tissues. This is important for several reasons. Firstly, it may identify high risk cancer patients and represent a useful prognostic marker. Secondly, the profiles of the TP53 mutations may allow speculation as to the causative mutagenic exposures which may ultimately lead to reduced exposures in the future.
RSM has been shown here to detect mutations in situations where previous studies have failed. This is a consequence of the sensitivity of RSM being much greater (1000-fold) than the sensitivity of standard methods used to detect DNA sequence changes (sequencing, SSCP, etc.). Hence, RSM may be a suitable tool for cancer research. Here we have described the detection of TP53 mutations, but one advantage of RSM is that it is readily applicable to other genes and, hence, early mutation detection in the APC gene (in colorectal cancer), Ras genes (in colorectal and pancreatic cancer), etc. is eminently possible. The data illustrated here on TP53 mutation detection may underestimate the total number of TP53 mutations in these tissues, due to the fact that mutations at codon 273, a frequent event in tumours (Figure 3), cannot be studied by RSM due to the lack of a RE site at this position. Extension of RSM analysis to this codon, either by the introduction of artificial RE sites at codon 273 or by the availability of new REs in the future, would increase the power of RSM to detect tumour-specific TP53 mutations.
We have shown here similarities between upper gastrointestinal tract TP53 mutation patterns and those induced experimentally by H2O2. This similarity was more marked for gastric tissue samples. Hence, this may suggest that the early stages of gastric cancer may be caused by ROS due to inflammation of the stomach, in part perhaps through infection with H.pylori. In the bladder, we can suggest that the lack of CpG mutations and the presence of transversions found here indicate exogenous mutagens, possibly nitrosamines. In the case of the colon tissues, not enough data are available to make any speculative estimates as to mutagen exposure.
The idea of pre- and post-expansion mutated cells explaining the detection rate of TP53 mutations by RSM is particularly interesting. It has recently been suggested that spontaneous mutations may be induced during the early growth and development phases of animals, as a consequence of their higher proliferation rates (Frank and Nowak, 2003), hence leaving young people with a clutch of pre-expansion mutations ready for clonal evolution. These early life mutations (which are an inevitability) do not pose a risk for cancer development, whereas once expansion occurs, leading to a large clone of mutated cells, there is a palpable risk. Given that DNA mutations can only be detected (even by sensitive methods like RSM) after clonal expansion, the threshold of detection (1:10 000) of RSM may prove to be useful in separating the expanded mutant cells that correlate with increased risk from pre-expansion cells, which pose a lower risk. This will only be shown after further studies are complete. Therefore, the fact that mutation systems are not sensitive may not be important so long as they can separate pre- and post-expansion mutations which directly modulate an individual’s cancer risk.
|
Acknowledgements |
|---|
I wish to thank Professor J.M. Parry for his help and advice over the course of these studies. In addition I would like to thank Claire Morgan, Gethin Williams and Ken Ivil who carried out the gastric, colon and bladder studies, respectively. Finally, thanks to the clinicians who provided the clinical samples, Professor J.N. Baxter, Mr J. Beynon, Dr A.P. Griffiths and Mr T. Stephenson.
|
Notes |
|---|
1Tel: +44 1792 295361; Fax: +44 1792 295447; Email: mailto:g.j.jenkins{at}swansea.ac.uk
*Recipient of the 2002 UKEMS Young Scientist Award and the 2003 EEMS Young Scientist Award
|
References |
|---|
|
TopAbstractIntroductionCollection of tissue samplesDNA extractionRSM procedureHelicobacter pylori detection in...TP53 mutations in pre-malignant...TP53 mutations in pre-malignant...TP53 mutations in pre-malignant...TP53 mutations in pre-malignant...Genetic stability of...Types and prevalence of...TP53 mutation as a...ConclusionsReferences |
|---|
-
Aguilar,F., Hussain,S.P. and Cerutti P. (1993) Aflatoxin B1 induces the transversion of G to T in codon 249 of the p53 tumor suppressor gene in human hepatocytes. Proc. Natl Acad. Sci. USA, 90, 8586–8590.
Albertini,R.J., Castle,K.L. and Borcherding,W.R. (1982) T-cell cloning to detect the mutant 6-thioguanine resistant lymphocytes present in human peripheral blood. Proc. Natl Acad. Sci. USA, 79, 6617–6621.
Ambs,S., Bennett,W.P., Merriam,W.G. et al. (1999) Relationship between p53 mutations and inducible nitric oxide synthase expression in human colorectal cancer. J. Natl Cancer Inst., 91, 86–89.
Ames,B.N., Lee,F.D. and Durston,W.E. (1973) An improved bacterial test system for the detection and classification of mutagens and carcinogens. Proc. Natl Acad. Sci. USA, 70, 782–786.
Bouffler,S.D., Kemp,C.J., Balmain,A. and Cox,R. (1995) Spontaneous and ionizing radiation induced chromosomal abnormalities in p53 deficient mice. Cancer Res., 55, 3883–3889.
Bramble,F.J. (1982) The treatment of adult enuresis and urge incontinence with enterocystoplasty. Br. J. Urol., 54, 693.[Web of Science][Medline]
Cadwell,C. and Zambetti,G.P. (2001) The effects of wild type p53 tumour supressor activity and mutant p53 gain of function on cell growth. Gene, 277, 15–30.[CrossRef][Web of Science][Medline]
Chen,J.X., Zheng,C.Y., West,M. and Tang,M. (1998) Carcinogens preferentially bind at methylated CpG in the p53 mutational hotspots. Cancer Res., 58, 2070–2075.
Cooper,D.N. and Krawczak,M. (1993) Human Gene Mutation. Bios Scientific Publishers, Oxford, UK.
Cotton,G.H. (1989) Detection of single base changes in nucleic acids. Biochem. J., 263, 1–10.[Web of Science][Medline]
Denissenko,M.F., Chen,J.X., Tang,M. and Pfeifer,G.P. (1997) Cytosine methylation determines hot spots of DNA damage in the human p53 gene. Proc. Natl Acad. Sci. USA, 94, 3893–3898.
Denissenko,M.F., Pao,A. and Pfeifer,G.P. (1998) Slow repair of bulky DNA adducts along the nontranscribed strand of the human p53 gene may explain the strand bias of transversion mutations in cancers. Oncogene, 16, 1241–1247.[CrossRef][Web of Science][Medline]
Etzioni,R., Urban,N., Ramsey,S., McIntosh,M., Schwartz,S., Reid,B., Radich,J., Anderson,G. and Hartwell,L. (2003) The case for early detection. Nature Rev. Cancer, 3, 243–252.[CrossRef][Web of Science][Medline]
Fearon,E.R. and Vogelstein,B. (1990) A genetic model for colorectal carcinogenesis. Cell, 61, 759–767.[CrossRef][Web of Science][Medline]
Filmer,R.B. and Spencer,J.R. (1990) Malignancies in bladder augmentations and intestinal conduits. J. Urol., 143, 671.[Web of Science][Medline]
Frank,S.A. and Nowak,M.A. (2003). Developmental predisposition to cancer. Nature, 422, 494.
Fukasawa,K., Choi,T., Kuriyama,R., Rulong,S. and Vande Woude,G.F. (1996) Abnormal centrosome amplification in the absence of p53. Science, 271, 1744–1747.[Abstract]
Gossen,J.A., de Leeuw,W.J.F., Tan,C.H.T, Zwarthoff,E.C., Berends,F., Lohman,P.H.M., Knook,D.L. and Vijg,J. (1989) Efficient rescue of integrated shuttle vectors from transgenic mice: a model for studying mutations in vivo. Proc. Natl Acad. Sci. USA, 86, 7971–7975.
Harris,C.C. (1998) Tumor suppressor genes: at the crossroads of molecular carcinogenesis, molecular epidemiology and cancer therapy. Lung Cancer, 18, 113.
Havre,P.A., Yuan,J., Hedrick,L., Cho,K.R. and Glazer,P.M. (1995) p53 inactivation by HPV16 E6 results in increased mutagenesis in human cells. Cancer Res., 55, 4420–4424.
Hainaut,P. and Hollstein,M. (2000) p53 and human cancer: the first ten thousand mutations. Adv. Cancer Res., 77, 81–137.[Web of Science][Medline]
Hernandez-Boussard,T., Rodriguez-Tome,P., Montesano,R. and Hainaut,P. (1999) IARC p53 mutation database: a relational database to compile and analyse p53 mutations in human tumors and cell lines. Hum. Mutat., 14, 1–8.[CrossRef][Web of Science][Medline]
23. International Agency for Research on Cancer (1994) IARC Monographs on the Evaluation of the Carcinogenic Risks to Humans. Vol. 61, Schistosomes, Liver Flukes and Helicobacter pylori. IARC, Lyon.
Jackson,A.L. and Loeb,L.A. (2001) The contribution of endogenous sources of DNA damage to multiple mutations in cancer. Mutat. Res., 477, 7–21.[Web of Science][Medline]
Jankowski,J.A., Wright,N.A., Meltzer,S.J., Triadafilopoulos,G., Geboes,K., Casson,A.G., Kerr,D. and Young,L.S. (1999) Molecular evolution of the metaplasia–dysplasia–adenocarcinoma sequence in the esophagus. Am. J. Pathol., 154, 965–973.
Jenkins,G.J.S. and Parry,J.M. (2000) Restriction Site Mutation (RSM) analysis of 2-acetylaminofluorene (2-AAF) induced mouse liver mutations and comparison with the measurement of in vivo micronucleus induction in the bone marrows of (2-AAF) treated mice. Teratog. Carcinog. Mutagen., 20, 107–117.[CrossRef][Web of Science][Medline]
Jenkins,G.J.S., Mitchell,I.deG. and Parry,J.M. (1997) Enhanced restriction site mutation analysis of 1,2-dimethylhydrazine induced mutations, using endogenous p53 intron sequences. Mutagenesis, 12, 117–123.
Jenkins,G.J.S., Chalestori,M.H., Song,H. and Parry,J.M. (1998) Mutation analysis using the restriction site mutation (RSM) assay. Mutat. Res., 405, 209–220.[Web of Science][Medline]
Jenkins,G.J.S., Takahashi,N. and Parry,J.M. (1999a) A study of ENU induced mutagenesis in the mouse using the restriction site mutation (RSM) assay. Teratog. Carcinog. Mutagen., 19, 281–292.[CrossRef][Web of Science][Medline]
Jenkins,G.J.S., Suzen H.S., Sueiro,R.A. and Parry,J.M. (1999b) The Restriction Site Mutation (RSM) assay. A review of the methodology development and the current status of the technique. Mutagenesis, 14, 439–448.
Jenkins,G.J.S., Morgan,C., Parry,E.M., Baxter,J.N. and Parry,J.M. (2001) The detection of mutations induced in vitro in the human p53 gene by a model Reactive Oxygen Species (ROS) employing the Restriction Site Mutation (RSM) Assay. Mutat. Res., 498, 135–144.[Web of Science][Medline]
Jenkins,G.J.S., Doak,S.H., Parry,J.M., D’Souza,F.R. Griffiths,A.P. and Baxter,J.N. (2002) Barrett’s oesophagus: the genetic pathways involved in its progression to adenocarcinoma. Br. J. Surg., 89, 824–837.[CrossRef][Web of Science][Medline]
Jenkins,G.J.S., Doak,S.H., Griffiths,A.P., Baxter,J.N. and Parry,J.M. (2003) Early p53 mutations in non-dysplastic Barrett’s tissues detected by the restriction site mutation (RSM) methodology. Br. J. Cancer, 88, 1271–1276.[CrossRef][Web of Science][Medline]
Kohler,S.W., Provost,G.S., Fieck,A. Kretz,P.L., Bullock,W.O., Sorge,J.A. Putman,D.L. and Short,J.M. (1991) Spectra of spontaneous and mutagen induced mutations in the lac I gene in transgenic mice. Environ. Mol. Mutagen., 18, 316–321.[Web of Science][Medline]
Kraemer,K.H. and Seidman,M.M. (1989) Use of supf, an Escherichia-coli tyrosine suppressor transfer-rna gene, as a mutagenic target in shuttle-vector plasmids. Mutat. Res., 220, 61–72.[Web of Science][Medline]
Krawczak,M. and Cooper,D.N. (1996) Single base pair substitutions in pathology and evolution: two sides to the same coin. Hum. Mutat., 8, 23–31.[CrossRef][Web of Science][Medline]
Lutz,W.K., Fekete,T. and Vamvakas,S. (1998) Position and base pair specific comparison of p53 mutation spectra in human tumors: elucidation of relationships betweens organs for cancer etiology. Environ. Health Perspect., 106, 207–211.[Web of Science][Medline]
Mirvish,S.S. (1995) Role of N-nitroso compounds (NOC) and N-nitrosation in etiology of gastric, esophageal, nasopharangeal and bladder cancer and contribution to cancer of known exposures to NOC. Cancer Lett., 93, 17–48.[CrossRef][Web of Science][Medline]
Mollaoglu,N., Wilson,M.J. and Cowpe,J.G. (2001) Extraction of DNA from cytological samples by scraping and smear method suitable for restriction site mutation analysis: a pilot study. Diagn. Cytopathol., 25, 83–85.[CrossRef][Web of Science][Medline]
Morgan,C., Jenkins,G.J.S., Ashton,T., Griffiths,P., Baxter,J.N., Parry,E.M. and Parry,J.M. (2003) The detection of p53 mutations in pre-cancerous gastric tissue using the RSM assay. Br. J. Cancer, 89, 1314–1319.[CrossRef][Web of Science][Medline]
Nurse,D.E. and Mundy,A.R. (1989) Assessment of the malignant potential of cystoplasty. Br. J. Urol., 64, 489.[Web of Science][Medline]
Parry,J.M., Shamsheer,M. and Skibinski,D. (1990) Restriction site mutation analysis, a proposed methodology for the detection and study of DNA base changes following mutagen exposure. Mutagenesis, 5, 209–212.
Perwez Hussain,S. and Harris,C.C. (2000) Molecular epidemiology and carcinogenesis: endogenous and exogenous carcinogens. Mutat. Res., 462, 311–322.[CrossRef][Web of Science][Medline]
Perwez Hussain,S., Amstad,P., Raja,K. et al. (2000) Increased p53 mutation load in noncancerous colon tissue from ulcerative colitis: a cancer prone chronic inflammatory disease. Cancer Res., 60, 333–337.
Pingoud,A. and Jeltsch,A. (2001) Structure and function of type II restriction endonucleases. Nucleic Acids Res., 29, 3705–3727.
Schneider,P.M., Stoeltzing,O., Roth,J.A. et al. (2000) P53 mutational status improves estimation of prognosis in patients with curatively resected adenocarcinoma in Barrett’s esophagus. Clin. Cancer Res., 6, 3153–3158.
Shahin,W. and Murray,J.A. (1999) Esophageal cancer and Barrett’s esophagus: how to approach surveillance, treatment and palliation. Postgrad. Med., 105, 111–127.[Medline]
Song,H., Jenkins,G.J.S., Ashby,J. and Parry,J.M. (2001) The application of the RSM assay to compare 1-ethyl-1-nitrosourea induced mutations between the endogenous p53 gene and the transgenic LacZ gene in MutaMouse testes. Mutagenesis, 16, 59–64.
Sueiro,R.A., Jenkins,G.J.S., Lyons,B.P., Harvey,J.S. and Parry,J.M. (2000) Benzo[a]pyrene induced mutagenicity depends on the sequence context in the flounder p53 gene. Mutat. Res., 468, 63–71.[Web of Science][Medline]
Suzen,S., Jenkins,G.J.S. and Parry,J.M. (1998) The application of the restriction site mutation technique to N-methyl-N-nitrosourea induced mutations in the rat. Teratog. Carcinog. Mutagen., 18, 171–182.[CrossRef][Web of Science][Medline]
Uchino,S., Noguchi,M. and Ochiai,A. (1993) P53 mutation in gastric-cancer—a genetic model for carcinogenesis is common to gastric and colorectal-cancer. Int. J. Cancer, 54, 759–764.[Web of Science][Medline]
Wabl,M., Burrows,P.D., von Gabain,A. and Steinberg,C. (1982) Hypermutation at the immunoglobulin heavy chain locus in a pre-B cell line. Proc. Natl Acad. Sci. USA, 82, 479–482.
Williams,G.L., Jenkins,G.J.S., Parry,J.M. and Beynon,J. (2002) The detection of early p53 mutations in colorectal polyps using the RSM method. Br. J. Surg., 86, S73.
Williams,L., Williams,J.G., Griffiths,A.P., Brown,T., Doak,S.H., Jenkins,G.J.S., Parry,E.M. and Parry,J.M. (2003) Interphase FISH to detect chromosomal abnormalities in gastric cancer progression. Gut, 52, A50.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  G. J.S. Jenkins, F. R. D'Souza, S. H. Suzen, Z. S. Eltahir, S. A. James, J. M. Parry, P. A. Griffiths, and J. N. Baxter Deoxycholic acid at neutral and acid pH, is genotoxic to oesophageal cells through the induction of ROS: the potential role of anti-oxidants in Barrett's oesophagus Carcinogenesis, January 1, 2007; 28(1): 136 - 142. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||