TP53 and C-MYC Co-Alterations – A Hallmark of Oral Cancer Progression / SIMULTANA ALTERACIJA TP53 I C-MYC GENA – OBELEŽJE PROGRESIJE ORALNIH KARCINOMA

Summary Background: Head and neck squamous cell carcinoma, including oral cancer, is the sixth most common cancer worldwide. Despite advances in surgery and treatment, the 5-year survival rate has not improved significantly. There- fore, reliable molecular markers for oral cancer progression are badly needed. Methods: We conducted a copy number analysis to esti- mate amplification status of c-myc, cycD1 and EGFR onco- genes, mutational PCR-SSCP analysis to determine activa- tion of H-ras oncogene and inactivation of TP55 tumour suppressor gene and methylation specific PCR analysis to evaluate hypermethylation of p16 and MGMT genes. Results: c-myc oncogene was amplified in 56.7%, cycD1 in 20% and EGFR in 16.7% of Oral Squamous Cell Carci- noma (OSCC) cases while H-ras was activated in 33.3% of samples. Amplification of c-myc was significantly associat- ed with the tumour grade 2. Interestingly, EGFR and H-ras alterations were mutually exclusive. p16 and MGMT were inactivated by hypermethylation in 30% and 13.3% of cases. Co-alteration of cycD1 and p16 were not observed in any of the analyzed samples. TP53 was inactivated in 56.7% of samples and was significantly associated with progression of OSCC, grade 2 and stage 2. Moreover, TP55 and c-myc oncogene were simultaneously altered in grade 2 OSCC. Conclusions: The most promising marker of OSCC pro- gression remains the TP53 tumour suppressor, which is the most frequently mutated gene in oral cancers. Since there is synergism between TP55 and c-myc, it seems that co- alteration of these two genes could be also a good marker of OSCC progression from gradel to grade 2 tumours. Kratak sadržaj Uvod: Skvamocelularni karcinomi glave i vrata (HNSCC) uključujud i skvamocelularni karcinom usne duplje (OSCC) ubrajaju se u šest najčesdh tipova humanih maligniteta. Uprkos značajnim napredcima u hirurškom i terapijskom tretmanu, stopa petogodišnjeg preživljavanja kod ovog tipa maligniteta nije značajnije popravljena. Upravo zato, defin- isanje pouzdanih molekularjnih markera progresije kod OSSC predstavlja apsolutni prioritetet. Metode: Amplifikacioni status c-myc, cycD1 i EGFR gena od reden je pomoču eseja za detekciju broja genskih kopi- ja, aktivacija H-ras onkogena i inaktivacija TP53 tumor supresora odredena je PCR-SSCP mutacionom analizom, a hipermetilacija promotora p16 i MGMT gena je ispitana metil specifičnim PCR-om (MSP). Rezultati: Amplifikacija c-myc onkogena detektovana je kod 56,7%, cycD1 onkogena kod 20%, a EGFR onkogena kod 16,7% analiziranih oralnih skvamocelularnih carcino- ma. Istovremeno, mutaciona aktivacija H-ras onkogena detektovana je kod 33,3% ispitanih uzoraka. Amplifikovani c-myc, statisticki značajno korelira sa gradusom 2 OSCC. Posebno intrigantan je bio nalaz po kom se onkogene akti- vacije u EGFR i H-ras genu medjusobno isključuju. Hipermetilacija promotora p16 gena detektovana je kod 30%, a MGMT kod 13,3% analiziranih uzoraka. Ko-altera- cije cycD1 i p16 gena nisu zapažene ni u jednom od ana- liziranih uzoraka. Inaktivacija TP53 gena detektovana je kod 56,7% uzoraka i utvrdeno je da statistički značajno korelira sa gradusom 2 i statusom 2 OSCC. Pored ovoga, utvrdeno je da statistički značajan broj uzoraka gradusa 2, sa aktiviranim TP53 genom ima istovremeno aktiviran i c-myc onkogen. Zaključak: TP53, najčešče mutirani gen u oralnim karcino- mima, ostaje za sada i najpouzdaniji marker progresije kod OSCC. Obzirom na detektovani sinergizam izmedu TP53 i c-myc gena, možemo reči da su istovremene promene u ova dva gena još pouzdaniji pokazatelj progresije OSSC iz gradusa 1 u gradus 2.


Introduction
Head and neck squamous cell carcinoma (HNSCC), including oral cancer, is the sixth most common cancer, with an annual incidence of more than 500,000 cases worldwide (1). Oral squamous cell carcinoma (OSCC), which arises from the oral mucosa lining, is the most frequent malignant neoplasm of the head and neck region and accounts for over 90% of these tumours (2). This aggressive malignancy is associated with high mortality and with severe physical, emotional, and psychosocial damage among the rare long-term survivors. Despite advances in surgery and treatment, the 5-year survival rate for patients diagnosed with oral squamous cell carcinomas has not improved significantly and remains at about 50% (3,4). The prognosis of OSCC is influenced by many factors, such as performance status, TNM staging and pathological grading of differentiation. However, these factors are not sufficient in predicting the outcome. This motivates the search for factors with prognostic relevance in order to better tailor the individual management of OSCC patients. Recent research has focused on identification of molecular prognostic markers that could stage patients in more meaningful prognostic groups.
Oral carcinogenesis is a highly complex multifactorial multi-step process that is driven by the accumulation of multiple genetic and epigenetic alterations. Majority of them represent the activation of oncogenes and the inactivation of tumour suppressor genes, meaning that numerous oncogenes and tumour suppressor genes are implicated in oral carcinogenesis. The most thoroughly studied oncogenes implicated in the pathogenesis and prognosis of OSCC are epidermal growth factor receptor (EGFR), and members of gene families -ras, myc and cyclins. EGFR, the biological receptor of EGF and TGF-a, is frequently overexpressed in oral cancer and was shown to result from EGFR gene amplification in 30% of oral cancers (5). Similarly, c-myc oncogene, a transcription factor that regulates important processes such as stimulation of proliferation and inhibition of differentiation is found to be amplified in 20 to 40% of oral cancers (6,7). Some studies showed that members of the ras oncogene family are overexpressed in oral cancer, too. However, H-ras is usually activated by point mutations in codons 12, 13 and 61. The frequency of H-ras mutations in OSCC varies and ranges from 5% in Western countries (8) up to 35% in Asian population (9). Finally, cyclins are essential in controlling the cell cycle. Among cyclins, cyclin D1 appears to be the most strongly implicated in human oncogenesis since its overexpression plays an essential role in the transition between G1 and S phase in proliferating cells (10). Amplification of cyclin D1 gene was reported in 25-70% of oral cancers (11).
Among tumour suppressors the most widely studied genes in OSCC pathogenesis are TP53 and p16. Protein products of these genes control the cell cycle and are involved in the inhibition of cell proliferation. Structural and/or functional alterations in these genes produce uncontrolled cell proliferation. Therefore, it is not surprising that reported frequencies of TP53 alteration ranges from 20-90% in HNSCC, depending on the methodologies, type of tumour material and heterogeneity of tumour site examined (12).
However, the available evidence about the mole cular markers still remains inconclusive. Accordingly, there is a continuous need to define the most important biological markers for OSCC progression and invasiveness.
cases. Co-alteration of cycD1 and p16 were not observed in any of the analyzed samples. TP53 was inactivated in 56.7% of samples and was significantly associated with progression of OSCC, grade 2 and stage 2. Moreover, TP53 and c-myc oncogene were simultaneously altered in grade 2 OSCC. Conclusions: The most promising marker of OSCC progression remains the TP53 tumour suppressor, which is the most frequently mutated gene in oral cancers. Since there is synergism between TP53 and c-myc, it seems that coalteration of these two genes could be also a good marker of OSCC progression from grade1 to grade 2 tumours.
Klju~ne re~i: c-myc, onkogeni, progresija tumora, TP53, tumor supresori, skvamocelularni karcinom usne duplje The aim of this study is to provide additional information regarding molecular markers with the greatest potential for predicting clinical outcome in patients with OSCC. These include regulators of the cell cycle: p16, TP53 and cyclin D1, DNA repair gene MGMT and genes of the commonly activated signaling pathway in tumorigenesis, the EGFR/H-ras/c-myc pathway. In addition, we also describe a modification of a method for copy number analyses which was used to investigate amplification of c-myc and cycD1 oncogenes.

Tissue samples and DNA Extraction
Genomic DNA used in this study was isolated from 30 fresh frozen, surgically resected SCC of oral cavity collected at the Clinic of Maxillofacial Surgery, School of Dentistry, University of Belgrade. For each obtained tumour sample written consent and approval were acquired according to the ethical standards laid down in the1964 Declaration of Helsinki, the International Ethical Guidelines for Biomedical Research Involving Human Subjects (CIOMS), Geneva 1993, and the Guidelines for Good Clinical Practice CPMP/ICH/135/95), September 1997.
Genomic DNA was extracted from thirty frozen tumour tissue samples and five blood control samples according to the standard phenol/chloroform extraction procedure described by Sambrook and colleagues (13). The quality of the extracted DNA was verified by agarose gel electrophoresis and the concentrations were assessed spectrophotometrically. Isolated DNA was stored at +4 °C until further analyses.

Copy number analysis by quantitative real time PCR
In order to estimate amplification status of the two most often amplified oncogenes in human tumours including OSCC, c-myc and cycD1, a TaqMan based assay was created. We used RNase-P gene as the internal control (reference gene). The rationale for this was that RNase-P gene is a single copy gene and is therefore most suitable for normalization in copy number quantification analysis.
The experiments were set in 96-well plastic plates, with samples prepared in duplicates, in total reaction volume of 20 mL, with primers /probe ratio 3:1 (0.1 mmol/L probe : 0.3 mmol/L primers), 1x TaqMan Master Mix and 150 ng of tested DNA per reaction. Each plate contained at least two controls, DNAs isolated from the blood of healthy individuals. This control samples were used as calibrators. PCR reactions were carried out in the ABI Prism 7500 Sequence Detection System at 50 °C for 2 minutes, 95 °C for 10 minutes, followed by 40 cycles at 95 °C for 15 seconds, and 60 °C for 1 minute. The experimental threshold was calculated based on the mean baseline fluorescence signal from cycle 3 to 15 plus 10 standard deviations. The point at which the amplification plot crosses this threshold is defined as Ct, which represents the cycle number at this point and is inversely proportional to the number of target copies present in the initial sample. A mean value of each Ct duplicate was used for further calculation. Each run included a no-template control, as well. The obtained results were analyzed by RQ Study Add ON software for 7500 v 1.3 SDS instrument with a confidence level of 95% (p<0.05).

Differential PCR
Amplification status of EGFR oncogene was determined by duplex PCR reaction that engaged two pairs of primers, one for the target gene (EGFR) and the other for the reference gene (b-actin). For differential PCR analysis, a 110-bp fragment of the EGFR gene was co-amplified with a 168-bp fragment of bactin (ACTB) gene. The primer sequences were as follows: EGFR_F 5'-AGCCATGCCCGCATTAGC TC-3' and EGFR_R 5'-AACCCTTCAACGTAAGG AAA-3' for EGFR, and ACTB_F 5'-CTCTTTTCTTTCCCGATAG-GT-3' and ACTB_R 5'-CTCCAGCTTCTCGTAGGGTC -3' for the ACTB. D-PCR was performed in the total reaction volume of 25 mL with 150 ng of DNA, 1× PCR buffer (50 mmol/L KCl, 10 mmol/L Tris-HCl, pH 8.3, 0.01% gelatin), 1.5 mmol/L MgCl, 0.2 mmol/L each dNTP, 1 mmol/L each of four primers, 1U Taq -Polymerase. Thermal cycling included thirty repeats of denaturation at 95 °C/1 minute, annealing at 58°C /1 minute and extension at 72 °C/1 minute, with initial denaturation (95 °C/10 minutes) before and final extension (72 °C/10 minutes) after the repeating temperature steps. Generated PCR products were applied to 9% polyacrylamide gel for electrophoresis, stained with silver-nitrate, photographed and analyzed by Image Quant 5.2 by comparing the median pixel intensity in a given/selected area of two bands (EGFR and ACTB) in the same lane (sample). When median pixel intensity of EGFR band was equal or higher then 25% of median pixel intensity of ACTB, it was interpreted as gene amplification.

PCR-SSCP analysis of TP53 and H-ras genes
The most frequently mutated exons of TP53 tumour-suppressor gene (5-9) and sequences surrounding the hot-spot codons of H-ras oncogene (12,13 and 61) were amplified by PCR and subjected to further mutational screening by single strand conformation polymorphism (SSCP). Primers and PCR conditions are given in Table I, and conditions for SSCP mutational screening have been described earlier (14). In brief, amplicons were mixed with double volume of formamide containing 0.05% xylene cianol and 0.05% brom phenol blue, denatured at 95 °C for 5 minutes, cooled on ice for 1 minute and loaded on non-denaturating polyacrilamide gels (PAA). Con centration of PAA gels ranged from 7-12% depending on the amplicon length, and they were prepared with or without glycerol (10% in final) depending on the temperature at which electrophoresis was run. In order to avoid false positives or negatives, samples were examined under at least two different SSCP conditions, including different crosslink (%C=5.26; %C=3.6), temperature, ion charge, etc. Genomic DNA isolated from the blood of five healthy individuals was used as negative control.

Bisulfite modification of DNA and methylation-specific PCR
The methylation status of promoter region of p16 and MGMT genes was determined by methylation-specific PCR (MSP). Extracted genomic DNA was modified by bisulfite treatment according to procedure described by Herman et al (15). In brief, 500ng of total genomic DNA was denatured with 3 mol/L NaOH and bisulfate modification done overnight (eg. eighteen hours) at 55 °C. Besides tumour specimens, totally methylated and unmethylated human DNAs were always included in the bisulfite treatment procedure as controls. To estimate methylation status of the promoter regions of p16 and MGMT genes, a set of four primers per each gene was used: one pair for the detection of methylated CpG-cytosine and the other for the identification of unmethylated CpG-cytosine. Primers for p16 were as follows: p16_MF 5'-TTAT -TAGAGGGTGGGGCGGATCGC-3' and p16_MR 5'-GACCCCGAACCGCGACCGTAA-3' for methylated CpG-cytosine producing an amplicon 150 bp long,

Statistical analysis
Significant differences between the data sets were determined by STATISTICA 6.0 software (StatSoft, Inc., Tulsa, USA). The correlations between clinicopathological parameters and alterations of TP53, MGMT, p16, c-myc, cycD1 and EGFR genes were evaluated using Fisher exact test. Statistical differences were considered significant when P was < 0.05 (*) and P<0.01 (**).

Amplification of c-myc, cycD1 and EGFR oncogenes
We determined amplification status of c-myc and cycD1 oncogenes by Quantitative Real Time PCR using modification of a method described by Akervall and colleagues (16). Namely, we used RNase P as a reference gene for the normalization and performed quantification by 2 -DDCt method (17). DNA from healthy individuals was used as a calibrator. We interpreted results of quantification > 2 as an abnormality, signifying oncogene amplification. Obtained results are summarized in Table II. Specifically, c-myc oncogene was amplified in 17 patients out of 30 (56.7%) while cycD1 was amplified in 6 patients or 20% of cases.
Unlike c-myc and cycD1, amplification of EGFR oncogene was assessed by differential PCR (Figure 1) and was found to be amplified in only 5 patients (16.7%).
Further, we analyzed relationship between altered oncogenes and histopathological characteristics (Table II) of patients. Amplification of c-myc showed to be significantly associated with the grade 2 of the disease ( Table III). Alterations of cycD1 and EGFR have not revealed any association.

Mutational status of H-ras oncogene and TP53 tumour suppressor gene
H-ras oncogene is usually activated by point mutations in codons 12, 13 and 61. Therefore, we analyzed mutational status of H-ras oncogene by PCR-SSCP and obtained that it was activated in 33.3% of patients ( Table II). Analysis of relationship with histopathological characteristics did not reveal any significant association.
On the other hand, mutational analysis of TP53 revealed that it was inactivated in 17 out of 30 (56.7%) patients (Table III). Moreover, inactivation of TP53 was significantly associated with progression of tumour, grade 2 (Table IV) and stage 2 ( Table V).

Methylation status of p16 and MGMT tumour suppressor genes
Methylation status of p16 and MGMT was assessed by MSP-PCR ( Figure 2). Obtained results revealed that p16 was hypermethylated in 9 patients (30%) and MGMT only in 4 out of 30 (13.3%). No correlation was found with histopathological characteristics of the patients.

Relationship between alterations of analyzed genes
In order to reveal possible association among studied genes, we further analyzed whether there were co-alterations between any of them during the evolution of tumour. The only significant association we found was the one between c-myc oncogene and TP53 tumour suppressor gene in patients with grade 2 OSCC. Namely, Fisher exact test revealed that frequency of TP53 mutations was significantly higher in grade 2 tumours with amplified c-myc (Table VI). In other words, c-myc oncogene and TP53 tumour suppressor gene were simultaneously altered in grade 2 tumours of oral cavity. All remaining combinations revealed no co-alterations of any significance. However, it is interesting to notice that EGFR was never amplified in samples that have had mutated Hras. Similarly, amplification of cycD1 and inactivation of p16 have not been detected in the same patient. In other words, alterations in genes from these two sets are mutually exclusive.   Grade  pT stage  M1  1  1  0  0  1  1  0  G2  T3  M2  0  1  0  0  0  0  0  G1  T1  M3 1

Discussion
In the present study we investigated the prognostic value of frequently altered oncogenes and tumour suppressor genes in 30 cases of Oral Squamous Cell Carcinoma. To that end, we conducted a copy number analysis to estimate amplification status of c-myc, cycD1 and EGFR oncogenes, mutational analysis to determine activation of H-ras oncogene and inactivation of TP53 tumour suppressor gene and methylation analysis to evaluate methylation status of p16 tumour suppressor and MGMT DNA repair gene. Obtained alterations were correlated with each other and with histopathological parameters.
Copy number analysis revealed unusually high frequency of c-myc amplification (56.7%) in this cohort, which was considerably above previously reported 20 -40% (6)(7)(8). Tumorigenicity of the c-myc protein is ascribed to the promotion of cell proliferation and inhibition of apoptosis (18,19). Its overexpression in OSCC is the result of gene amplification and is usually associated with poorly differentiated tumours (20). Our results confirm this finding. Namely, amplification of c-myc oncogene was significantly associated with advanced tumours of grade 2 compared to grade 1, indicating its role in tumour progression. However, its correlation with survival of OSCC patients is not clear (21).
Contrary to c-myc, EGFR and cycD1 oncogenes were amplified with relatively low frequency, 16.7% and 20.0% respectively, in this set of samples. At first sight, this was an unexpected result because both of them are frequently amplified in OSCC (5,11). However, if we take into consideration mutational oncogenic activation of H-ras, which appeared at very high frequency for European population of 33.3%, the picture becomes more complex but clearer and even expected. Namely, H-ras is usually induced by EGFR, which mediates extracellular mitogenic stimuli. Downstream event of Ras activation may actually be induction of cycD1 and promotion of cell replication (22). Ras activates cycD1 promoter via an AP-1 like sequence (23) and achieves predominant CycD1 nuclear localization. As a consequence, the cells may acquire ability for non-adherent growth (24) or independent cell cycle progression (25). Thus, activation of either of these oncogenes could have the same tumorigenic consequence, that is continuous cell replication, since they act in the same signaling pathway. Therefore, oncogenic activation of all of them at the same time could not be justified from biological point of view and, maybe, we should consider them as a package. In favor of this is our finding that EGFR and H-ras were never altered in the same tumour sample. According to our results, alterations in these two oncogenes are mutually exclusive. That could be the explanation for such a low frequency of EGFR amplification. Lower frequency of cycD1 amplification (20%) then reported by other authors (11) could be the consequence of another, additional finding. Specifically, we have not observed co-alteration of cycD1 and p16 in any of the analyzed samples. This is important because cyclin D1 activation and p16 inactivation have the same effect, i.e., an increase in pRb phosphorylation and progression from G1 to S phase in the cell cycle (26). Hence, alteration of either of these genes will have the same outcome and there is no need for simultaneous changes in both. In other words, if p16 is inactivated, oncogenic activation of cycD1 is not necessary and would not be biologically rational.
One of the tumour suppressor genes most frequently involved in human OSCC is TP53 (27). We found that it was mutated in 56.7% of tumour samples, which is within reported frequencies (12). Inactivation of TP53 was significantly associated with more advanced primary tumours, grade 2 (p= 0.0004) and pT stage 2 (p=0.0298). Many reports support this result (12,28), but there are many which show that inactivation of TP53 is an early event (29,30). Opinions about this issue are contradictory. Moreover, we showed co-alteration of TP53 and cmyc, i.e., inactivation of TP53 significantly correlated with the amplification of c-myc in grade 2 tumours. The interactive involvement of c-myc and TP53 with cell proliferation and apoptosis, when altered by mutational events, seems to be a key regulatory element of oncogenesis and co-alteration of these two important genes is considered to be a relatively early    (31)).
In conclusion, despite major advances in our understanding of the molecular pathology of OSCC, numerous gaps have still to be filled. The most promising marker remains the TP53 tumour suppressor, which is the most frequently mutated gene in OSCC. Since there is synergism between TP53 and c-myc, it seems that co-alteration of these two genes could be also a good marker of OSCC progression from grade1 to grade 2 tumours.