Human Molecular Genetics, 2002, Vol. 11, No. 12 1449-1453
© 2002 Oxford University Press
Promoter switch: a novel mechanism causing biallelic PEG1/MEST expression in invasive breast cancer
1Department of Pathology, Conway Institute of Biomolecular and Biomedical Research, University College Dublin, Belfield, Dublin 4, Ireland, 2Department of Histopathology and 3Department of Surgery, Mater Hospital, Dublin 7, Ireland and 4Molecular Medicine Unit, University of Leeds, St James's University Hospital, Leeds LS9 7TF, UK
Received February 11, 2002; Accepted March 27, 2002
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
We have previously reported on the biallelic expression of the imprinted PEG1/MEST gene in infiltrating carcinomas of the breast. Putative loss of imprinting (LOI) of PEG1/MEST has subsequently also been implicated in the aetiology of lung adenocarcinomas and colon cancer. Taking advantage of our previous study, identifying seven infiltrating carcinomas of the breast, displaying biallelic PEG1/MEST expression, we have analysed the allelic usage of the two alternative PEG1/MEST transcripts encoding isoforms 1 and 2, separately. In addition, expression levels of the two transcripts have been measured by real-time RT–PCR, in order to elucidate the mechanism behind the switch from monoallelic transcription in normal breast tissue to biallelic expression in invasive cancer. The isoform 1 transcript is imprinted in both the paired normal tissue and the breast carcinomas. In contrast, the isoform 2 transcript is biallelically expressed, or in one case expressed from the opposite allele to isoform 1, raising the possibility that isoform 2 is polymorphically imprinted in normal breast tissue. In all the paired normal samples, isoform 1 is predominantly expressed, explaining the monoallelic profiles of these samples. However, in four of the seven biallelic carcinomas, isoform 2 is expressed at higher levels than isoform 1, indicating that a switch in expression from isoform 1 to isoform 2 is responsible for the biallelic profiles in these samples. Our results not only suggest a novel mechanism leading to biallelic expression detected when analysing the common 3'-UTR of the PEG1/MEST transcriptional unit, they are also indicative of the existence of further alternative PEG1/MEST transcripts.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
Genomic imprinting is a mechanism by which a number of genes throughout the genome are monoallelically expressed according to their parental origin. Loss of imprinting (LOI) in the form of reactivation of a previously silent/imprinted allele of the growth-promoting gene IGF2 and inactivation of H19 was first detected in Wilms' tumours (1,2) and has since been reported in several adult cancers, including breast cancer (3–5). In vitro reversal of LOI of both H19 and IGF2 has suggested good therapeutic potential (6). In addition, LOI of IGF2 has been proposed to be an easily detectable premalignant marker of colorectal cancer (7,8). Therefore, the investigation of the allelic usage of imprinted loci is of great importance.
Since our original report documenting biallelic expression of PEG1/MEST in infiltrating carcinomas of the breast (9), an alternative non-imprinted PEG1/MEST transcript has been isolated from human adult blood lymphocytes (10). The two alternative transcripts have been denoted isoform 2 (non-imprinted) and isoform 1 (paternally expressed) respectively (10), based on the assumption that they encode different isoforms of the PEG1/MEST protein. This nomenclature has been adopted in the current study.
Taking advantage of our previous work, identifying seven informative tumours displaying biallelic PEG1/MEST expression, we have investigated LOI versus promoter switch as the possible mechanisms underlying biallelic PEG1/MEST expression in breast cancer. For this purpose, a protocol allowing the allelic usage of each transcript to be evaluated separately was employed. This, in conjunction with real-time RT–PCR analysis of the relative expression levels of isoforms 1 and 2, has provided evidence for promoter switch being a novel mechanism resulting in biallelic PEG1/MEST expression in breast cancer.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
Utilizing isoform-specific primers (Fig. 1), allelic usage of isoforms 1 and 2 was investigated separately. Analysis of the paired samples from patients with biallelic PEG1/MEST expression, established that isoform 1 remained imprinted in the cancer tissue (Fig. 2). Hence, isoform 1 has not undergone LOI, indicating that an alternative mechanism must be involved in the biallelic expression.
|
|
Isoform 2 was biallelically expressed in six out of the seven biallelic samples (T4, T11, T18, T39, T53 and T98) (Fig. 3B). The corresponding normal tissue also expressed isoform 2 biallelically (Fig. 3A) in five of the six cases (one case, N18, could not be analysed owing to there being insufficient RNA). This indicates that biallelic isoform 2 expression is not a result of LOI. However, in one case (T44), isoform 2 was expressed from the opposite allele of isoform 1 (Fig. 3B). This case also displayed monoallelic isoform 2 expression in the normal tissue (Fig. 3A), indicating the possibility that isoform 2 is polymorphically imprinted in breast tissue.
|
The monoallelic PEG1/MEST expression reported previously (9), when investigating the common 3'-UTR directly, reflects the allelic usage of isoform 1. This indicates that isoform 1 is the predominant transcript in normal breast tissue. Owing to the fact that isoform 2 is expressed either biallelically or from the opposite allele to isoform 1, the possibility exists that the biallelic profiles of the breast carcinomas detected originally (9) is a result of a switch in expression from isoform 1 to isoform 2. To investigate this further, we established the relative expression levels of the two isoforms by isoform-specific real-time RT–PCR.
In all normal samples analysable, isoform 1 mRNA levels were at least 10-fold higher than isoform 2 levels (Table 1). In two benign lesions with monoallelic PEG1/MEST profiles, the relative expression of isoform 1 remained high compared with isoform 2 (data not shown). In contrast, two of the biallelic carcinomas (T4 and T53) displayed higher isoform 2 than isoform 1 mRNA levels, and in two samples (T39 and T44) the expression levels in the cancer tissue were comparable. In conjunction with the allelic-usage results, this suggests that the biallelic PEG1/MEST profiles could be a result of promoter switch in four of the seven biallelic infiltrating carcinomas (Table 1). In three samples, the mRNA levels of isoform 2 remained substantially lower than the isoform 1 levels, and since isoform 1 remains monoallelically expressed, the biallelic profiles must be attributed to an as-yet unidentified transcript. The possibility of another transcript arising from the PEG1/MEST transcription unit indicates that the expression pattern at this locus maybe as complex as that reported for the GNAS1 locus (11–13).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
In the current study, we found isoform 1 to be imprinted in both normal and breast cancer tissue. In contrast, isoform 2 was biallelically expressed in six cases and expressed from the opposite allele to isoform 1 in one case. Whether this indicates that isoform 2 is reciprocally imprinted to isoform 1 in a proportion of breast carcinomas will have to be investigated further. The possibility exists that the observed monoallelic expression is a result of structural changes in the form of a mutation present in both the normal and tumour tissues of this patient. Since isoform 1 transcription can be detected from the opposite allele, the structural change must be limited to the region upstream of exon 1.
Real-time PCR established that isoform 1 is the predominantly expressed transcript in normal breast tissue, whereas isoform 2 is expressed at comparable or higher levels than isoform 1 in four out of the seven carcinomas.
Based on these results, we can conclude that the biallelic PEG1/MEST profiles detected originally in infiltrating carcinomas are not a result of LOI of isoform 1. Three biallelic carcinomas retain predominant expression of the imprinted isoform 1. In these three cases, the originally detected biallelic expression has to be attributed to another transcript. COPG2 is a gene orientated in tail-to-tail fashion with PEG1/MEST, and the 3'-UTR of the two genes overlap with 52 bp (14). However, the RFLP utilized in this study is not situated within this overlap. Hence, COPG2 expression does not interfere with the imprinting analysis. Therefore, as-yet unidentified transcript(s) must account for the biallelic profiles of these three cases. In the four remaining carcinomas, where isoform 2 expression levels are either comparable to or higher than isoform 1 levels, we propose that a switch in expression to isoform 2 is contributing to (or causing) the biallelic profiles.
It is important to establish whether the differential expression of isoform 2 is caused by differences in cellular composition of carcinomas and normal tissue, or is the result of an increase in isoform 2 levels in specific cell types. We have detected both isoform 1 and isoform 2 expression in primary cultures of breast epithelial cells (data not shown). In addition, isoform 2 is expressed in blood lymphocytes (10). However, we know from our previous study that the infiltrating tumour lymphocytes are unlikely to influence the expression profiling (9). Expression analysis of different cell types microdissected from breast tissue will give a more detailed knowledge of the mechanism(s) leading to elevated isoform 2 mRNA levels in infiltrating carcinomas.
It is difficult to suggest what the overall effect of the observed differential expression of the PEG1/MEST transcripts might be. In normal development, isoform 1 is predominantly expressed in embryonic tissues, whereas expression of the two isoforms is comparable in adult blood lymphocytes (15). The expression pattern observed for human and murine PEG1/MEST by mRNA in situ hybridization indicates that PEG1/MEST is an oncofetal neoangiogenesis factor (16). However, the probe utilized was part of the 3'-UTR. Therefore differentiation between isoforms 1 and 2 was not possible. Although Kosaki et al. (10) denoted the two alternative PEG1/MEST transcripts, isoforms 1 and 2, uncertainty exists as to whether the putative proteins are indeed isoforms or unrelated proteins. Conceptional translation shows that isoform 1 encodes a 335-amino-acid protein of the 
/ß hydrolase fold family (17). Exon 1, which is specific to isoform 1, is a coding exon. The full-length cDNA sequence of isoform 2 is not known. There is a start codon in exon 2 inframe with the reading frame of isoform 1. If this codon is the translational start site for isoform 2, the alternative protein product would be a PEG1/MEST isoform lacking the first nine amino acids of isoform 1. However, the start codon is not in a good context. On the other hand, there are stop codons within exon A or immediately at the beginning of exon 2 for all three reading frames. Therefore, if the translational start site of isoform 2 is located 5' to the currently known sequence of exon A, the two transcripts would encode different proteins. Further investigation of the coding potential of isoform 2 is of great importance in establishing the effect of the observed switch in expression.
The effect of differential expression of PEG1/MEST transcripts between normal tissue and cancer still has to be established. Evidence is accumulating that it may be a frequent occurrence in several different cancer forms. Since our original report of biallelic PEG1/MEST expression in breast cancer, PEG1/MEST has been reported to be biallelically expressed in lung adenocarcinomas and colon cancer (18,8). In lung cancer, 11 out of 13 (85%) informative samples have been reported to have biallelic PEG1/MEST expression. Lymphocytic infiltrate was ruled out as cause of the observed profiles, but the two isoforms were not investigated separately (18). Hence, the possibility exists that the mechanism leading to biallelic expression is similar to what we have observed for breast cancer. In colon cancer, 7 out of 20 samples displayed biallelic PEG1/MEST expression, and, of these, 5 of 7 matched-normal mucosa also showed biallelic expression of PEG1/MEST (8). LOI of isoform 1 was ruled out as the mechanism. On the other hand, isoform 2 was reported to have undergone LOI in the tissues with biallelic expression.
Future studies will focus on the isolation of other transcripts arising from the PEG1/MEST transcriptional unit, and the investigation of their imprinting status and coding potential. This would further elucidate the mechanism behind the biallelic expression observed when analysing the 3'-UTR of PEG1/MEST and also contribute to establishing the effect of this frequent occurrence in breast and lung cancer.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
Tissue samples
Seven PEG1/MEST biallelic infiltrating carcinomas (four infiltrating ductal carcinomas, one papillary carcinoma, one mucinous carcinoma and a squamous cell carcinoma) and paired normal breast tissue were analysed in this study. Six of the seven samples originated from the cohort described in our previous work (9). One additional biallelic sample was included. This sample was T44, an infiltrating ductal carcinoma with a high-grade ductal carcinoma in situ component. The lymphocytic response in this sample was moderate and lymphovascular invasion was present.
Ethical approval had been obtained in the acquisition of the samples, conforming to the Declaration of Helsinki.
Isoform-specific allelic-usage analysis
Isoform-specific evaluation of PEG1/MEST allelic usage was carried out utilizing a protocol similar to that previously described by Nishihara et al. (8).
Aliquots of 1 µg total RNA were DNase-treated using DNaseI (GibcoBRL). cDNA was synthesized using the Superscript II reverse transcription kit (GibcoBRL) and oligo dT primer (GibcoBRL). For each sample, a RT-negative control was set up to check for contaminating genomic DNA. In order to facilitate the most efficient generation of full-length cDNA, synthesis was carried out for 2 hours at 42°C.
RT–PCR reactions were set up using reverse primer R2324 (5'-GACTAAGACAATGAAATGTGGC-3') in combination with forward primer exon 1 (5'-GCGGCGGGCGGCATGGG-A TA-3') or exon A (5'-CCTGTAGGCAAGGTCTTACCTG-3'), generating isoform 1- or isoform 2-specific products respectively. In each reaction, 2 µl oligo dT-primed cDNA was used as template. The reaction volume was 50 µl, containing 1 µM of each primer, 2.5 mM MgCl2, 200 nM dNTP mixture, 1x TaKaRaLA Taq buffer and 2.5 units TaKaRaLA Taq enzyme (TaKaRa). The reactants were subjected to 5 minutes initial denaturing at 95°C, followed by 35 cycles (95°C for 30 seconds, 55°C for 1 minute, 72°C for 3 minutes), with a final extension for 10 minutes at 72°C. Cycling was carried out on a DNA Engine (PTC-2000).
Each of the isoform 1 and isoform 2 products was gel-purified separately using the Geneclean kit (Bio 101) according to manufacturer's instructions. The purified RT–PCR products were subsequently used as template in a nested PCR and digested with AflIII restriction enzyme (GibcoBRL) as previously described (9,16). The restriction profiles were evaluated by densitometry (9).
Real-time PCR
Real-time PCR was performed using an ABI7700 sequence detection system (PE Applied Biosystems) in the presence of SYBR-green. This fluorochrome incorporates stoichiometrically into the amplification product, providing real-time quantification of double-stranded DNA PCR product. Primers for each alternative transcript were designed, to amplify an 80–120 bp fragment with 59°C annealing temperature (Primer Express, PE Applied Biosystems). The reverse primer was identical for both transcripts (exon 2: 5'-GAAGACTTCCATGAGTGAAGGGC-3'). A forward primer was designed for each transcript: isoform 1 (exon 1b: 5'-GATAACGCGGGCCATGGTG-3'), and isoform 2 (exon A: 5'-CCTGTAGGCAAGGTCTTACCTG-3'). Optimization of the real-time PCR reaction was performed according to the manufacturer's instructions. For each analysis, transcription of the gene of interest was compared with transcription of the housekeeping gene GAPDH, which was amplified in parallel (GAPDH-F: 5'-AACAGGGACACCCACTCCTC-3' and GAPDH-R: 5'-CATACCAGGAAATGAGCTTGACAA-3').
|
ACKNOWLEDGEMENTS |
|---|
The work was supported by the Danish Research Council and the Danish Cancer Society.
|
FOOTNOTES |
|---|
* To whom correspondence should be addressed. Tel: +353 1 7162811; Fax: +353 1 2692016; Email: amanda.mccann{at}ucd.ie
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
1 Rainier, S., Johnson, L.A., Dobry, C.J., Ping, A.J., Grundy, P.E. and Feinberg, A.P. (1993) Relaxation of imprinted genes in human cancer. Nature, 362, 747–749.[Medline]
2 Ogawa, O., Eccles, M.R., Szeto, J., McNoe, L.A., Yun, K., Maw, M.A., Smith, P.J. and Reeve, A.E. (1993) Relaxation of insulin-like growth factor II gene imprinting implicated in Wilms' tumour. Nature, 362, 749–751.[Medline]
3
McCann, A.H., Miller, N., O'Meara, A., Pedersen, I., Keogh, K., Gorey, T. and Dervan, P.A. (1996) Biallelic expression of the IGF2 gene in human breast disease. Hum. Mol. Genet., 5, 1123–1127.
4 Yballe, C.M., Vu, T.H. and Hoffman, A.R. (1996) Imprinting and expression of insulin-like growth factor II and H19 in normal breast tissue and breast tumour. J. Clin. Endocrinol. Metab., 81, 1607–1612.[Abstract]
5 Wu, H-K., Squire, J.A., Catzavelos, C.G. and Weksberg, R. (1997) Relaxation of imprinting in human insulin-like growth factor II gene, IGF2, in sporadic breast carcinomas. Biochem. Biophys. Res. Commun., 235, 123–129.[Web of Science][Medline]
6
Barletta, J.M., Rainier, S. and Feinberg, A.P. (1997) Reversal of loss of imprinting in tumour cells by 5-aza-2'-deoxycytidine. Cancer Res., 57, 48–50.
7 Cui, H., Horon, I.L., Ohlsson, R., Hamilton, S.R. and Feinberg, A.P. (1998) Loss of imprinting in normal tissue of colorectal cancer patients with microsatellite instability. Nat. Med., 4, 1276–1280.[Web of Science][Medline]
8 Nishihara, S., Hayashida, T., Mitsuya, K., Schulz, T.C., Ikeguchi, M., Kaibara, N. and Oshimura, M. (2000) Multipoint imprinting analysis in sporadic colorectal cancers with and without microsatellite instability. Int. J. Oncol., 17, 317–322.[Web of Science][Medline]
9
Pedersen, I.S., Dervan, P.A., Broderick, D., Harrison, M., Miller, N., Delany, E., O'Shea, D., Costello, P., McGoldrick, A., Keating, G. et al. (1999) Frequent loss of imprinting of PEG1/MEST in invasive breast cancer. Cancer Res., 59, 5449–5451.
10 Kosaki, K., Kosaki, R., Craigen, W.J. and Matsuo, N. (2000) Isoform-specific imprinting of the human PEG1/MEST gene. Am. J. Hum. Genet., 66, 309–312.[Web of Science][Medline]
11
Hayward, B.E., Kamiya, M., Strain, L., Moran, V., Campbell, R., Hayashizaki, Y. and Bonthron, D.T. (1998) The human GNAS1 gene is imprinted and encodes distinct paternally and biallelically expressed G proteins. Proc. Natl Acad. Sci. USA, 95, 10038–10043.
12
Hayward, B.E., Moran, V., Strain, L. and Bonthron, D.T. (1998) Biderectional imprinting of a single gene: GNAS1 encodes maternally, paternally, and biallelically derived proteins. Proc. Natl Acad. Sci. USA, 95, 15475–15480.
13
Hayward, B.E. and Bonthron, D.T. (2000) An imprinted antisense transcript at the human GNAS1 locus. Hum. Mol. Genet., 9, 835–841.
14
Blagitko, N., Schulz, U., Schinzel, A.A., Ropers, H-H. and Kalscheuer, V.M. (1999) 
2-COP, a novel imprinted gene on chromosome 7q32, defines a new imprinting cluster in the human genome. Hum. Mol. Genet., 8, 2387–2396.
15 Kobayashi, S., Uemura, H., Kohda, T., Nagai, T., Chinen, Y., Naritomi, K., Kinoshita, E., Ohashi, H., Imaizumi, K., Tsukahara, M. et al. (2001) No evidence of PEG1/MEST gene mutations in Silver–Russell syndrome patients. Am. J. Med. Genet., 104, 225–231.[Web of Science][Medline]
16 Mayer, W., Hemberger, M., Frank, H-G., Grümmer, R., Winterhager, E., Kaufmann, P. and Fundele, R. (2000) Expression of the imprinted genes MEST/Mest in human and murine placenta suggests a role in angiogenesis. Dev. Dyn., 217, 1–10.[Web of Science][Medline]
17
Kobayashi, S., Kohda, T., Miyoshi, N., Kuroiwa, Y., Aisaka, K., Tsutsumi, O., Kaneko-Ishino, T. and Ishino, F. (1997) Human PEG1/MEST, an imprinted gene on chromosome 7. Hum. Mol. Genet., 6, 781–786.
18 Kohda, M., Hoshiya, H., Katoh, M., Tanaka, I., Masuda, R., Takemura, T., Fujiwara, M. and Oshimura, M. (2001) Frequent loss of imprinting of IGF2 and MEST in lung adenocarcinoma. Mol. Carcinogen. 31, 184–191.[Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  P. Sathyan, H. B. Golden, and R. C. Miranda Competing Interactions between Micro-RNAs Determine Neural Progenitor Survival and Proliferation after Ethanol Exposure: Evidence from an Ex Vivo Model of the Fetal Cerebral Cortical Neuroepithelium J. Neurosci., August 8, 2007; 27(32): 8546 - 8557. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. M. Valleley, S. F. Cordery, and D. T. Bonthron Tissue-specific imprinting of the ZAC/PLAGL1 tumour suppressor gene results from variable utilization of monoallelic and biallelic promoters Hum. Mol. Genet., April 15, 2007; 16(8): 972 - 981. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Hikichi, T. Kohda, T. Kaneko-Ishino, and F. Ishino Imprinting regulation of the murine Meg1/Grb10 and human GRB10 genes; roles of brain-specific promoters and mouse-specific CTCF-binding sites Nucleic Acids Res., March 1, 2003; 31(5): 1398 - 1406. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||