CpG site or CpG island, which is important - (Jun/02/2005 )
Hi, everyone
I am new here. I read through the whole forum in this section, but can not get an answer. So I bring it up here. It may be a naive and stupid question.
I've recently cloned a new gene, and find its mRNA is upregulated in most of colon cancer tissues (27/33 so far, I am still working on more patient samples). I want to figure out if the methylation play a role in the regulation of this gene expression. However, the situation is: I tried all of the three CpG island prediction program using defalut setting, and they all predict no CpG island in my promoter. Then I change the setting to less stringent, the program does find a CpG island, but it's more than 4Kb away from the 1st exon. There is some CpG sites near the 1st exon, but i don't know if their methylation status will affect the transcription. CpG site or CpG island, which one is more inportant? I attched the CpG pattern pf my promoter here. The last 65 nt is part of exon 1. Please give me some comments.
I have problem to attach the CpG pattern. The predict CpG island is 216bp
CMF,
CpG islands are more important, are there any CpG island on the 3' end,
also have you tried using the prediction programs with the settings reverse and complement on? I found that cpg islands appear with this on, i am not too sure why as the reverse complement should be the same as the input sequence 
with the island you have found, is it unique? it could well be a repetitive element and methylation of these have been shown to regulate genes in the mouse at least.
Nick
Hi, Nick,
Thanks, but how can I know if the CpG island is a repetitive element?
I am now trying to put the full gene sequence (61K) into the prediction program, I will update how it goes.
No CpG island in 3' end
But I did found a CpG island 771 bp right after 1st exon (%GC=64.7, 0.651, 515bp) , and another CpG island 5KB further down stream (%GC=73.3, 0.657, and 431bp). Both CpG islands are in intron I. The protein start codon is in Exon2. And some CpG islands in other introns. Which one to look at? Any coment?
Another question, Did anyone here sucessfully show that promoter hypomethylation is associated to mRNA upregualtion in cancer tissues or cells? any literature?
your cpg island sequence can be queried on repeatmasker here http://repeatmasker.org
sounds like you have a number of cpg islands to look at now. I would look at the largest one right after the first exon first.
and there are many many papers discribing hypomethylation of promoters of oncogenes say, that have resuultant upregulation of mRNA all the time!
good luck
nick
But I did found a CpG island 771 bp right after 1st exon (%GC=64.7, 0.651, 515bp) , and another CpG island 5KB further down stream (%GC=73.3, 0.657, and 431bp). Both CpG islands are in intron I. The protein start codon is in Exon2. And some CpG islands in other introns. Which one to look at? Any coment?
Another question, Did anyone here sucessfully show that promoter hypomethylation is associated to mRNA upregualtion in cancer tissues or cells? any literature?
I guess this should be the literature you are looking for, prostate cancer, hypermethylation, and up-regulation.
http://www.ncbi.nlm.nih.gov/entrez/query.f...9168&query_hl=3
Increased heparanase expression is caused by promoter hypomethylation and up-regulation of transcriptional factor early growth response-1 in human prostate cancer.
Ogishima T, Shiina H, Breault JE, Tabatabai L, Bassett WW, Enokida H, Li LC, Kawakami T, Urakami S, Ribeiro-Filho LA, Terashima M, Fujime M, Igawa M, Dahiya R.
Department of Urology, University of California-San Francisco and Veterans Affairs Medical Center, 4150 Clement Street, San Francisco, CA 94121, USA.
PURPOSE: Heparanase degrades heparan sulfate and has been implicated in tumor invasion and metastasis. The transcription factor, early growth response 1 (EGR1), is associated with the inducible transcription of the heparanase gene. We hypothesize that CpG hypomethylation in the heparanase promoter coupled with up-regulation of EGR1 levels may induce heparanase expression in human prostate cancer. EXPERIMENTAL DESIGN: Cultured prostate cancer cell lines (Du145, DuPro, LNCaP, and PC-3) with and without 5'-aza-2-deoxycytidine treatment, 177 prostate cancer samples, and 69 benign prostatic hyperplasia (BPH) samples were used. The frequency and level of heparanase promoter methylation were analyzed by methylation-specific primers which covered the core binding motif of EGR1 (GGCG) or SP1 (GGGCGG) or both. RESULTS: In cultured Du145, DuPro, LNCaP, and PC-3 cell lines, mRNA transcripts of heparanase were significantly increased after 5'-aza-2-deoxycytidine treatment, suggesting that promoter methylation was involved in the regulation of heparanase mRNA transcript. Significantly higher methylation was found in BPH samples than in prostate cancer samples (P < 0.0001), whereas mRNA transcripts of the heparanase gene were inversely lower in BPH samples than in prostate cancer samples (P < 0.01). EGR1 expression in prostate cancer tissues was significantly higher than in BPH tissues (P < 0.001) and correlated with heparanase expression (P < 0.0001). Moreover, multiple regression analysis revealed that up-regulation of EGR1 contributed significantly more to heparanase expression than did promoter CpG hypomethylation in prostate cancer samples (P < 0.0001). CONCLUSIONS: To our knowledge this is the first comprehensive study demonstrating that increased heparanase expression in prostate cancer tissues is due to promoter hypomethylation and up-regulation of transcription factor EGR1.
But I did found a CpG island 771 bp right after 1st exon (%GC=64.7, 0.651, 515bp) , and another CpG island 5KB further down stream (%GC=73.3, 0.657, and 431bp). Both CpG islands are in intron I. The protein start codon is in Exon2. And some CpG islands in other introns. Which one to look at? Any coment?
Another question, Did anyone here sucessfully show that promoter hypomethylation is associated to mRNA upregualtion in cancer tissues or cells? any literature?
En, I would suggest an control experiment:
1. locate all genes that are upregulated in >= 27/33 tissues;
2. do the CpG island prediction for all of them;
3. locate the overrepresented CpG island in this population, of which, YFG might also have (might be complicated, need a statitical test);
The idea is that if hypermethylation in the promoter is the reason for the up-regulation, most likely there is a transcription factor that is responsible for the binding, then other genes with the same cis-element will subject to the same regulation.
Hi CMF,
CpG sites don't make much sense if they are not considered in the context of CpG island.
The CpG island you found in 1st intron could be important to the transcription of your gene. Is the promoter sequence for your gene experimentally verified? Is the 1st exon a real exon and are there many EST sequences aligned to it (view your gene in UCSC or Ensembl genome browser you will find out)? If not, the 1st exon may not be a real one and the 1st intron may be the promoter or alternative promoter.
Yes, recently many papers (including papers from our lab) reporting that hypomethylation is responsible for overexpression of some genes in cancer.
You can prove your hypothesis by in vitro study in which you can demethylate no expressing cells with Aza-cytindine, or induce methylation in overexpressing cells with siRNA targeting the CpG island.
Good luck!
Thanks, Everyone. It's a super busy day. I will get back to you fantastic guys next monday! Have a good weekend.