northern blot, mRNA and the study of splicing - how can I study mRNA? (Feb/09/2007 )
Hi, I'm planning some experiments and as I don't have much experience in this I'd like to know if someone can give me a hand...
I'm planning to study the splicing of a protein (that has been the main item of investigation for the last years in the laboratory where I work), the idea came up after doing a NB and finding a larger second band, for the tissue we study, and even a third one in other tissues. The problem is that the NB is the only information we have, we only know that the probe (a fragment of the sequence that codes for the main protein) not only hibridizes with the known mRNA but also with other bigger fragment.........so we came with the following idea: first of all to get mRNA from the tissue of interest (is it possible to get it from a cell line culture from the tissue?), then to create the cDNA (is there any form to get rid of the mRNA I'm not intersted in?), then a gel and the only thing that occurred to me to do after that is to recover from the gel the DNA of the expected size (is it possible? or there might be too many fragments I'm not interested in??) and to do a "mini" library with them, so then by doing a colony blot or something like that I could recover the fragment of interest, sequence and clone it and finally used it to do others studies in vitro and in vivo.
so......the question is: is it possible to do it? is any of the steps ridicules???
The main point is to know the sequence difference -first- and then the functional differences between both transcripts. I first thought of designing primers based on the exon sequences from the known mRNA and use them to see if I get different amplicons but I don't if the primers will work or those fragments are absent in the second one!!...............I have to do something involving both mRNA I think....maybe if I could hibridized both of them I could find some differences....
well what do you think???
Hello,
After doing a northern I also had three additional signals, suggesting either alternative splicing or a familiy. Before I followed any experiment I blast in pubmed for sequence homologies and found already by that three other transcripts.
Since the genome is sequenced I think its easier to first look their. You will find for sure the sequence, though not yet identified. In my case, the sequences that matched the size of the detecte northern signal were putative proteins. Not yet identified or analysed. This computer work can take you a few days but believe me it will save you a lot of time instead of trying to identify your transcpripts by colony blots.
Once you have the sequence, you can design primer specific for these transcripts. But let me answer your question in work-flow.
1. Normally you can use a cell line to isolate RNA and to do a reverse transcrpition to obtain a cDNA library. It worked very well in my case. I'm studying acute-phase response in liver, and used liver cancer cell line. But if you prefere to not risk any transcriptional changes between your tissue in-vivo and a cell line you should isolate RNA from tissue if you have acces to it. (I assume you do since you did with it a northern). It's not really more work than isolate it from cells. You also don't really need mRNA. Normally reverse transcription works well with total RNA. Depending on the Kit you can put as much as 5µg total RNA (Corresponding to 100-250ng mRNA)
2. To get rid off the RNA after Reverse Transcription can be easily done by treating your cDNA with RNAse A. But in my opinion it's not a necessary step. I never did it and had no problems with it.
3. As I said before, to find your corresponding transcript will take you a lot of time and luck. If you were lucky in the database. And you should be, since all transcripts are there. Or at least you will find a homology in the genome. So based on that I woul design primers and do the second specific PCR.
4. Once you have your bands just sequence it.
By that way we discovered 10 new proteins. So I hope this helped you a bit. I know it sounds boring but nowadays scientist cannot avoid bioinformatics. It's boring and annoying but really helpful,
2.
[quote name='Vibrio' date='Feb 9 2007, 07:57 PM' post='87542']
Hello,
After doing a northern I also had three additional signals, suggesting either alternative splicing or a familiy. Before I followed any experiment I blast in pubmed for sequence homologies and found already by that three other transcripts.
By that way we discovered 10 new proteins. So I hope this helped you a bit. I know it sounds boring but nowadays scientist cannot avoid bioinformatics. It's boring and annoying but really helpful,
2.
Hi! thank you very much for the information......it looks like that my case is quite similar to yours! I'm working in pancreas response during acute phase of pancreatitis. 
In fact, one of the NB bands corresponds to a protein differentially expressed in stress response. It has been the basic element of my laboratory investigation during the past years.
So I have all the information of these protein (the smaller transcript in the NB), by searching for homologous sequences in pubmed, you mean to use the whole sequence of these one to search the database?
What do I have to use, the cDNA sequence? against the whole genome??
could you explain it to me?? 