The microRNA profiling had done for cancer cells treated with (X) drugs, and we want to understand the mechanisms of this drug.
A lot of functional experiments were done previously in my lab, and they found that (X) drug induce EMT, proliferation, and invasion. In vivo, increase the tumor size and cause some development impairment for the embryos.
These findings led us also to think about an epigenetic mechanim induced by this drug. The main focus is on TGFß and Wnt signalings.
To make my question more precise, I have a list of target genes from both pathways for two candidates of miRNA were predicted by using bioinformatics algorithms. So now I can test which target gene could be selected for further experiments. If there any specific approach for Wnt and TGFß signaling.
This is a strange question. I can understand the desire to find targets of candidates microRNAs that fit into a hypothesis. There are justifiable reasons to approach the problem this way. But technically, I am not sure this is the correct way. Similar questions here on ResearchGate are heading the same direction: how deep is the desire to understand microRNA biology. I see that this may sound judgmental or impractical. But shouldn't you use a less biased approach to define the microRNA target genes. Going into the experiment with so much ballast, so much prejudice about the outcome (it should be something related to Wnt, EMT, TGFbeta) may not be as scientific as one would wish for. You will find on these "endless" lists of predicted targets, the genes your are looking for. But are they really the important ones and you just don't want to see the real targets?
It is hard to give a clear answer. It depends on how good the evidence is that makes you focus on certain pathways, how good the evidence is that certain predicted targets fit into this pathway. All pretty shaky assumptions. There is still a lot of gut feeling involved , a lot of luck.
Dear Thomas,
Thanks for your answer, but I have justifiable reasons to ask
First of all, I don't want to talk about my hypothesis in open area !
but we have strong evidence to look for miRNA candidates and their target genes. I did the microRNA profiling using a high throughput method and we have a list of top up /down regulated miRNAs analysed by very professional bioinformatic facility.
The profiling did for Cancer cells treated with (X) drug and we want to understand the mechanisms of this Drug, because when the patients take this drug make the tumour more aggressive and induce metastasis.
a lot of functional experiments were done previously in my lab and they found that (X) drug induce EMT, proliferation and Invasion. in vivo increase the tumour size and cause some development impairment for the embryos.
these findings let us to think about epigenetic regulation caused by this drug. the focusing on TGFß signalling and Wnt because they are the master regulator of EMT.
To make my question more Clear.. I have a list of target genes from both pathways for 2 candidates of miRNA were predicted by using bioinformatics algorithms. So how I can test which target gene could be selected for further experiments.. if there any specific experiments for Wnt and TGFß signalling.
Thanks for the clarification. Why not manipulate your microRNAs of interest in a cell line of relevance and measure which predicted target genes behave the way they are supposed to by qRT-PCR or Western blotting. You would go through the predicted target list and pick as many as you can handle.
This should narrow down the targets a lot. The qRT-PCR approach is not perfect but many targets are affected on the mRNA level. If you do both: increase microRNA expression and decrease its activity, you may actually get for good responders a total of a 2-fold change.
If you are not sure whether your experiments work, you may have to generate a microRNA luciferase sensor, co-transfect it into your cells and measure whether the sensor responds to the microRNA inhibition or overexpression. This is your gold standard. The sensor makes the microRNA work like a siRNA and you should see a clear response.
Some targets will not respond much on the mRNA level. Then it becomes more laborious.
Do you have mRNA profiling data from the cells treated with the drug? Comparing mRNA and microRNA profiles may give you additional clues. But the transcriptional changes induced by the drug X may be too overwhelming to identify a microRNA activity signature in it.
In the end you can do a lot of things: treat with the drug after you manipulated the microRNAs of interest and see whether the pre-treatment alters the outcome of the drug treatment. In a simple world, the manipulation of the microRNA should "dramatically" alter the cells' responses and change the EMT etc phenotypes...
Hope this helps a little bit.
Hi Alia,
as Thomas Andl said, you will need to perform gain- and loss- of-function experiments to validate the targets you find in your model.
If you want to narrow your list of genes to check by qPCR, I would suggest to look for predicted targets which expression is correlated with that of your miRNA of interest. You might find numerous positive correlations as a large proportion of miRNAs–mRNA interactions involve feedback loops (see for example
http://genesdev.cshlp.org/content/22/18/2535.abstract?ijkey=29475619d69d4dcf3a0112d116c3ddb7754678b6&keytype2=tf_ipsecsha
or
http://www.ncbi.nlm.nih.gov/pubmed/24037378?dopt=Abstract)
However, our team recently showed that a significant proportion of miRNA-correlated transcripts, in particular those that are predicted to be direct targets, are indeed causally regulated by our miRNA of interest by using this kind of approach (http://genome.cshlp.org/content/24/5/850.long)
Hope this will help
- Badges
- Science topic
- Similar topics
- Biological Science
- Biochemistry
- Biomolecules
- Nucleic Acids
- RNA
- Small RNA
- microRNA