I'd lay good odds that your qPCR primers are not working, but your gene is expressed. The most common reason is poor RT-PCR/cDNA synthesis quality. Not all mRNAs are poly-adenylated, so you may need to use random hexamers, or a mix of random hexamer and oligo dT. Low abundance transcripts or very large transcripts might also be poorly represented or absent from your cDNA. Another potential source of false negatives is failure of your qPCR primers to amplify your cDNA of interest, possibly due to splicing or SNP-based sequence variations. This tends to crop up more often with Taqman primers (b/c they query 3 different sequences), but can do happen with standard SyBR green primers, too (some oligos just don't anneal that efficiently). It is critical that you validate your qPCR reagents against a positive control (easiest system is usually HEK293s transiently transfected with your GOI).

Alternatively, your antibody may be yielding a false positive. As incredible as it may seem, I have actually observed this on rare occasions (and even seen changes upon siRNA treatment that were later confirmed to be artifactual). This scenario is unlikely, but it would be useful to confirm your Western results using another antibody against your protein of interest, to make sure your gene is truly expressed, and targeted by the siRNA.

The fact you can see a decrease in the level of your protein of interest suggests that your knockdown is working well, so yes, I'd say that your result is valid (as long as you're not seeing any off target effects - the decrease in your targeted protein would suggest that at least some of any observable phenotype will be attributable to your knockdown). In fact, as protein is arguably the key functional unit, it's probably the most important thing to show decreasing following RNAi knockdown. If you still feel it's necessary to demonstrate mRNA knockdown, you need to sort out your RNA detection - what approach are you using - qRT-PCR or northern? Obviously, these approaches have different sensitivities and success in each case will be influenced by RNA quality and quantity (did you include any positive controls or had the RNA been used successfully in another application?). If you're convinced that a simple northern is not sensitive enough (many transcripts are indeed undetectable using this aporoach, though you can try increasing the amount you load per lane), then may be you should try qRT-PCR. Good luck.

I would check your mRNA detection method. I would assume you are using qRT-PCR which is very sensitive. It is very rare that a gene undetectable using PCR can then be detected by western, normally its the other way around (you have mRNA but no protein). Does the mRNA detection method work for any sample?

Thank you for the replies, Sam and Martin.

I used qRT-PCR to detect mRNA at normal level, which came out undetectable. ELISA method also showed undetectable levels.

I included another cell line as positive control for qRT-PCR and it showed high level of expression of the particular protein.

(I am a new research student. So I apologise in advance for my naive questions)

Since the dogma is transcription to translation, then l also wonder about the possibility of not detecting the transcript but the translate. l guess you should try out new primers and different primer concentrations since this has been shown to have some effect on qRT-PCR amplification. You could also try using a probe in your qRT-PCR as this increases detection.

The protein must be made somehow from a gene. In that sense it must be detected. Unless the gene is under some form of control and knocking down that control leads to transcription and then translation. In that sense you should be looking at the transcription control factor.

I'd lay good odds that your qPCR primers are not working, but your gene is expressed. The most common reason is poor RT-PCR/cDNA synthesis quality. Not all mRNAs are poly-adenylated, so you may need to use random hexamers, or a mix of random hexamer and oligo dT. Low abundance transcripts or very large transcripts might also be poorly represented or absent from your cDNA. Another potential source of false negatives is failure of your qPCR primers to amplify your cDNA of interest, possibly due to splicing or SNP-based sequence variations. This tends to crop up more often with Taqman primers (b/c they query 3 different sequences), but can do happen with standard SyBR green primers, too (some oligos just don't anneal that efficiently). It is critical that you validate your qPCR reagents against a positive control (easiest system is usually HEK293s transiently transfected with your GOI).

Alternatively, your antibody may be yielding a false positive. As incredible as it may seem, I have actually observed this on rare occasions (and even seen changes upon siRNA treatment that were later confirmed to be artifactual). This scenario is unlikely, but it would be useful to confirm your Western results using another antibody against your protein of interest, to make sure your gene is truly expressed, and targeted by the siRNA.

that's really weird. usually, protein would be undetectable while mRNA can be detected. qPCR primers should be checked. in other way, antibody specificity need be confirmed by overexpression of your gene.

good luck

Dear Melissa,

You should do a time-course mRNA analysis (like 12h, 18h, 24h,48h and 72h after transfection) , then you can see the down-regulation according to time and not a completely lack of expression as you mentioned. This would make you more confident with your results. Good luck

Viviane