The first thing I would test is your RT-PCR primers, it is possible they amplify another transcript or the genomic DNA. Are your primers spanning an intron-axon junction and how is your no RT control look like? If your RT turn out not to be the issue then it is possible your siRNA is inhibiting translation rather then cause mRNA cleavage like some miRNA's do, it's not suppose to happen with fully complementary siRNA's but these were famous last words of many a scientist. Do you have an antibody to test the level of your target protein? If it's getting reduced then good for you if not then you probably have some off target effect which can happen even with the supposedly most specific siRNA constructs.

siRNAs lead to unspecific effects such as activation of the interferon signaling pathway that could lead to the increase of proportion of cells in G0/G1 or off-target effects that could explain your phenotype.

Do you have a control siRNA that does not target any mRNA in your cells? If yes, do you observe the same phenotype?

How long do you incubate your cells with your siRNA before checking the expression of your target? If you wait too long, the mRNA could reach its steady-state level because of degradation and/or dilution of your siRNAs. Usually, you can detect an effect two days after transfection.

If you want to detect an activation of interferon signaling, you can check the expression of oas1 or ifit1 that are known to be activated by such signaling.

Another explanation (but I do not believe to much to it), is that your siRNAs have a miRNA effect, it's meaning that your siRNA block the translation without degrading your mRNA leading to decrease of your protein level. Can you check the protein level by Western blot?

I hope it will help.

Kind regards

siRNAs do have non-specific effects and as pointed out by Guillaume, they can lead to interferon signalling, resulting in the observed phenotype. I would check to see if the negative control siRNA (non-targeting control) behaves in a similar way. If that is the case, then it suggests that it could be due to the cytotoxicity of the siRNA/transfection and not due to the target knock down. Also, did you have a positive control for the knock down, meaning that do you know for sure that the siRNA you have used is efficient in knocking down the target? For the specificity aspect, I would also perform a BLAST search (with relaxed parameters) to see if the siRNA sequence could potentially target anything else. It might give you a rough idea. Another thing to try would be to use another siRNA (different sequence) for your target and see what happens. All the best!

I would suggest separate the two things apart.

The decrease in mRNA of your target proteins is the presumption of your experiment design. If this is not true, or if you cannot say it is true, the phenotype changes makes no sense.

Rather, I would speculate your phenotypic changes comes from system errors that are not eliminated by your controls. Carefully examine the system of your Real-time PCR by setting positive and negative controls. Once this system, including RNAi and gene expression detection system, is stablized and ready, then we move to the next step.

Hope this could help

1. Is your target protein decreased? You can have decrease in protein without measurable decrease in mRNA. If so, then that will explain your observed phenotype.

2. If no specific decrease in target protein is observed, then you have a non-specific effect as described by other responses to your query.

3. Additional controls are needed- does your control or scramble siRNA gives you the same phenotype?

4. try another siRNA seq against your target, and try the same seq and controls in another cell line.