I don't know what all you tried. In my own experience, it helped several times to just use different primers and/or different master mixes (just try some different brands, you will usually get free samples upon request).

Another option is to use sequence-specific primers for RT, and/or to make a nested PCR (what is actually a sequence-specific pre-amplification).

100ng of cDNA is also...quite a lot. I typically use 8 (assuming 1:1 conversion), and I work with some fairly low-abundance transcripts.

1600ng RNA in 20ul synthesis, diluted 1/20 afterward (so 400ul final, 4ng per ul), use 2ul per well (10ul reactions).

Under these conditions, 18S should be giving you Cq values of 8-10, not 18. This suggests you may be adding too much cDNA, which can (counter-intuitively) inhibit the reaction efficiency.

Have you tried a dilution series with your cDNA (start at 100ng per well, work down in 2-fold dilutions to something like 1 or 2ng)?

Do this for both your GOI and your reference genes, ideally.

Also, do the same with a positive control that you absolutely know SHOULD express PCSK9 (genecards suggest liver is pretty good, and you can probably buy human liver cDNA).

100 ng is too much, try to make it about 30 ng for target and 15 ng for housekeeping gene, getting mutiple peak may indicate that your primer is not specific, try to do a serial dilution to check the effeciency of your assay

John Hildyard : Yes, I have tried serial dilutions starting at 20ng, 40, 50 and 100ng, but did not observe amplification in the lower concentrations and hence decided to use 100ng to see if it yielded any results. Also, the same cDNA was used for amplification of other targets and the housekeeping genes and we did not face any issues, which made me think that perhaps low expression was the reason. I understand that liver cells would be the perfect positive control in this case, however, due to budget constraints, I may not be able to procure additional requirements. Hence, I used endometrial tissue cDNA which was available with another researcher in my lab to check the efficacy of my primers. I will however, try serial dilutions again to figure out the optimum concentration of cDNA required for this particular GOI.

Also, I am getting Cq values of 12-15 for 18S, the 18-29 mentioned were for other target genes.

Thanks for replying.

Jochen Wilhelm Thank you for replying. I have tried different primer and template cDNA concentrations (cDNA upto 50ng), however, could not optimise the amplification conditions since I could never obtain single peaks or bands upon doing agarose gel electrophoresis either. Hence, I used endometrial tissue cDNA which was available in my laboratory to check if the issue was low abundance of PCSK9 transcript in blood (since it has been reported earlier), and was able to obtain single peaks for the first time. And yes, the primers used were sequence-specific but haven't attempted a nested PCR; will definitely try and see if that works. Also, will try to procure different master-mixes.

As a side note - can you comment on whether the melt curve would shift if the sample type is changed (in this case, endometrial tissue and blood?)

Rasha Abd Ali Hussein Al-khalidi : Thank you for replying. I have started from 20ng cDNA, increasing to 100ng since I could not observe a single peak, despite troubleshooting the assay conditions. I will do a serial dilution again to see if there is something that I may have missed.

The melting temperature depends (a bit) on the buffer conditions in the PCR.

Samples from different tissues may contain different residual salts and have a slight impact on the pH as well. So yes, there might be a shift (not drastic, but possibly recognisable).

I agree with some of the answers, Multiple Tm peaks indicates that your primer isn't specific. Also, variation in technical repeats indicates pipetting error - and this shall give you high standard deviation. Is that what you're getting in your results? How do you design your gene of interest (GOI) primers? Do you simply copy paste from existing literature or do you carefully choose them via NCBI Blast and IDT?

Udesh Dhawan Yes.I design my own primers using IDT, Oligocalc and NCBI Blast, do not copy paste from existing literature. Also, I have worked with 5 other GOIs and several housekeeping genes. Did not observe the issue of high standard deviation in those. This is my 2nd set of primers and they're also yielding similar results, which led me to believe that low abundance of my transcript was perhaps responsible for varying results, and hence consistent high Ct values in samples (28-34).