Hello Rachel,

Forgive me if I am wrong, but it looks like this is the first or one of the first times you do qPCR. qPCR is 'simple', but also serious business, and if what you do is not correct your data will suffer accordingly. I would strongly recommend you to read the basic papers to get a good grasp, so you will see that actually your questions can be answered elsewhere on those papers. Do come back when you need troubleshooting or guidance.

Michael Kubista is one of the great ones at qPCR, and I encourage you to revise his papers on the TATAA biocenter webiste, especially:

http://www.tataa.com/wp-content/uploads/2012/09/Clinical_Chemistry_55_101816%E2%80%931823_2009.pdf

http://www.tataa.com/wp-content/uploads/2012/09/Methods_50_231%E2%80%93236_2010.pdf

http://www.tataa.com/wp-content/uploads/2012/07/miqe-bustin-et-al-clin-chem-2009.pdf

There you will find whether using a reference gene is correct or not, guidance on how to validate that reference and whether you should use more references, and how to approach the dilutions. I would also suggest you to check papers on qPCR from fecal matter, because your data seems to point at a problem with RNA purity (i.e. your extraction and purification method) that is affecting the Cq of your reference gene (>28 cycles is way too high for a normal one), and that could be one key for your experiments.

As a note on your quantification approach: first you extract RNA, then you assess quantity (ng/ul) and purity (260/280 and 260/230, check https://biosci-batzerlab.biology.lsu.edu/Genomics/documentation/3130_NanoDrop_tips.pdf), then you do the RT, then dilute and then qPCR. Measuring cDNA concentration with a nanodrop will not be useful, because your RT reaction contains protein and salts from the buffer that will affect the readings.

Cheers,

Ruben

You will need to have the same input RNA for your reference gene assessment. Otherwise, you won't know whether the relative quantity is different due to biological variation or due to differences in starting material.

I would not assume the dNTPs are responsible for high readings for NTC, the nanodrop (or spectrophotometry) is not used to detect carry-over/amplicon contaminants. That is what electrophoresis is for. Food for thought, what if the dNTP was contaminated via accidental double pipetting?

If you "normalize" or that is, input the same material for cDNA synthesis, say 500 ng, then you can be fairly sure that the reference gene study will give you stable gene.

So spec your RNA, input same RNA into cDNA, run reference gene stability, then use your stable genes. In the subsequent cDNA synthesis, you do not have to use same RNA input because you have previously shown that gene X is stable (since it's relative), but to me this is just a lazy approach that should be avoided. It's always good to visually confirm after a run a tight clustering of reference genes. You can't do that with a lazy, "1 ug in this tube, 200ng in that, 400 in this tube" approach. You also run the risk of overloading the reaction input parameters particularly for samples with high RNA contents, waste of RNA material as well as RT kit.

https://go.qbaseplus.com/qpcr-training-normalization-strategies-genorm-global-mean

Just to add to Can's answer, and as general guidelines:

• for each qPCR reaction 5-50ng of reverse transcribed RNA is usually a good compromise between having too little and overloading.
• If you want to be more precise, you can run a dilution series of cDNA, and use that concentration that yields a Cq of 20-25 cycles for your gene of interest (more than 25 cycles and you start risking bad quantification, less than 20 and you are actually wasting your cDNA if it is in limited amounts).

Cheers,

Ruben

So, the first time I did the RT I thought I had contaminated my no template reactions (doing one RT for each miRNA, so 2 RTs per sample). I still went on with the qPCR, loading 100ng per reaction (trusting the nanodrop reading). My no template reactions were negative, so the nanodrop reading is not my template. All other reactions showed amplification with cts from 32 to 38 for my reference miRNA.

So my second run I tried standardizing the initial RNA amounts (loaded 500ng of total RNA for each RT), then loaded the results without any dilution in the qPCR (RT products were 1450 to 4000 ng/microliter, loaded 1 microliter). This time my reference miRNA came from 27-31. However this time the curves (this is a taqman assay) were clearly showing that I loaded to much cDNA.

My main question at this point is:

-should I dilute my total RNA more (with the risk of loading too little of the miRNAs that actually interest me - bear in mind that my samples are feces so I will always have plenty of other RNAs in the extraction)

-should I keep using the same load of total RNA but dilute the cDNA? Can I trust the nanodrop and dilute based on the reading, or should I use a fixed dilution like 1:10?

Also, I am planning to evaluate the target miRNA based on the ratio between target and reference miRNA, is this correct? I am hoping that this would eliminate the intrinsic bias of how much of my fecal sample is actually from shedded cells from the host and how much is microbiota or food.