Personally, I would always use the same amount of RNA in each reaction, and the kit instructions here are simply...wrong.
Assuming your RNA is clean (good 260/230 ratio is most important: aim for 2.0+, but anything above 1.7 or 1.8 is usually ok), then the only difference between '1ul of RNA at 120ng.ul-1' and '10ul of RNA at 12ng.ul-1' is that the latter contains an extra 9ul of water.
If I had two samples with those concentrations, I would do two cDNA synthesis reactions with 120ng of RNA, using 1ul of the first (+9ul of water) and 10ul of the second (with no added water).
You'll eventually be using reference genes to normalise anyway, but there's no point in deliberately starting out with highly divergent RNA amounts if you can possibly avoid it. Much easier to normalise (and much less prone to error) if all your cDNA gives values in a similar range.
Also, you don't list your volumes: I assume they are small, and you're dealing with samples that don't yield much RNA (13ng.ul-1 is not a lot of RNA, unless you've got a milliliter of it, or something). If your volumes are actually fairly high (100-200ul) then you could concentrate up your RNA by precipitation (isopropanol or ethanol) and redissolving in a smaller volume.
As to your final question, that is very much a 'how long is a piece of string' question. 100ng of RNA is a reasonable amount, even if you factor in that 80%+ of it will be ribosomes. For genes with high expression (GAPDH, actin etc) you'll be absolutely fine. For genes with lower expression you may be into slightly questionable territory.
Of course, this also depends on how much of your cDNA you put in each well. I would recommend you dilute your cDNA at least 1/5, possibly 1/10 or 1/20 to dilute out the cDNA synthesis buffer (it can interfere with PCR, in my experience), but most mastermixes are quite generous in volume of cDNA permitted.
For reference, I usually use 800ng of RNA, final vol 10ul, then dilute 1/20 to give 200ul of cDNA stock, then use 2ul of this in 10ul PCRs (so ~8ng cDNA per well, assuming 1:1 conversion). This usually suffices for even fairly low copy-number transcripts, and you'll be in this ballpark (even a factor of ten lower will only increase your Cq values by 3.3).
Personally, I would always use the same amount of RNA in each reaction, and the kit instructions here are simply...wrong.
Assuming your RNA is clean (good 260/230 ratio is most important: aim for 2.0+, but anything above 1.7 or 1.8 is usually ok), then the only difference between '1ul of RNA at 120ng.ul-1' and '10ul of RNA at 12ng.ul-1' is that the latter contains an extra 9ul of water.
If I had two samples with those concentrations, I would do two cDNA synthesis reactions with 120ng of RNA, using 1ul of the first (+9ul of water) and 10ul of the second (with no added water).
You'll eventually be using reference genes to normalise anyway, but there's no point in deliberately starting out with highly divergent RNA amounts if you can possibly avoid it. Much easier to normalise (and much less prone to error) if all your cDNA gives values in a similar range.
Also, you don't list your volumes: I assume they are small, and you're dealing with samples that don't yield much RNA (13ng.ul-1 is not a lot of RNA, unless you've got a milliliter of it, or something). If your volumes are actually fairly high (100-200ul) then you could concentrate up your RNA by precipitation (isopropanol or ethanol) and redissolving in a smaller volume.
As to your final question, that is very much a 'how long is a piece of string' question. 100ng of RNA is a reasonable amount, even if you factor in that 80%+ of it will be ribosomes. For genes with high expression (GAPDH, actin etc) you'll be absolutely fine. For genes with lower expression you may be into slightly questionable territory.
Of course, this also depends on how much of your cDNA you put in each well. I would recommend you dilute your cDNA at least 1/5, possibly 1/10 or 1/20 to dilute out the cDNA synthesis buffer (it can interfere with PCR, in my experience), but most mastermixes are quite generous in volume of cDNA permitted.
For reference, I usually use 800ng of RNA, final vol 10ul, then dilute 1/20 to give 200ul of cDNA stock, then use 2ul of this in 10ul PCRs (so ~8ng cDNA per well, assuming 1:1 conversion). This usually suffices for even fairly low copy-number transcripts, and you'll be in this ballpark (even a factor of ten lower will only increase your Cq values by 3.3).
Whether you are a student or a researcher,,,,,,, this might be of some interest to you. Not as an answer to your question, ,,,but as a source of ideas to consider.
John explained very well how to proceed. To decide the amount of RNA to be retrotranscribed you must take into account how many genes you have to analyse as well as their expression level: if You analyse just 3-4 genes (including reference genes), and their expression level is fair, so that the Ct is
I usually use 1000ng. Depending on the expression of your gene (aacording to the literature) use 15-30ng of cDNA per qPCR reaction. In my experience it is always better to have lower Ct values in RT-qPCR. Results are more reproducible.
I don’t think 100ng would be enough. Remember that you should have at least enough material for two reference genes plus any gene you are interested in.
Based on the protocol, the amount of RNA should be 1 ug, but I would like to use 1.5 ug if my target gene is more than 2.5kb or the expression level is relative low, because I think with high concentration of RNA, it is easy to obtain relative more cDNA pattern of target gene(s). sometimes, it is also necessary to prolong the extension time when doing reverse transcription.
Just an addition to the many good answers already posted here: I have used 250 to 500ng of good quality isolated total RNA with a commercially available RT kit in a 10uL reaction (random primers). After 15 min RT reaction, I dilute the sample with 10uL water. From this final cDNA sample, I use 1µL in a 20µL qPCR reaction. Control mRNA transcripts like GAPDH and G6PD will display Ct values of around 18-25 with that approach in my set up (I don't have the exact numbers here at the moment, but they do come up early enough ;))