Header Leaderboard Ad


RNA quality for library generation: Low 260/230 ratio



No announcement yet.
  • Filter
  • Time
  • Show
Clear All
new posts

  • RNA quality for library generation: Low 260/230 ratio


    I am having issues isolating quality RNA for gene expression profiling.

    I isolate total RNA from samples stored in Trizol followed by an extra purification step via Qiagen RNeasy Minielute columns (Cat. # 74204). I obtain small amounts (20-240ng) of RNA according to a nanodrop with a RIN number greater than 9 on the bioanalyzer.

    However, on the nanodrop I get a peak a 230 (charcterisitc "shoulder" at 225nm), and a bump and 270nm. I have heard that these two peak problems are normal when isolating small amounts of RNA via Trizol and near impossible to remove.

    I would like to use the RNA to make libraries for gene expression profiling on the Illumina HiSEQ but do not want to submit low quality RNA. Is there a way that I can check ahead of time to make sure that whatever the source of these peaks will not affect library construction (RT-PCR?).

    I would greatly appreciate any ideas on how to further purify my RNA (butanol/ether experiences?), any past experiences as to whether or not such nanodrop readings predicted poor library construction, and other ways to better purify small samples already stored in Trizol.

    Thank you for your time,


  • #2
    Hi Katie,
    How is your RNA look like on bioanalyzer? Looking at your RIN number, it seems to me that your RNA is good for library preparation. Also don't worry about sending the RNA to the sequencing facility as they will look at your bioanalyzer traces and let you know if the RNA looks good for sequencing or not.


    • #3
      Nanodrop "quality" required for facility submission

      Hi Upenda,

      Thank you so much for the response!

      My RNA looks normal on the bioanalyzer. However, if I cannot get my RNA "quality" to appear better on the nanodrop through protocol suggestions from the Qiagen technical support, do you know of a couple of less-expensive experiments that I can try on practice samples of RNA isolated in the same manner to see if the possible contaminants will effect library construction? I am planning on doing RT-PCR (to see if generating cDNA or cDNA amplification is effected), but I am concerned that the only way to truly see if the cause of the unusual nanodrop curve is going to hinder library construction is to just invest in trying to actually generate libraries from a couple of the samples themselves. What are your thoughts on this?

      Thank you again for the help!!!



      • #4
        Dear Katie,

        I have also isolated small amounts of RNA (50-300 ng/ul) with the Qiagen RNeasy Mini kit, and would often get low 260/240 ratios (around 1 or even less), no matter how careful I was to avoid getting Trizol and other junk in my samples.

        We went through a lot of trouble to improve our protocol. Perhaps you have seen these before, but here are some tips that helped us a lot:

        - Make sure Buffer RLT and centrifugation steps are done at room temp (it may crystalize if too cold)
        - When washing, roll the column around to make sure the entire inside gets washed

        (as seen at: http://answers.yahoo.com/question/in...6155235AAi9J1h)

        After doing these things, we saw great improvement in 260/230 ratios (now 1.8-2).

        Also, for qRT-PCR, we have recently switched to a non-Qiagen RNA isolation, the older way where RNA is purified with 3M sodium acetate and the pellet is washed with ethanol. This method has been very successful for us and gave us higher total RNA yield as well as consistent good purity, at least from Nanodrop. It is much less expensive compared to the Qiagen method. We haven't done any HTS stuff with this method to isolate RNA yet, but maybe it is something you might consider trying?

        Good luck!


        • #5
          Cleaning-up small amounts of RNA for Illumina RNAseq

          Dear Drosoform,

          Thank you so much for the reply. Your comments confirmed the following of what the Qiagen tech support said:

          "With regards to your query below, low 260/230 ratios are usually due to carryover of small amounts of guanidine thiocyanate, often in combination with very low RNA concentration. I would also suggest the following:
          1) Performing multiple RPE wash steps to remove residual salts.
          2) Perform slightly longer and slightly faster centrifugation steps.
          3) Ensure that the procedure is performed at room temperature (please do not place samples on ice or spin at 4 deg C).
          4) Ensure that the sample is not too cold prior to addition of Buffer RLT.
          5) Ensure that the RLT buffer does not contain any precipitates. If there are precipitates, equilibrate the buffers at room temperature prior to use.

          They then went on to say in an attached newsletter the following: "In our experience, increased absorbance at 230 nm in RNA samples is almost always due to contamination with guanidine thiocyanate, a salt which absorbs very strongly at 220–230 nm and is present at very high concentrations in the lysis buffer or extraction reagent (e.g., TRIzol®) used in most RNA purification procedures. Our experiments show that the A260/A230 ratio of an RNA sample is strongly reduced when guanidine thiocyanate is present even at submillimolar concentrations (Figure 6A). However, we also found that concentrations of guanidine thiocyanate of up to 100 mM in an RNA sample do not compromise the reliability of real-time RT-PCR, even when using PCR chemistries that are sensitive to inhibitors (Figure 6B). Similar observations have been reported by other researchers (2)."

          However, after optimizing my RNeasy purification with suggestions from the tech support and your post I was not able to improve my 260/230 ratios. I think this is because the extra RPE washes dramatically reduced my RNA yield.

          I may try your suggestion to purify with 3M sodium acetate and wash with ethanol, and I may also try a butanol/ether purification that I found in the following paper: "A simple and loss-free method to remove TRIzol contaminations from minute RNA samples" Helmut Blum et al 2008.

          However, I may just use my protocl as is after speaking with some local PI's they have said they constructed RNAseq libraries from samples with similar nanodrop curves. Upon further inquiry Qiagen tech support has said " If you have a very small amount of RNA, this will skew the 260/230 values, no matter how many times you try to clean it up. You can try to perform ethanol or isopropanol precipitation, but your RNA pellet will be very small and hard to see. I would suggest that you use the RNA for the downstream application already."

          I feel that the best thing to do would be to quantitate my samples again via Qubit, see if an RT-PCR works, and if it the sample has enough RNA and works for the RT-PCR I will continue with Illumina library construction to see if whatever is contaminating my quality RNA hinders quality library construction.

          Thank you for your help, and I will keep this thread updated on whether or not the RT-PCR and library construction is successful. If it is not successful I will keep this thread updated on which methods, in my hands, have worked to further purify small amounts of RNA stored in TRIzol for such future applications.

          Thank you again for your time!



          • #6
            Hi Katie,
            The 270 nm peak is from the phenol in Trizol. The 230 nm peak can be various substances, but given that Trizol contains guanidium, it probably is a salt thereof. However, acetate (eg, from sodium acetate) can also absorb in that area. More details about various contaminants that absorb in the UV here.

            I always advise people doing RNAseq to do a final DNAse treatment followed by a column clean up. I like the this kit for that. However, like many other columns of this type, you can get a little salt carry-over that gives you some UV absorbance around 230 nm. It probably means you have not quite gotten the protocol down perfectly. But as long as the peak isn't sky high, the salt concentration is probably not going to interfere with the first step of the TruSeq RNA prep kit.



            • #7
              Hi Phillip,

              Thank you for the response. I have been very concerned about the 270nm peak and even switched to Phase Lock Gel tubes for the phenol chloroform extraction just in case my technique is to blame. However I still cannot get rid of the little extra bump at 270nm. The only way that I have been able to get rid of the abnormal nanodrop RNA curve is to increase my RNA concentration (cell number) with the same protocol. It is my belief that the contaminant at 270nm is still there but masked by the larger 260 curve in this case. Do you have any other suggestions as to how I might optimize my protocol to get rid of the possible phenol? I am very open to suggestions and would like to try everything possible before touching my precious samples!

              My 230nm peak isn't sky-high, but relative to my 260nm peak it still significantly lowers my 260/230 ratio. The curve actually looks very similar to Figure 1A in the paper that I cited earlier (A simple and loss-free method to remove TRIzol contaminations from minute RNA samples Helmut Blum et al 2008).

              I have incorporated a DNase treatment on the RNeasy columns. I have heard that without this people have had up to 40% of their reads map to introns with Trizol-based isolation procedures and am trying to be as careful as possible!

              Thank you again,



              • #8
                Dear Drosoform-

                I mis-typed earlier: I isolate a maximum of 20-40ng/uL of RNA. Have you worked with lower than 50ng/uL RNA? I am so sorry for the earlier typo.


                • #9
                  Hi Katie,
                  Might help your sense of well-being to calculate what concentration of phenol your preps contain. I think you will find that it is very low and unlikely to affect downstream assays -- except for UV spectrophotometry.
                  But I am surprised you would have phenol carry-over after running a RNeasy column, even at the low level to make it detectable via UV spec.
                  In any case I think you want to think about your QC assays in a different way. Your samples could be contaminated with all manner of substances -- as long as they don't absorb in the UV, you will be blissfully unaware. So it doesn't make sense to fixate on any one QC assay too much. When I see RNA samples that have a little phenol left over, I focus on the nano chip result. If that looks good for total amount of RNA, I declare victory and move on.
                  The concentration of RNA you mention does sound low, but as long as you are over 1 ug of total RNA, you will likely be fine. The thing that will screw up the first step of TruSeq RNA is stuff that would interfere with hybridization of the polyA RNA to the oligo dT, etc.
                  By the way, I did not look at the phenol clean-up method you invoke, but if it is the butanol/ether one -- sure, that should work fine. Just remember to water saturate your butanol prior to extractions unless you want to reduce your volume because it will pull some water out of your samples. (We used to use non-water-saturated butanol as a method to concentrate aqueous solutions.)



                  • #10
                    Hi Phillip,

                    I really appreciate your replies and will keep this thread posted as to how my library prep goes!

                    Thank you,



                    • #11
                      Hi Kathie and everyone,

                      Any new insights into the low 260/230 ration problem using Trizol? I have identical problem: small sample size> small amount of RNA (which is OK) and very low 260/230 ratio (less than 0.1).

                      Thanks a lot,


                      • #12
                        Hi Wacguy,

                        I started quantitating my samples with a Qubit as the nanodrop that I was using should only be accurate down to ~20ng. I also did RT-PCR to see if the contaminant(s) at 230nm would inhibit the reaction (it did not). I have not yet tried to construct libraries for RNAseq to see if the contaminant inhibits this but can keep you posted! I still also check quality with a Bioanalyzer and routinely get samples with RIN > 9.

                        I am planning to use these RNA samples that I have isolated with TRIzol, but for samples that I am isolating now I have switched to Qiagen's RLT plus + BME. The RNA appears much more clean on the Bioanalyzer with this switch. Also, my samples always look better when processed fresh than stored in lysis buffer if you can process them right away. Are you storing your samples in TRIzol? If so, are you mixing really really well upon thaw (otherwise phenol can isolate in the aqueous phase) and letting the TRIzol come to room temperature before phase separation?

                        If you would like, I can post my TRIzol and RLT plus protocol here that I have had the most success with the little tricks that have reduced salt contamination in the RNA sample. How much RNA are you working with? From what source? What is your isolation protocol?



                        • #13
                          Hi Kathie,

                          Thanks a lot for such fast reply. I'm new to RNA world. I used Trizol and no kit to get all RNA types. I ground my small piece of root w/ liquid nitrogen and added the Trizol; (do you think I should wait a few secs till the ependorf is at RT w/ the powder and then add the Trizol?). I don't store the samples in Trizol. After adding it, I add 1/5 of a volume of Chloroform, centrifuge and transfer the aqueous phase, add Glycoblue and precipitate w/ ethanol and NaAc O/N. Next day I centrifuge in 4oC for 30' and wash twice w/ 75% EtOH and then dissolve in water. I got 5-10 ng/micrl (w/the nanodrop) which was surprisingly high but as I mentioned, extremely low 260/230 ration. Qbit was less friendly to such low RNA concentrations. Did you try the Butanol method; I'm considering of trying it but we don't have ether here although Butanol can be discarded w/o the ether I think: (http://www.ncbi.nlm.nih.gov/pmc/arti...00185-0196.pdf). I'd love to read about your tricks, maybe I can skip the Butanol part that seems a bit tedious to me.

                          Thanks a lot again,


                          • #14
                            Hi Guy,

                            The Qubit is more specific for RNA than your 260nm reading on the Nanodrop. Have you considered incorporating a DNase treatment? In my experience, it is really hard to not get DNA contamination with TRIzol extraction and this can contribute to your 260nm reading and make you think you have more RNA than you do.

                            I did not try the Butanol method for the same reasons you mentioned. However, if you don't need to get rid of whatever the contaminant is being detected at 230nm (salt from TRIzol?) I wouldn't mess with it. Have you tried your downstream application yet? Have you looked at your sample quality via Bioanalyzer yet? How does that look? Have you tried dedicating more of your sample to the Qubit reading? The one that I have access to can detect as low as 5 ng of RNA so I usually use 2 uL of my samples to quantitate. I discovered I had much less than 20 ng/uL of RNA after DNase treatment. You can test for DNA contamination in your samples that have not been treated with DNase with an RT-PCR +/- the reverse transcriptase.

                            I don't think you need to wait with the powdered sample before adding the TRIzol. Just make sure that the TRIzol+sample is at room temperature before the phase extraction.

                            My tricks are basically modified versions of a Qiagen column clean-up kit with an on-column DNase treatment. However, they never gave me very high 260/230 ratios. I think this is because my sample was too small for an accurate nanodrop reading. As my 260 value decreases and my 230 value stays the same it's really effects the ratio. I have decided to just move forward with my downstream applications and see if it will work as my RNA is high quality. I will post my TRIzol and RLT protocols later tonight.

                            Good luck in the meantime!



                            • #15
                              Hi Katie,

                              Just wondering if you ever managed to get your RNAs to work? I have the exact same problems with mine, A260/230 <0.5 even though my RNA concentration and absolute yield are workable (although still <10ng/uL). Would be massively grateful if you would post the modified RLT protocol? I'm also using the RNeasy Plus Micro kit, with multiple cleanups and concentration steps, but with no significant improvement to the ratio (also no improvement to my A260/280). Quite frustrating!