Announcement

Collapse
No announcement yet.

RNA quality for library generation: Low 260/230 ratio

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • #16
    RNAzol

    Hi Tanals,

    I'm using ENAzolRT and it gives really good results; high concentration of RNA (in respect to the small tissue size I am starting with) and good 260/280=1.7 260/230=0.8 values, not perfect. The RNA was tried for qPCR w/ 3 primer sets and there were no problems so I guess it is OK.

    Guy

    Comment


    • #17
      For the record, we rarely look at Nanodrop readings for our library preps.
      Quantification information from it is often thrown off by any increased A270 values. We either use the Qubit or the Bioanalyser. The Bioanalyser isn't brilliantly accurate, but as long as we can get >100ng based on it's reading we seem to produce more than enough library. RIN's on the Bioanalyser are important, however, as you need non-degraded transcripts to avoid 3' bias.
      You also have to remember, the first steps of the TruSeq RNA protocol is a mRNA isolation on Oligo dT beads. This in non-enzymatic and includes an ethanol wash, which should remove any left over nasty stuff in your extractions.

      Comment


      • #18
        Man, I wish no one had ever invented the idea of checking two 'diagnostic ratios' with UV spec. Seems to just encourage people to ignore the full spectrum that the nanodrop gives them and concentrate only on these (often meaningless) metrics.

        UV spec is a powerful tool but it seems to have accumulated too much protocol 'baggage' from the 1960's (or whenever) to be of much use these days. C'mon guys, think for yourselves. If you want to use a nanodrop, take a few of the UV-absorbing substances in the lab (acetate salts, phenol (@ 0.1% in water), guanidine salts, beta mercaptoethanol) and run spectra on them. Hang them by your desks so you can learn them as well as your mother's face.

        While you are at it, calculate 260/280 and 260/230 values for each so you can see how easily these common lab compounds can confound this assay. Here, let me lay down the guantlet -- until you do this you are not worthy to use a UV spectrophotometer. At best you are an automaton malfunctioning every time you are given a sample that is just a little outside of perfection.

        Or, you can read my post where I ran some of these spectra. Perhaps that will suffice. Better yet, find some other UV absorbing compound and post its spectrum to that thread!

        --
        Phillip

        Comment


        • #19
          I found it most useful to look at the nanodrop curves, bioanalyzer trace and RIN, qubit concentrations, and RT-PCR +/-reverse transcriptase while I optimized my protocol. I also started to work with a technician that processes hundreds of RNA samples a week that have been stored a variety of ways. She gave me tips from samples that are always her favorite to process. My protocol and tips are written below.

          I am submitting my samples for microarray + nugen amplification because I could only get a 5 ng total RNA for my smallest samples. I will let you know if it worked once I find out!

          RNA extraction: RLT plus (Qiagen #11048449) +BME version

          Tips:
          -Use of RLT plus gives me more RNA and much more clean RNA instead of TRIzol
          -Processing fresh is best for the above reasons as well. If freeze, do a quick thaw at 37 degrees Celsius
          -Make EtOH fresh. I make sure that it's less than one week old. Some prefer to make it that day.
          -DNase is very sensitive to physical denaturation so be very careful (gently invert only to mix, no vortex or pipetting up and down)
          -If centrifuge has the option, put centrifuge "soft" mode OFF as many of the spins are short. This helps get more RNA-looking nanodrop curves (by getting rid of contaminants more)
          -Get out all tubes and reagents (RNase-zap these, pipetts, and hands) before begin. Don't get too crazy as can hurt RNA too! Make sure to let all dry or kim wipe off at least before working with it. Getting out the RNeasy tubes now is important too as all of the steps with this are supposed to be at room temp and not being so can effect isolation.
          -Low binding tubes and tips are good so don't lose any of precious RNA isolated.
          -Discard flow through by pipetting to reduce salt contamination on outside of RNeasy column that could contaminate at later time.

          Protocol:
          1) Sort cells, pellet down 350 xg 20 mins, aspirate off supe, resuspend in 1mL RLT plus (Qiagen #11048449) + 10 uL BME (freshly mixed by vortexing together before adding to sample). Can look-up and optimize how much RLT+BME to add, but is a lysis buffer. I went for the over kill with my 30K sorted cells to keep it standard across my samples. Only danger I could think of with this step is not using enough lysis buffer. RLT plus got better reviews by the technician cited above than the straight RLT that comes in the RNeasy kit specified below.

          2) Mix really well by vortexing sample on max speed for 30 seconds - 1 minute. Don't bother pipetting up and down here or at later steps as just gets RNA into tip and potentially lost. Some will say not to vortex here as can shear long RNA and DNA so harder to see if have contaminating DNA by PCR. I optimized not having contaminating DNA without the vortex step (shaking vigorously instead), then switched to this to get maximum lysis (and saw max amount of RNA isolated with vortexing).

          3) Spin down 10K 1 minute to get all of sample back down to bottom of tube. Can store in -80 here if need to, but I like to process fresh and saw best results if just did it all now.

          4) Pipett to QIAshredder (Qiagen # 79654) to homogenize and pellet down crap (proteins, etc) to be avoided when transfer to new tube. Spin down 2.5 minutes on max speed at room temperature. I like to circle where all of the protein will pellet down to so I know to avoid that part of the tube when aspirating up everything I can with my pipett on the other side. I use the same tube with the same spot facing towards the outside so I know where to avoid. I used to not do this step, but it did help make everything a little bit more clean.

          4) Transfer to new tube. Again, I like to use siliconized tubes (Low bind) to avoid RNA loss.

          5) Add 70% EtOH 1:1 (1 mL here) and shake vigorously about 15 seconds to mix. I kept the shake instead of vortex here because I didn't see a difference with vortexing at this step and tried to avoid vortexing for reasons specified earlier. If use 1 mL, will likely do your sample in two tubes. half in one, half in another. A common mistake it to not take the exact proof of the EtOH stock into account when making your 70% EtOH. The one I use is 200 proof molecular biology grade. Make sure you're really working with 70% EtOH and not lower.

          6) Quick spin to get your sample back down to bottom again.

          7) Bind samples to column by transferring to Qiagen RNeasy miniElute column (Qiagen # 79254). Spin 10K xg at room temp for 1 minute. I like this column because of the very small elution volume (14 uL) and the subsequent concentration of my RNA.

          8) Discard flow through by pipetting, not decanting. This just got my RNA curves a little bit better by reducing salt contamination on the outside of my column that spins down at later steps.

          9) Repeat steps 7-8 until all of sample has been run through column.

          10) DNase treat on column (Qiagen RNase free DNase set recommended by RNeasy column packet #79254) according to directions and repeated below:
          -Wash column with 350 uL RW1 from RNY ME kit at 10K xg 15 seconds.This doesn't come in any of the Qiagen catalogue numbers specified above and I found it in another kit we had lying around. I put RW1 directly on column instead of running down sides to avoid having salt from RW1 at higher column areas. This is very salty, I think, based off of nanodrop curves (see below reasons in RPE wash step).
          -Add (10 uL DNase + 70 uL RDD mixed by gentle inversion if any) to the column. Do not pipet up and down as can hurt DNase.
          -Incubate at room temp 15 minutes
          -Wash with 350 uL RWI again and spin 10K xg 15s
          -Can repeat DNase treatments or increase incubation time to 20 mins if necessary. I optimized this in my hands by doing RT-PCR +/-reverse transcriptase to get rid of contaminating DNA. One treatment at 15 mins was sufficient and it's what I do for now on. My PCR method could only detect down to 0.1 ng DNA, though, as I saw by running specific amounts of DNA in other gel wells to know how low I could detect with this method.
          As people said above, though, you can get rid of DNA in the mRNA selection for the RNAseq... but without DNase treatment you might think you're looking at a lot more RNA than you are. I kept it for accurate quantitation so I knew exactly which gene expression profiling platforms were options for me.

          11) Wash column with 500 uL RPE at 10K xg at room temperature for 1 minute. Make sure to invert (and roll between fingers) RPE on column for ~2 minutes before spin to get rid of all possible salt still stuck to tube. For inversions, put into fresh tubes to prevent getting old liquid/salts on the outside of the RNeasy column that could contaminate at later time/spin. Some like to increase the RPE amount to 700 uL. If have a lot of salt contaminants (230 nm UV, can do this step multiple times. With the RW1 wash needed for the DNase treatment, I found it was very salty and I needed to do two RPE washes to get rid of it. I only needed to do one RPE wash without the DNase treatment.

          12) Repeat RPE wash step 11. See above. I still put it into a fresh tube for the inversion here again.

          13) Wash column with 500 uL 80 % EtOH, spin 10K xg at room temperature for 1 minute. Some have optimized their isolation to get rid of this step. I haven't seen a difference with or without it either, but have kept it.

          14) Transfer column to new tube.

          15) Spin with top open to dry for 5 minutes at 16K xg at room temperature. Any hint of EtOH and won't pull off RNA.

          16) Transfer column to new tube. Elute RNA by adding 14 uL RNase free water to column. Incubate at room temperature 1 minute with lid closed and spin 16K xg for 1 minute.

          17) Repeat step 16 by using the 14 uL flow-through. I got twice as much RNA this way sometimes.

          18) Inspect RNA samples via nanodrop (UV ID of contaminatns and quantity estimate if have enough RNA for this. I liked it to see if my curves looked like normal RNA curves.), bioanalyzer (quality of RNA/degredation), qubit (quantity if need specific and can do as low as 5 ng/multiple uL), and RT-PCR (+/-DNA contaminants). I did all of these while optimizing my isolation with practice samples of the same number of cells that I had for my samples, but now with my real samples I just do bioanalyzer. If that estimates that I can quantitate by qubit and not waste all of my sample in the process I do that. If it doesn't, I just use the bioanalyzer estimate and will see how the microarray core wants to proceed from there.

          Please let me know if any of this isn't clear. I have a similar TRIzol isolation protocol too if you would like me to post this. This is basically just a slightly modified version of the Qiagen RNeasy column and DNase treatment protocols... but with specifics added in that tech support, other labs, and personal experience has let me add in. Good luck!!!

          -Katie

          Comment


          • #20
            PS. Using RLT plus vs. TRIzol and isolating fresh are two things that have started to give me RIN values of 10 every single time. My only ones lower than that were not done fresh and/or processed with TRIzol. Good luck again!

            -Katie

            Comment


            • #21
              Hi Katherine

              Thanks for your responses. I've basically done pretty much as you did and still got the same results. I did run an extraction without the DNase step (either on-column or solution based, which are part of either Qiagen's micro or mini kit) and I'm pretty sure that the salt contamination comes from the wash buffer here. As our facility will not accept samples unless the A260/230 ratio passes their QC, I'm going to try a set without DNase treatment and see how that goes. Your trizol protocol may be worth a go, do you mind posting it too? I'm using this with phase-lock gel tubes, have you ever tried those?

              Thanks
              Tanals

              Comment


              • #22
                Hi Tanals,

                I actually started with a TRIzol protocol that included the phase lock tubes. It is written below. The DNase treatment was very important for me because a lot of my 260 reading was coming from DNA, and I thought I had a lot more RNA than I really did. If you treat it on the column it avoids losing RNA.

                Our facility told me the same thing, but I just said "I optimized my protocol by looking at the nanodrop curves and ratios, then used the Qubit to quantitate because it is more accurate for my tiny amounts of RNA" then submitted that with my bioanalyzer trace's and RINs. They took the samples without question, then, but I just submitted them so I can't tell you if it worked for the microarray (switched to nugen amplification and affymetrix microarray because of low amounts of RNA, time of processing for our grant we want to submit soon, and experience of our facility).

                One thing that can effect your ratios with a TRIzol extraction is the phenol so be careful if you think it is salt because you don't know that for sure. Phenol generally gives you a pretty specific curve, though... A good example that helped me is in that paper I referenced early on in this thread. I found the shaking really well after having it frozen is very important for proper separation.

                The phase extraction (that, in my hands, couldn't get rid of the DNA anyways) combined with the DNase treatment seemed like just extra steps that I could be losing my RNA at. When I switched to the RLT lysis I got more and higher quality RNA. I'm sure it's different in everyone's hands, but the switch really helped me. Maybe a neighboring lab has some you can try before buying your own? TRIzol is also not the best for very small sample sizes. It was ok until I got down to about 50,000 sorted lymphocytes by flow cytometry, or 20,000 spleenocytes that were counted by microscope and then put into the tube (my practice samples like this likely had more cells than my sorted lymphocytes). Every cell type is different, though, so you will have to see for your cells.

                TRIzol protocol our facility gives us and that I optimized in my hands for my small samples:

                Note: This wasn't as good as the RLT protocol above if you can change.

                Tips similar to before:
                -Do not vortex as this could shear long RNA or DNA so that it is harder to detect.
                -Make EtOH fresh.
                -DNase sensitive.
                -Centrifuge mode off of "soft" if have it.
                -Get everything ready and RNase-zapped ahead of time.
                -Low binding tubes
                -Discard flow through from wash steps by pippetting to reduce salt contamination.

                Protocol:
                1) Sort cells, spin down, aspirate off supe (as before), and resuspend in 1 mL TRIzol.
                2) Store -80 for up to 1 month
                3) Thaw quickly at 37 degrees Celsius. Important sample is at room temp before phase extraction.
                4) Mix really really well by pipetting and shaking vigorously or phenol will stay in aqueous phase. This was something that was really messing with my nanodrop curves and readings so mix really well.
                5) QIA shredder to homgenize and pellet down crap/protein as before
                6) Meanwhile, spin phaselock gel tubes in a separate centrifuge at 16K xg for 1 minute. Position tubes so all handles facing outwards.
                7) Add 250 uL chloroform to samples before adding to phase lock tubes. Shake vigorously ~15 seconds.
                8) Add sample with chloroform to phase lock gel tubes. Let sit room temp 3 minutes.
                9) spin 16K xg at 4 degrees Celsius for 5 minutes.
                10) Add 250 uL chloroform again. Invert gently to mix. Let sit room temperature 3 minutes.
                11) Spin 16K xg at 4 degrees Celsius 10 minutes
                12) Transfer top clear layer to new tube
                13) Let sample and RNeasy columns come back to room temperature for ~1-2 minutes before continuing on.
                14) Add 70% EtOH 1:1, shake to mix.
                15) Bind samples to column by transferring to Qiagen RNeasy miniElute column (Cat # 74204)
                16) Spin 10K xg room temperature for 1 minute. Do sample in two parts to fit into column.
                17) Discard flow through. Repeat for remaining samples.
                18) DNase treat on column as specified in RLT protocol above.
                19) Wash with RPE 2x and 80% EtOH 1x as specified in RLT protocol above.
                20) Finish same way as with RLT protocol above.

                Good luck again!

                Katie

                Comment


                • #23
                  Hello, kindly what is meant by these ratios in RNA sample? And can it be used for gene expression? 260/280=2.05
                  260/230=1.09
                  ng/ul =11.5

                  Thanks

                  Comment


                  • #24
                    Originally posted by HMM View Post
                    Hello, kindly what is meant by these ratios in RNA sample? And can it be used for gene expression? 260/280=2.05
                    260/230=1.09
                    ng/ul =11.5

                    Thanks
                    Hi HMM,
                    The first two ratios are of spectrophotometric readings at those wavelengths. 260nm is ostensibly the "lambda max" wavelength--the crest of the peak-- of nucleic acid bases. Is it really though? Eh, close enough. 280nm is a maxima for protein. 230nm isn't really a maximum for anything in particular but there is a bench-scientist-popular meme that signal there derives from "organics", whatever that is supposed to mean in this context.
                    I mainly regard these ratios as meaningless. I won't bother repeating the bench folklore about their significance . Fairly easy these days to look at the full UV spectrum of your sample. Do that and learn what the spectrums of various pure substances that absorb in the UV look like.
                    "ng/ul" is nanograms per microliter. This one, if determined fluorimetrically using an appropriate dye actually does mean something. It is the concentration of the solute of interest (RNA in this case) in your sample. Of course 90% of the time it wasn't determined fluorimetrically, but spectrophotometrically. This is much less trustworthy. Because lots of chemicals used in the purification of RNA have a signal at 260nm. phenol (from Trizol, for example) being the prime confounder for RNA.

                    So these 3 rations, by themselves cannot directly tell you anything about gene expression. They are more sample "metrics" representing hurdles you might need to clear to convince a core lab to accept your sample for downstream processing (including gene expression.)
                    Regards,
                    Phillip

                    Comment

                    Working...
                    X