Tech Clinic #2: Gel Extraction – Avoid/Rescue a Bad 260/230 Ratio

Gel extraction — what could be easier?

Now we have quick and easy gel extraction kits, we no longer need to use time consuming old fashioned methods like electro-elution or “freeze and squeeze”. Thank goodness.

When Gel Extraction Goes Wrong

But even the simplest of procedures can go wrong.

Maybe you were distracted, confused or thought that gel extraction was so easy that you tried to multi-task one too many things alongside it. (We’ve been there!)

However it happened, the good news is that if you make a mistake with these simple kits, it is still easy to fix.

A question from DJ…

We received the following question from one of our readers, DJ.

By mistake during my gel extraction I eluted with hot water (a tip I got from this site) before I had gotten rid of the buffer PE (primarily ethanol).

I get a decent amount of dna for my ligation purposes but the 260/230 is around 0.03-.1. I evaporated all the liquid in a vacuum centrifuge.

Then I resuspended in 70% ethanol to desalt it. I stuck it back in the vacuum centrifuge. When I resuspended it with water and nano-dropped it, the same 260/230 readings were returned to me.

What could I be doing wrong here that is not allowing salt to be evaporated from my DNA? Thanks! Love the site.

As Nick described in the early days of Bitesize Bio, a low 260/230 ratio is indicative of several possible contaminants.

EDTA, guanidine salts and oligosaccharides can all absorb around the 230 wavelength. The PE wash step is used to remove the left over gel and the salts from the column.

EDTA is usually not a component of wash buffers. But since the buffer formulations in these kits are proprietary, there is no way to know if a low level of EDTA is contained in PE. EDTA is a nuclease inhibitor so adding it into buffers may be a way of ensuring DNase activity is eliminated.

When mistakes occur, it can be hard to troubleshoot when you don’t know the formulas for the solutions you are using.

But the chemistry is really simple (bind silica with chaotropic salts, wash away salts with ethanol, dry to remove the ethanol, elute in low salt buffer or water) and fixing slip ups does not have to mean starting over.

So what could have happened to your sample?

Since you eluted too early in the protocol, your sample would have contained guanidine salts, agarose and prehaps EDTA.

When you dried it down, you also concentrated those chemicals. With a 70% ethanol wash to remove residual salt, you may have removed some of it, but probably not all of it because when guanidine solidifies, it becomes difficult to resolubilize and usually needs some heating. So you did get rid of the ethanol, but probably not the guanidine.

If the contaminants were simply salts or EDTA, then ethanol precipitation would be a better way to clean up the DNA instead of drying it down.  If the contaminant was gel residue, ammonium acetate should be used for the precipitation salt because it will keep oligosaccharides soluble.

Forget the chemistry – just use a kit

If all of that chemistry is sounding like a bit too much hard work, you will be relieved to know that there is another solution that you could try if this happens again:

Simply re-bind the DNA to a new gel column by adding back the QG binding buffer or if you have a PCR clean up kit in your lab, you could use that to clean up the DNA on a new column. You should be able to recover 80% back (according to manufacturer data).

More gel extraction help…

You can get more tips on gel extraction (15 in total!) in a couple of articles, one bymyself and one by Nick.

But now that I am in gel extraction troubleshooting mode… here are a few more:

1. Always perform the additional QG wash to remove residual gel debris, especially if you pushed the limit of the max allowed for gel size, and then perform a second ethanol (PE) wash to make sure all the salt is removed.

You can perform a second wash with 80% ethanol instead of the PE wash so you don’t run out. Of course, dry the column before elution, but at least if you do forget, then the speed vac method will work to clean it up.

2. Only 100% anhydrous ethanol should be used in silica column wash buffers. Using denatured alcohol (typically the outer container will say 95% alcohol) will cause many problems with the drying step.

Denatured alcohol contains chemicals such as isopropanol or benzene that do not evaporate quickly and this carries over into the DNA.

You know when you used the wrong ethanol when your DNA smells funny or it won’t freeze in the -20C. And when you try and load it on a gel, it floats, even in loading dye.

The UltraClean Gel Spin Kit and UltraClean 15 DNA Purification Kit are examples of two gel extraction products that provide wash buffers already containing 100% ethanol  so you don’t need to add your own and mistakes like this can’t happen.

3. Remember your spec has a limited linear range!

If you are using a Nanodrop to quantify the DNA, the readings tend to not accurately reflect the purity if the sample is near the lower level of detection for the instrument. I usually see this effect at around 10ng/ul and below.

When the DNA concentration is very low, the 260 reading is not peaking so the scan looks more like a straight line instead of a bell curve. But if the scan looks like a ski slope (starting out high at the 220-230 side and then sloping down as it gets to 260), then you do have some salt or other contamination in there. The scan will tell you more information, so the 260/230 may be fine and it may just be an effect of low yield.

If you are using a spectrophotometer, the same effect will apply. And, of course, for UV analysis of DNA, if you eluted the DNA in a buffer with EDTA, such as TE, then blank the instrument in TE.

4. For more tips on quantifying DNA with a spectrophotometer, check out this Cold Spring Harbor Labs Protocol.

5. Go easy on the multi-tasking… you only have one pair of hands!

We’d like to know what you think of our gel extraction tips, or if you have any ideas or questions of your own. Tell us yours by leaving a note in the comments section below this article.


  1. Mike Smith on August 31, 2010 at 3:47 pm

    You said “If the contaminant was gel residue, ammonium acetate should be used for the precipitation salt because it will keep oligosaccharides soluble.”

    I’ve never heard that ammonium acetate will specifically keep oligosaccharides soluble vs other precipitation salts (i.e. sodium acetate, sodium chloride, etc). Do you have a reference for this that compares ammonium acetate to other salts?

  2. Suzanne Kennedy on July 18, 2010 at 12:01 am

    Hi Robert,
    No- there will be no problem using PB to wash the column. Buffer QG is guanidine thiocyanate and PB contains guanidine HCl mixed with ethanol. Buffer PB is normally used to help remove extra protein from the plasmid prep. It will not remove the DNA from the column in a gel extraction experiment.
    What might be happening is that the low ratios are caused by the GuThio in QG not completely washing out of the column. This is a strong salt that can be difficult to resolubilize when it dries on the membrane during centrifugation. The ethanol wash is supposed to solubilize the salt and remove it but it is not always enough.

    Maybe the wash with PB gets better resolubilization of the GuThio and switches it for GuHCl, which is easier to wash out with ethanol. We have seen in our own lab that GuThio is much more difficult to wash away than GuHCl.

    Typically RNA extraction kits use GuThio because it is a stronger chaotrope/denaturing salt. But you’ll notice that the RNA kits have 3-4 washes too. The first washes are with a lower GuThio salt buffer and then two ethanol washes.

    Kits using GuHCl to bind can get away with less wash steps.

    I’m glad you solved the problem with your kit. Thanks for sharing!

  3. Robert Rutledge on July 17, 2010 at 4:28 pm

    Hi Suzanne,

    I use the Qiagen kit for gel extraction but usually get very bad (less than 1.0) 260/230 ratios. I have tried each of the suggestions, but nothing seemed to make a big difference. Recently, I accidently washed the gel extraction column using PB buffer from the Qiagen plasmid prep kit instead of the recommended QG buffer. I expected low yield and bad 260/230 ratios. Instead, I had a great 260/230 ratio (around 2). I have since repeated this several times and in parallel with the regular gel extraction protocol and each time the results are the same: excellent 260/230 values. I am wondering if there is some unrecognized problem to using the PB wash that I am missing? The yields are comparable to the normal method and the DNA performs fine in sequencing and cloning steps.

  4. Suzanne on February 28, 2010 at 6:56 pm

    Thanks Edgardo,
    We love the feedback and if you have any topics you would like us to write about or want to write about yourself, please let us know.

  5. Edgardo Rivera on February 28, 2010 at 6:41 pm

    Thanks for the tips. I’ve been having low yields with some of my gel extraction. I’ll definitely try the tips from the gel extraction tech clinics. Articles from this website have been really useful in the past, specially the PDM media. This website is really a good resource. In my breaks from experiments I like to surf around bitesizebio in order to relax while being productive. I hope one day I can contribute back with some of my findings.

  6. Ross Durland on July 2, 2009 at 7:00 pm

    When assessing DNA purity by absorbance, I always prefer to look at the entire absorbance spectrum from ~ 220 – 340 nm, rather than single wavelength measurements. The shape of the spectrum is much more informative than 260/280 or 260/230 absorbance ratios.

    Also, extending the scan out to ~ 340 nm allows you to detect baseline shifts relative to your blank. If the absolute absorbance is low, these can easily affect both the absorbance ratios and the calculated concentrations.

Leave a Comment