UV spectra of some common contaminants of DNA or RNA solutions.

[pmiguel]

Spectrophotometers, such as the Nanodrop, can quickly produce a UV spectrum from a 2ul drop of solution without a cuvette. Gone are the days when one relied solely on absorbance at 260 and 280 alone to gauge the quality of ones sample. Now the sight of a perfect UV spectrum can be a great source of encouragement:







But what about cases were your spectrum does not approach that almost platonic level of perfection?

Common wisdom is that good DNA and RNA will give you a A260/A280 ratio of 1.8-2.0. Lower values for this metric are frequently attributed to protein contamination. Less commonly, one uses A260/A230 as a metric with above 2.0 is considered good. Much below that cutoff is frequently attributed to "organics", whatever that means. (Yeah, what does "organics" mean? My guess? I think the "organics cause low A260/A230" is just a laboratory Urban Myth. But I deal more with the A260/A230 ratio, and what really causes low ones, later.)





DTT.
(Thanks to JF at IU for digging up a spectrum for DTT at 454 Titanium training years ago.)
Dithiothreitol is a reducing agent. Not commonly added to DNA or RNA solutions. But many enzymatic steps in DNA or RNA processing protocols will employ it. DTT does have a UV spectrum:









PHENOL

Ah phenol, long a favorite protein denaturant of old-school molecular biologists. Frequently combined with chloroform and, if you are true purist, a little isoamyl alcohol as an "anti-foaming" agent. (The latter being important if you, for some mysterious reason, eschew centrifuges for the true tool of a practitioner of liquid-liquid extraction: the sep funnel.) There must be 100 ways to use phenol in the lab, and 1000 ways to mis-use it. In my most curmudgeonly of hearts, I know that without a mastery of the phenol extraction, one is not a true molecular biologist. But, realistically, that attitude probably results from the inconvenience and difficulty of using phenol. Phenol is definitely in my "avoid if possible" list of reagents.

Here are some of the fun characteristics of phenol, either "lore" I have picked up along the way, or stuff I have observed during my first hand interactions with phenol:

• At room temperature, when pure and unoxidized, phenol is a colorless crystal.
• As it oxidizes, phenol turns yellow, then red, then brown.
• The lore is that even yellow phenol is too oxidized to use -- it will damage the DNA it is used to extract.
• Phenol is only semi-miscible with water (below pH 9 or so--somewhere above that pH it becomes fully miscible). Mixtures of phenol + water (or other aqueous solution) form emulsions when vigorously mixed (for example, with a vortexer) but separate back into two phases (the "aqueous" phase and the "organic" phase) given time. Usually the organic phase is of higher density and therefore is the bottom phase. Centrifugation is generally used to speed the separation of phases.
• Prior to its use, one generally "equilibrates" phenol by adding an equal volume of a buffered aqueous solution (usually Tris) and allowing the mixture to "melt" in a 50 oC water bath. During "melting" the phenol absorbs water from the aqueous phase. After absorption of water, the phenol remains liquid, even at 4 oC in the refrigerator.
• If you equilibrate phenol with an excess of water, instead of a buffer solution, such that you have two phases, the aqueous phase will be acidic. In this acidic state, a solution of DNA mixed with the phenol, will result in the DNA dissolving preferentially into the organic phase. But RNA will remain in the aqueous phase. This is the basis of the "acid phenol" method of RNA extraction which has been commercialized as the "Trizol" reagent.
• Typically, phenol is meant to be used for DNA extractions--where one wants the DNA to remain in the aqueous phase and only protein to fractionate into the organic phase. To accomplish this, the phenol must be "equilibrated" with a more basic aqueous solution, usually Tris. Most protocols I have seen call for the investigator preparing the phenol to sequentially mix the phenol with equal volumes of 0.1 M Tris at the desired pH. After each mixture, the aqueous phase is tested. If the pH is too low, the aqueous phase is removed and another volume of 0.1 M Tris is added, mixed and the phases allowed to separate. Good luck getting this to work in less than substantial fraction of the lifetime of the universe. My guess is that this is a protocol "honored in the breach" and the typical technician tasked with equilibrating the phenol with either use a higher concentration solution of Tris, use Tris at a pH higher than the final desired pH or both.
• Phenol will blister your skin, given time. It also has anesthetic properties, so it does not initially burn. Last I checked, it was still an ingredient in some sore throat remedies.
• 8-hydroxyquinoline is sometimes added to phenol both as an anti-oxidant and as a colorant. Its bright yellow color helps you distinguish your aqueous from your organic phase. But the yellow color could also mask the yellow color phenol will take on when it oxidizes. So, initially, 8-hydroxyquinoline protects your phenol from oxidation, but when your phenol does oxidize, you will not be able to tell. Or maybe with the addition of 8-hydroxyquinoline, oxidation is no longer a problem.

How about phenol's UV spectrum? Pure phenol pH spectrum's metrics make it look like pretty good DNA, if you don't check that 260/270 ratio. Here is the spectrum of a 0.1% solution (1ul of molten phenol diluted to 1ml) Notice that the 260/230 ratio tells you nothing! Why? Phenol does have an absorption peak near 230, but the shoulder of its 270 nm peak is so much larger, that ratio of 260/230 looks like good DNA. So, don't look for phenol at 230 nm, look at 270 nm!





The question remains, where does that low 260/230 nm absorbance myth come from? One site has even done the nanodrop spectrum of phenol + RNA and their spectra do show high absorbance at 230! See here, for example.
How can that be? I propose some alternate hypotheses below. But one major consideration is the pH of the solution into which the phenol is dissolved:



Given that the spectra look nearly identical for 0.1% v/v phenol in water, compared to 0.1 M HCl, I think we can presume that phenol dissolved in water is nearly completely protonated. Dissolving into a 0.1 M NaOH? (theoretical pH 13 before addition of phenol) shows the 224 nm local lambda max shifting slightly and broadening extensively. At crazy-high alkalinity (1 M NaOH?) the spectrum looks barely recognizable.

For extracting protein from DNA, phenol is typically pHed to 8.0. As I mention above, the standard written protocols make reaching this pH nearly impossible. Why? A bottle of phenol containing 500 ml is about 5.6 mols of phenol. 500 ml of 0.1 M Tris is, well, 0.05 moles of Tris. That would be 5.6 moles of acid (phenol) against 0.5 micromoles of free hydroxide (pH 8 0.1 M Tris). I leave the math to the reader, but I'm guessing the original author of the canonical "equilibration of Tris" protocol, never actually tested it.

In any case, the spectra of phenol in 0.1 M Tris pHed to 7.8 or 8.2 scarcely differs from that of phenol dissolved in water.


8-hydroxyquinoline
The earlier mentioned 8-hydroxyquinoline also has a UV spectrum. I don't have any in my lab, so I pulled the spectrum below from NIST.


Salts of acetic acid
(acetate)

The other major player as a contaminant of solutions of nucleic acids is acetate. As in sodium acetate, potassium acetate or ammonium acetate? Here is the spectrum of a solution of 100 mM sodium acetate:



Sodium Acetate

That is quite a bit of acetate--100 mM. But a common method for precipitating DNA is adding 0.3 vol of 3M NaOAc? prior to adding 2 volumes of ethanol. So your resuspended DNA could easily end up as a 100 mM solution of NaOAc? were you not very careful washing your pellet.

Also, I think salts of acetate might explain a couple of other UV spectrum legends. First that a high 230 nm reading comes from residual ethanol. This is fairly easily tested, and not the case. However, acetate salts are almost always used as co-precipitants in an ethanol precipitation. If you have large amounts of residual acetate in your sample after a precipitation, you might also have some ethanol contamination... So, if you squint, you could say this one has an element of truth to it.

The second is that if you equilibrate your phenol with tris acetate, then you might end up with enough in the aqueous phase to give that hefty 230 nm reading that is frequently attributed to phenol.

Tris

Tris is the worst buffer that everyone uses. Why use a buffer that shifts its pKa 0.3 pH units lower for every 10 ºC increase in temperature? No, really. Why use it? The very act of "pHing" Tris (adding acid to bring a solution of it to a desired pH) releases enough heat to throw the pH of the solution off. I think the answer is that everyone uses Tris because everyone uses it. Most protocols are built around Tris. Tris is fairly cheap. So, basically, we are stuck with it.



How about its UV spectrum? Actually not too bad. At very high concentrations >100mM it absorbs at 220 nm (and presumably shorter wavelengths). I really only bring it up because it is difficult to know how much Tris ends up dissoving into phenol during equilibration with buffer. Maybe not enough to turn your DNA solution to 1M Tris. But 0.1 M? Maybe.




Others? Post the spectrum of your favorite contaminant!

--
Phillip

[dietmar13]

hello,

i get consistently this spectrum (attachment) after extraction of RNA.

my procedure:
add 10x V (1 ml) QIAzol http://www.qiagen.com/Products/Catal...inginformation) to 100 µl H2O solution of exosomes (contains RNasin® Plus RNase Inhibitor, http://at.promega.com/en/products/dn...e-inhibitor%2f)
add 140 µl chlorophorm
15', 15k rpm, 4°C
extract water phase two times with chlorophorm
add 2 µl glycogen
precipitate with 0.8V isopropanol (RT)
30', 15k rpm, 4°C
wash pellet twice with 70% EtOH (-20°C)
dry and dissolve in H2O

it seems not to be phenol, glycogen (this has no UV-spectrum), ...
perhaps it is DTT (this is in the RNasin), but the spectrum looks different.
how could i get rid of this contamination or if it is DTT, does DTT intefere with enzymatic reactions (RT, fragmentation for RNA-seq library preparation, ...)?

[JamieHeather]

I don't have a spectra to hand, but the one that always gets me is guanidine thiocyanate leftover from gel extractions (see figure 4 here). Increasing the washes usually drives most of this down.

[dietmar13]

possibly - guanidine thiocyanate is in the QIAzol.

do you know if this interferes with enzymatic reactions (RT) or RNA-seq library protocols...

[JamieHeather]

Buffer QG also contains it.

I suppose if there's lots left over then it would inhibit any downstream enzymatic reaction. However I regularly perform PCRs on gel-extracted DNA, which certainly work, so presumably not enough is carrying over to noticeably stop reactions.

[pmiguel]

Originally Posted by dietmar13

hello,

i get consistently this spectrum (attachment) after extraction of RNA.

my procedure:
add 10x V (1 ml) QIAzol http://www.qiagen.com/Products/Catal...inginformation) to 100 µl H2O solution of exosomes (contains RNasin® Plus RNase Inhibitor, http://at.promega.com/en/products/dn...e-inhibitor%2f)
add 140 µl chlorophorm
15', 15k rpm, 4°C
extract water phase two times with chlorophorm
add 2 µl glycogen
precipitate with 0.8V isopropanol (RT)
30', 15k rpm, 4°C
wash pellet twice with 70% EtOH (-20°C)
dry and dissolve in H2O

it seems not to be phenol, glycogen (this has no UV-spectrum), ...
perhaps it is DTT (this is in the RNasin), but the spectrum looks different.
how could i get rid of this contamination or if it is DTT, does DTT intefere with enzymatic reactions (RT, fragmentation for RNA-seq library preparation, ...)?

That is a little remnant Qiazol. Note the characteristic local maximum at ~270nm. That is phenol. Then you have some other UV absorbent solute at a shorter wavelength. I agree with dietmar13 and JamieHeather that it is guanidine isothiocyanate, most likely.

Your biggest problem at this point is that the phenol has a shoulder at 260nm, so you can't tell how much RNA is there. Well, I should not say "can't". It would be difficult.

If you were going to make a library out of this sample, I would suggest cleaning it up with something like the Zymo DNA-free RNA kit. Basically it is just a DNAse digestion followed by a column clean-up.

--
Phillip

[Jon_Keats]

Great thread Phillip,

These are exactly many issues I've encountered in the past. My lab it actually told to set the nanodrop vertical bar at 270 to make sure the peak absorbance is not at 270. Also, I agree 260/230 is very valuable metric that most people disregard. Usually low number correspond to short drying times of isolated DNA.

[Solorem]

Could anyone please tel me ifni can DNA extraction kits to extract gDNA from RNAlater preserved cattle skin tissue, i am conducting a PhD research on polymorphism and mRNA expression levels of HSP 70 and 90 in Nigerian and Indian cattle. i am about to commence gDNA extraction for polymorphism study and genotyping of HSP 70 and 90 genes from the samples next week. I came to India with the samples from Nigeria. I took the samples between January- early march, 2017. i stored the samples in refrigerator before my travelling, i transported the sample with ice park to create a refrigerated environment for the sample. I intend doing mRNA expression study on the samples after completion of polymorphism study of the genes. i take time to give these preamble so that whoever that want to guide me will have detail information to work with. So kindly guide me on the use of extraction kits as phenol-chlorofom assay is not available in the lab that is hosting me. looking forward for your kind suggestions. Thank you