We often have this problem when working with HIV, RSV, and MuLV Gag proteins. Depending upon what you eventually want to do with this protein, such as make co-complexes with a specific nucleic acid, adding benzonase or equivalent nuclease can be a problem with down stream processing. Two things that work depending upon your downstream application. 1) Lysis and extraction in high salt 0.7-1M NaCl followed by Polyamine P (PEI) precipitation of the nucleic acids, either from the crude lysate or cleared supernatant. When using Polyamine P you need to make sure that the stock is pH'd, is very basic right out of the bottle. Typically a 10% stock is good then between 0.1-0.5% final to precipitate. I tend to play around to find the optimum. This can still be used even when your protein goes down in the pellet. In fact, this was one of the keys to purifying the Klenow fragment for crystallization in the bad old days. In that case you can back extract with high salt, in the case of the klenow fragment it was AmSO4. Another potential issue here is that you may not be able to use cation exchange chromatography immediately after PEI precipitation since this will bind strongly to that resin (in fact it is the key to the displacement chromatography method in cation exchange). If the following step is Ni affinity chromatography or anion exchange there is no problem with a little PEI. 2) You can perform hydrophobic interaction chromatography. Loading the protein in high salt to bind it to the column, nucleic acids come through in the flow through and Bob's your uncle. I generally prefer using a method that captures the protein since often the nucleic acids are of comparable or heterogeneous size and SEC does not completely remove them.

I'd suggest you to use polyethyleneimine precipitation. It usually helps, although you may need to get rid of it afterwards. I usually do 0.3% (v/v) PEI precipitation followed by ammonium sulphate precipitation of my protein. For more info, you can check this protocol: http://ecoliwiki.net/colipedia/index.php/Methods:Polyethyleneimine_precipitation

Hope it helps.

Cheers,

Modesto.

do you treat your lysates with DNAse prior to purification?

Would enzymatic digestion with or without enzymes covalently linked to a solid phase, beads etc., work?

You can try the following:

1- use an heparin binding chromatography, it strongly binds to DNA-binding proteins and displaces bound DNA. We have used this successfully.

Other possible methods:

2- use an anion excange chromatography, it may retain the DNA and not your protein, depending on the pI of your protein and the affintiy of your protein

3- add polyethilenimine, it may selectively precipitate the DNA (or may coprecipitate your protein as well).

4- unfold the protein with urea and refold it after removing the DNA by the one of the methods mentioned above.

I'd suggest you to first treat your protein sample with benzonase (a mix of DNase and Rnase) during 30 min at 37°C. In a second time, you could try an heparin binding chromatography (ge healthcare).

Hi, if you're recombinantly expressing the target protein with tag (eg 6xHis) for purification you can also purify the protein under denaturing conditions, follow by on (purification) column refolding of the protein before the protein is eluted.

You should try DNAse, but some times it is not useful if DNA is completely masked by the protein. In that case you should try DNAse with reducing agent or some mild detergents, but same time you have to take care of your protein as well as DNAse activity.

Addition of high salt (NaCl/KCl) of ~1-2M to the buffer followed by a size exclusion chr. step would be helpful to remove the bound DNA.

If you need denatured proteins only to determine mol.masses or for a structural work (trypsinolysis +Ms/Ms), use phenol/SDS deproteinization, move out water phase with DNA(RNA), add acetone (90% final conc.) to precipitate proteins and use sample buffer for Laemmli SDS/PAGE or 4-6 urea (= salt) to dissolve proteins. Slow dialysis helps to keep proteins in soluble form in the appropriate buffer.

Good luck!

All the above sound like good suggestions. We had a particular problem with DNA co-purifying with GST-tagged transcription factors in GST pull-down experiments. We found that adding ethidium bromide to a final concentration of 20 ug per mL.effectively blocked any protein DNA interactions.

I would recommend either PEI precipitation or streptomycin sulfate precipitation (or both depending on the affinity of your protein to nucleic acids). The disadvantage of PEI is that you will have to remove it by ammonium sulfate precipitation (see Modesto's well documented answer). you can also play with salt concentration to favor or disrupt proteins-DNA interactions. If your protein is a DNA binding or enzyme working on DNA I would avoid DNase treatments (even traces of DNase can become a nightmare ...). You can have a look at one of my papers where I used differential precipitations to purify a protein:

https://www.researchgate.net/publication/11487730_Cleavage_properties_of_an_archaeal_site-specific_recombinase_the_SSV1_integrase?ev=prf_pub

Article Cleavage Properties of an Archaeal Site-specific Recombinase...

I would use Benzonase (removes both DNA and RNA) as it was always successful for us, for 90 minutes at 4 C, rotating wheel (make sure you don’t have detergent in your buffer, and that is supplemented with appropriate cofactor, Mg2). If you are still not satisfied with your sample purity, try using pre-packed heparin columns. It acts as a cation exchanger (high content of anionic sulfate groups, mimics the similarly polyanionic structure of the nucleic acid), and you can elute your protein samples with gradient salt buffers.

All the given suggestions are efficient. Having the same problem in the past with our recombinant protein expressed in E. Coli, we found that it could works fine also if you prepare your crude extract using sonication. Many short steps of sonication keeping the bacterial colture on ice (to avoid warming up the protein extract) may be quite helpful in cutting nucleic acids in small fragments that are easier to get rid off...

It is fast and has a good efficience. Of course it may be not sufficient if your protein binds to DNA strongly, and in that case you'll have to go to one of the other described methods.

Good luck with your work..

I'd try the Benzonase option and dialyse against a dessert spoon of activated charcoal in your protein's favourite buffer to remove the fragments.

Apart from many good suggestions above, try cation exchange chromatography. If your protein binds to the negatively charged solid support, the nucleic acid contamination would come down significantly due to charge repulsions within the column.

Hi, I have a less technical suggestion which may do the trick - add an excess of phosphate to outcompete the binding constant of the enzyme for the phosphate backbone of the DNA, then add DNAse to digest. After this a brief size-exclusion step to clean up (something like a NAP-5 column should do it).

We often have this problem when working with HIV, RSV, and MuLV Gag proteins. Depending upon what you eventually want to do with this protein, such as make co-complexes with a specific nucleic acid, adding benzonase or equivalent nuclease can be a problem with down stream processing. Two things that work depending upon your downstream application. 1) Lysis and extraction in high salt 0.7-1M NaCl followed by Polyamine P (PEI) precipitation of the nucleic acids, either from the crude lysate or cleared supernatant. When using Polyamine P you need to make sure that the stock is pH'd, is very basic right out of the bottle. Typically a 10% stock is good then between 0.1-0.5% final to precipitate. I tend to play around to find the optimum. This can still be used even when your protein goes down in the pellet. In fact, this was one of the keys to purifying the Klenow fragment for crystallization in the bad old days. In that case you can back extract with high salt, in the case of the klenow fragment it was AmSO4. Another potential issue here is that you may not be able to use cation exchange chromatography immediately after PEI precipitation since this will bind strongly to that resin (in fact it is the key to the displacement chromatography method in cation exchange). If the following step is Ni affinity chromatography or anion exchange there is no problem with a little PEI. 2) You can perform hydrophobic interaction chromatography. Loading the protein in high salt to bind it to the column, nucleic acids come through in the flow through and Bob's your uncle. I generally prefer using a method that captures the protein since often the nucleic acids are of comparable or heterogeneous size and SEC does not completely remove them.

I have worked with many RNA binding proteins, they do co-purify with RNAs and/or DNAs. After the first purification step (according to the tag on your protein, if you have one) I would use a DEAE column. The DEAE resin binds to nucleic acids quite well, to improve binding of the nucleic acids to the column make sure that you load your sample onto the column with a low ionic strength (50-100 mM salt). Your protein could also bind to the DEAE according to its pI but in my experience nucleic acids bind really tight. I did not like to precipitate the nucleic acids at the beginning of the purification, because in my hands this always co-precipitated some of the target protein as well. Hope this will help.

A slight note. 1) PEI can typically be removed by AmSO4 precipitation of your protein and then you can use cation exchange. 2) My recollection is that streptomycin sulphate can also be use to remove nucleic acids, however I have not used that in over 20 years so I cannot help in that regard. 3) Make sure you get the correct size polyamine, i.e. Polymin P is the tradename.

The simplest way is to remove DNA from protein using a matrix such as Nitrocellulose. Nitrocellulose does not bind double stranded DNA and Proteins bind well. Through exclusion chromatography one could remove DNA and elute Proteins using salt gradient containing detergents.

Best of luck

We purify RNA polymerase and other DNA binding proteins for crystallography and need highly purified material – this includes removal of any non-specifically bound DNA.

We typically use a combination of the following (some of it has been suggested before):

1. If you purify from the native source (i.e. not over expressed in E. coli or the like) you can treat your sample with small amounts of DNAseI (typically 130ug of DNAse/100g of cells). I have never had an issue with DNAse sticking around during subsequent steps

2. We always do Polyethyleneimine (PEI) precipitation (Polyethyleneimine, ~ M.N. 60,000, 50 wt.% aq. solution, branched, Acros Organics). As has been said before some empirical testing will be required. We use a 10% stock (pH’ed to 8.0) and typically use 0.1-0.5% final concentration – add it slowly while stirring your sample. You can even dialyze the PEI stock to get rid of low molecular weight PEI if you’re worried about removing it. Our target protein precipitates with the PEI! We wash our PEI pellets with low salt buffer (0.2-0.5M salt) and elute the target with high salt (1M) – that way you already do a crude first purification. Again, empirical testing will be required. You need to AmmSO4 precipitate your protein to get rid of the PEI afterwards (see next point).

3. AmmSO4 precipitation. You can wash your AmmSO4 pellets using a solution containing at least as much AmmSO4 as you used to precipitate in the first place – that way you can elute DNA out of your pellets. You can monitor your progress by measuring the absorbance at 260 – worked well in the past for me.

4. Heparin column. As has been said before – DNA binding proteins like to bind to Heparin. Elute with a salt gradient. In the worst case (if your protein binds with very high affinity to DNA) you separate the bound (in the flow-through) from the free protein (binds Heparin matrix).

5. Ion-exchange chromatography has worked in the past for me.

If your protein goes into inclusion bodies, then I would do a freeze (-80) thaw (RT water bath) lysis of the E. coli if they carry a pLys plasmid, or you can add add exogenous lysozyme. Then after the cells have lysed completely (the pLys plasmid helps to insure that), briefly (2 x 1 second) blend the lysate to shear the chromosomal DNA to reduce the viscosity. Then pellet the IBs from the lysate at ~12k for ~20 min. Then you can wash the IBs with urea (0.2-0.5 M, depending on solubility of protein) and 1 M NaCl, followed by two DI water washes to get rid of the NaCl. Lastly do the benzonase incubation on the resuspended IBs to degrade the DNA and RNA at RT. The urea/NaCl pre-cleaning steps reduce how much benzonase you need. With IBs you want to make sure you get the pellet completely suspended after each wash/centrifuge step. There are often little chunks that I will suck up with a 10 mL pipette from a 1 L beaker containing the IB prep and put in a 1.5-15 mL cent. tube and then use a 1 mL pipettor to break up the chunks by pipetting up and down repeatedly. The IB chunks are held together by DNA/RNA.

If the protein has a His-tag, I've used 1.0 M NaCl washes when the protein is bound to the affinity column. His binds Ni-affinity tightly in high salt to allow for binding, washing, and elution.

As others have suggested, I've also found benzonase to help. Works well if you follow with a high salt on-column wash, for example 500mM NaCl on a Ni-chelating column if you have a His-tag.

thank you - I should have mentioned that I prefer to avoid nucleases as it complicates subsequent assays. In my experience nucleases are difficult to completely remove.

The majority of protein systems that we study are expressed as inclusion bodies, which offers a unique advantage relative to soluble proteins as the washing of extracted IBs usually constitutes an intrinsic purification step eliminating nucleic acids. Following refolding and purification, we often measure a relatively low 280/260 OD ratio which is indicative of residual DNA/RNA impurities. A method that we employ to eliminate the remaining nucleic acid contaminants involves applying the protein sample in low ionic strength buffer to a Q-sepharose column. The protein normally elutes at salt concentrations less than 0.5 M depending on the solution pH and protein pI, whereas nucleic acids bind tightly to the column and require much higher salt concentrations to detach from the resin. In principle, a similar purification protocol could be applied to proteins expressed in the soluble fraction as a means of removing residual nucleic acid contamination.

Avoid E. coli expression systems, not only do you have the chromosomal DNA to deal with but also the plasmid DNA. If your protein is a cytoplasmic protein, than an insect cell/baculovirus system will dramatically reduce the amount of contaminating DNA in a cytoplasmic lysate since the DNA is in the nucleus.

John, in the Kowalczykowski lab we also worried about nucleases, and the only thing that got rid of nucleases was a Mono Q column as a final polishing step. This final step also removed any remaining DNA as well.

Are there cases where a protein has to be taken to 37 degrees or partially denatured to release bound DNA or RNA or can it always be removed by the methods noted in here without raising temperature or adding denaturants?

To my knowledge, raising temperature (particularly to 37°C which can be pretty cold for thermophilic proteins ! ) is not a strategy used to dissociate protein-DNA/RNA complexes. I would not consider denaturants as a good solution. Increasing salt concentration at the crude extract step is a good alternative as far as your POI is stable in such conditions (and of course that you are not dealing with halophiles!). As was already pointed out (if I remember well), you can play around with salting in/ salting out strategies which will not only remove contaminating nucleic acids but may also contribute to the purification.

Hence, to strictly answer your last question, DNA/RNA can *almost* always be removed by the different strategies listed above.

I concur that raising the salt should work. I have used up to 1.2M and then PEI precipitated. I assume that you know that it is not covalently attached to your protein?

One other out of the box thought, if nothing else works, you could try a strategy used for nucleotide binding proteins. Specifically, if you know that your protein binds relatively weakly to say a short olignucleotide then saturate the solution with this nucleotide to displace the tightly bound ones. This general approach is pretty good at stripping the last bit on ADP for GroEL and DnaK (there we use a GTP analog or other lower affinity nucleotide). However, raising the salt or changing the pH is probably the most likely to work.

Okay, "if" the protein is His-tagged and it is a DNA binder you can do this: first add 5-10 mM EDTA to the lysate. This will remove metal co-factors needed by nucleases and by proteins that bind DNA. Then you can either dialyze the EDTA away or use an IMAC media that can handle EDTA (Roche cOmplete His tag media-requires pH 8.0 for binding His-tags), then bind the His-tagged protein to an IMAC column (or better, batch bind to media, batch wash and then load onto column), then do a 0-1 M NaCl gradient wash (check peak fractions for DNA), if the DNA does not come off then you can try using urea (up to 6 M) to partially-fully denature the protein. Then do a "decreasing" gradient wash at 5oC to remove the urea and allow the protein to refold on the column. This on-column refolding works really well because the protein does not contact other proteins which can interfere with the refolding process. If you use the cOmplete His-tag media you can keep 5 mM EDTA in the NaCl and urea washes to help scrub off any metals and bound DNA.

Dear Dr. Tainer,

some DNA enzymes (polynucleotide transferases, i.e. DNA topoisomerases, recombinases, resolvases, etc.) form so-called 'relaxation complexes' where proteins will be covalently linked to DNA under protein-denaturing conditions. Review on this subject was published in Nucl. Acids Res. (1998), v.26, N21, 4791-4796. Moreover, even under natural conditions we have some covalent intermediate compounds of these enzymes and DNA in the living cell. Sometimes, increasing ionic strength (up to 0.5-2 M salt, no other protein denaturants) helps to dissociate most of complexes of these enzymes and DNA.

I have worked with DNA methyltransferases and restriction enzymes. PEI treatment in the presence of 1M NaCl/KCl works perfectly fine. In this step, one can get rid off all the bound DNA.

Use DNase in extraction buffer

As others I recommend to add DNase I to your lysis buffer and use an anion exchange column like HiTrapQ for removing the DNA. Your protein will be in the flowthrough.

Dear John Tainer

Try to add benzonase (Merck 1.01654.001) to the cell lysis buffer. See our article Gesser et al. 2007 J Invest Derm. We purified nuclear proteins for Western blotting and 15-30 min treatment on ice with benzonase 1-2 microl / 100 microl sample buffer (see our sample buffer with 7 M urea) can be recommended.

Best regards, Borbala Gesser

You can also from the beginning dissolve the nuclear fraction in Tris buffer with 7 M Urea. You can centrifuge down DNA (14.000g) and thereafter dialyze the supernatant for pure water. After 5-6 changes of water, you can freeze dry the sample and redisolve the proteins in a new buffer for gel electrophoresis. You can also try to separate the sample in the Tris buffer with 7 M Urea, on a gel filtration column from the DNA.

Cheap alternative for Benzonase is here: http://www.protean.cz/en/recombinant-protein/11/endonuclease-s-marcescens

If your protein constructs have tags (GST or His), you can try on-column washing with 1 M NaCl. I utilize this successfully when purifying certain recombination proteins. Also, a high-resolution ion exchange should work as well at the final step of purification.. Either Bio-Rad ENrich Q/S or GE Mono Q/S.

I would recommend benzonase.  Benzonase works well when Mg2+ is added to the lysate. 

If the protein is His-tagged, then you can use 6M GuHCl/0.3-1 M NaCL to solubilize the protein (of course it will be unfolded) and bind to a His-tag column while disassociating the nucleic acids, and then wash extensively "in batch mode first followed by pouring into a column and continuing to wash (can measure wash for nucleic acids), before using a gradient/step gradient to remove the GuHCl and reduce NaCL to 0.3 M to refold the protein on the column, followed by elution of the protein. The batch mode washing steps prevent nucleic acids from getting trapped in a column matrix where they can bleed out during protein elution.

You can also try precipitating DNA by adding poly ethyleneimine (0.1% final) into your centrifuged lysis, mix well and incubate for 5 min on ice, then centrifuge again.

Benzonase

I think use Urea for protein that you want to crystallized would be not so good. Benzonase is working well, but if you are working on DNA binding protein you should try : lysed your cells with 1 M NaCl buffer with benzonase treatment, then add heparine column step after hisTrap and before GF.

Why not to try a filtration through the DEAE-paper filter at salt concentration lower 0.4 M?