We give a lot of troubleshooting help on DNA and RNA isolation here at Bitesize Bio because almost everything we do in molecular biology requires DNA or RNA at the very first step. These days, most labs use commercial kits, which employ spin columns, for the isolation of nucleic acids. The spin columns contain a silica resin that selectively binds DNA/RNA, depending on the salt conditions and other factors influenced by the extraction method.
These kits make the whole process much easier and faster than the methods of old, when things are going well, but the downside of using a kit is that we don’t always know what is in the mysterious and proprietary set of solutions that each company uses in its kit, which makes troubleshooting more difficult.
So in this article, I’ll explain in some detail how silica spin filter kits work and what is going on at each step. I’ll also go over some common problems specific to silica columns that can be overcome or avoided with just a little extra understanding.
Lysis:
The lysis formulas may vary based on the whether you want DNA or RNA, but the common denominator is a lysis buffer containing a high concentration of chaotropic salt. Chaotropes destabilize hydrogen bonds, van der Waals forces, and hydrophobic interactions. Proteins are destabilized, including nucleases, and the association of nucleic acids with water is disrupted setting up the conditions for the transfer to silica.
Chaotropic salts include guanidine HCL, guanidine thiocyanate, urea, and lithium perchlorate.
Besides the chaotropes, there is usually some detergents involved, to help with protein solubilization and lysis. There can also be enzymes used for lysis depending on the samples type. Proteinase K is one of these, and actually works very well in these denaturing buffers; the more denatured the protein, the better Proteinase K works. Lysozyme, however, does not work in the denaturing and so lysozyme treatment is usually done before adding the denaturing salts.
One comment about plasmid preps, the lysis is very different than extraction for RNA or genomic DNA because the plasmid has to be separated from the genomic DNA first and if you throw in chaotropes, you’ll release everything at once and won’t be able to differentially separate the small circular DNA from the high molecular weight chromosome. So, in plasmid preps the chaotropes are not added until after lysis and the salts are used for binding. An excellent in-depth article on alkaline lysis is here and also another article on the difference between genomic DNA and plasmid is available for further reading.
Binding:
The chaotropic salts are critical for lysis, but also for binding, as we discussed. Additionally, to enhance and influence the binding of nucleic acids to silica, alcohol is also added. Most of the time this is ethanol but sometimes it may be isopropanol. The percent ethanol and the volume has big effects. Too much and you’ll bring in a lot of degraded nucleic acids and small species that will influence UV260 readings and throw off some of your yields. Too little, and it may become difficult to wash away all of the salt from the membrane.
The important point here is that the ethanol influences binding and the amount added is optimized for whatever kit you are using. Modifying that step can help change what you recover so if you are having problems and want to troubleshoot recovery, that can be a step to evaluate further.
Another way to diagnose problems is to save the flow-through after binding and precipitate it to see if you can find the nucleic acids you are searching for. If you used an SDS-containing detergent in lysis, try using NaCl as a precipitant to avoid contamination of the DNA or RNA with detergent.
Washing Steps:
Your lysate was centrifuged through the silica membrane and now your DNA or RNA should be bound to the column and the impurities, protein and polysaccharides, should have passed through. But, the membrane is still dirty with residual proteins and salt. If the sample was from plants, there will still be polysaccharides, maybe some pigments too, left on the membrane, or if the sample was blood, the membrane might be tinted brown or yellow.
The wash steps serve to remove these impurities. There are typically two washes, although this can vary depending on the sample type. The first wash will often have a low amount of chaotropic salt to remove the protein and colored contaminants. This is always followed with an ethanol wash to remove the salts. If the prep is something that didn’t have a lot of protein to start, such as plasmid preps or PCR clean up, then only an ethanol wash is needed.
Removal of the chaotropic salts is crucial to getting high yields and purity DNA or RNA. Some kits will even wash the column with ethanol twice. If salt remains behind, the elution of nucleic acid is going to be poor, and the A230 reading will be high, resulting in low 260/230 ratios.
Dry Spin:
After the ethanol wash, most protocols have a centrifugation step to dry the column. This is to remove the ethanol and is essential for a clean eluant. When 10 mM Tris buffer or water is applied to the membrane for elution, the nucleic acids can become hydrated and will release from the membrane. If the column still has ethanol on it, then the nucleic acids cannot be fully rehydrated.
Skipping the drying step results in ethanol contamination and low yields. I do not see ethanol absorbance on the Nanodrop, so it won’t show up in your readings. The main indicators of a problem are that when you try to load the sample onto an agarose gel, the DNA will not sink. Even in the presence of loading dye. Another indicator is that if you put the sample in the -20C, it doesn’t freeze.
Elution:
The final step is the release of pure DNA or RNA from the silica. For DNA preps, 10 mM Tris at a pH between 8-9 is typically used. DNA is more stable at a slightly basic pH and will dissolve faster in a buffer. This is true even for DNA pellets. Water tends to have a low pH, as low as 4-5 and high molecular weight DNA may not completely rehydrate in the short time used for elution. Elution of DNA can be maximized by allowing the buffer to sit in the membrane for a few minutes before centrifugation.
RNA, on the other hand, is fine at a slightly acidic pH and so water is the preferred diluent. RNA dissolves readily in water.
What other things can go wrong:
Low yields: If you experience yields lower than you expected for a sample, there are many factors to think about. Usually it is a lysis problem. Incomplete lysis is a major cause of low yields. It could also be caused by incorrect binding conditions. Make sure to use fresh high quality ethanol (100% 200 proof) to dilute buffers or for adding to the binding step. Low quality ethanol or old stocks may have taken on water and not be the correct concentration. If the wash buffer is not made correctly, you may be washing off your DNA or RNA.
Low Purity: If the sample is contaminated with protein (low 260/280) then maybe you started with too much sample and the protein was not completely removed or dissolved. If the sample has poor 260/230 ratio the issue is usually salt from the bind or the wash buffer. Make sure that the highest quality ethanol was used to prepare wash buffers and if the problem continues, give the colun an additional wash.
Some samples have a lot more inhibitors compared to others. Environmental samples are especially prone to purity issues because humic substances are solubilized during extraction. Humics behave similarly to DNA and are difficult to remove from the silica column. For this type of sample, specialized techniques exist to remove the protein and humics prior to the column step.
Degradation: This is more of a concern for RNA preps and an article that gives specific advice is here. Mainly with RNA, degradation occurs from inproper storage of the sample or an inefficient lysis, assuming of course that you eluted with RNase-free water. For DNA preps, degradation is not a huge problem because for PCR, the DNA can be sheared and it works fine. But if you were hoping to not have so much sheared DNA, then you may have used too strong a lysis method.
PCR Clean-up Special Considerations: The easiest of all the techniques, because it is simply adding a high concentration of binding salts (typically between 3-5 volumes of salt per volumes of PCR reaction) and centrifugation through the column. So when PCR Clean-up kits fail, it can be particularly frustrating. The first question I ask people is “did you check the results of the PCR on a gel?” because you cannot UV check a PCR reaction and have an accurate reading. There is way too much in a PCR reaction absorbing UV at 260: nucleotides, detergents, salts, and primers. In my experience, a failure of a PCR clean-up kit to work frequently is caused by a PCR reaction that has failed and so there was nothing to clean up. But if you know you had a strong PCR product, the best approach is to just save your flow-through fraction after binding. If the DNA doesn’t bind, that’s where it is. You can always rescue it and then clean it up again. And then call tech support and ask for a replacement kit.
These kits make the whole process much easier and faster than the methods of old, when things are going well, but the downside of using a kit is that we don’t always know what is in the mysterious and proprietary set of solutions that each company uses in its kit, which makes troubleshooting more difficult.
So in this article, I’ll explain in some detail how silica spin filter kits work and what is going on at each step. I’ll also go over some common problems specific to silica columns that can be overcome or avoided with just a little extra understanding.
Lysis:
The lysis formulas may vary based on the whether you want DNA or RNA, but the common denominator is a lysis buffer containing a high concentration of chaotropic salt. Chaotropes destabilize hydrogen bonds, van der Waals forces, and hydrophobic interactions. Proteins are destabilized, including nucleases, and the association of nucleic acids with water is disrupted setting up the conditions for the transfer to silica.
Chaotropic salts include guanidine HCL, guanidine thiocyanate, urea, and lithium perchlorate.
Besides the chaotropes, there is usually some detergents involved, to help with protein solubilization and lysis. There can also be enzymes used for lysis depending on the samples type. Proteinase K is one of these, and actually works very well in these denaturing buffers; the more denatured the protein, the better Proteinase K works. Lysozyme, however, does not work in the denaturing and so lysozyme treatment is usually done before adding the denaturing salts.
One comment about plasmid preps, the lysis is very different than extraction for RNA or genomic DNA because the plasmid has to be separated from the genomic DNA first and if you throw in chaotropes, you’ll release everything at once and won’t be able to differentially separate the small circular DNA from the high molecular weight chromosome. So, in plasmid preps the chaotropes are not added until after lysis and the salts are used for binding. An excellent in-depth article on alkaline lysis is here and also another article on the difference between genomic DNA and plasmid is available for further reading.
Binding:
The chaotropic salts are critical for lysis, but also for binding, as we discussed. Additionally, to enhance and influence the binding of nucleic acids to silica, alcohol is also added. Most of the time this is ethanol but sometimes it may be isopropanol. The percent ethanol and the volume has big effects. Too much and you’ll bring in a lot of degraded nucleic acids and small species that will influence UV260 readings and throw off some of your yields. Too little, and it may become difficult to wash away all of the salt from the membrane.
The important point here is that the ethanol influences binding and the amount added is optimized for whatever kit you are using. Modifying that step can help change what you recover so if you are having problems and want to troubleshoot recovery, that can be a step to evaluate further.
Another way to diagnose problems is to save the flow-through after binding and precipitate it to see if you can find the nucleic acids you are searching for. If you used an SDS-containing detergent in lysis, try using NaCl as a precipitant to avoid contamination of the DNA or RNA with detergent.
Washing Steps:
Your lysate was centrifuged through the silica membrane and now your DNA or RNA should be bound to the column and the impurities, protein and polysaccharides, should have passed through. But, the membrane is still dirty with residual proteins and salt. If the sample was from plants, there will still be polysaccharides, maybe some pigments too, left on the membrane, or if the sample was blood, the membrane might be tinted brown or yellow.
The wash steps serve to remove these impurities. There are typically two washes, although this can vary depending on the sample type. The first wash will often have a low amount of chaotropic salt to remove the protein and colored contaminants. This is always followed with an ethanol wash to remove the salts. If the prep is something that didn’t have a lot of protein to start, such as plasmid preps or PCR clean up, then only an ethanol wash is needed.
Removal of the chaotropic salts is crucial to getting high yields and purity DNA or RNA. Some kits will even wash the column with ethanol twice. If salt remains behind, the elution of nucleic acid is going to be poor, and the A230 reading will be high, resulting in low 260/230 ratios.
Dry Spin:
After the ethanol wash, most protocols have a centrifugation step to dry the column. This is to remove the ethanol and is essential for a clean eluant. When 10 mM Tris buffer or water is applied to the membrane for elution, the nucleic acids can become hydrated and will release from the membrane. If the column still has ethanol on it, then the nucleic acids cannot be fully rehydrated.
Skipping the drying step results in ethanol contamination and low yields. I do not see ethanol absorbance on the Nanodrop, so it won’t show up in your readings. The main indicators of a problem are that when you try to load the sample onto an agarose gel, the DNA will not sink. Even in the presence of loading dye. Another indicator is that if you put the sample in the -20C, it doesn’t freeze.
Elution:
The final step is the release of pure DNA or RNA from the silica. For DNA preps, 10 mM Tris at a pH between 8-9 is typically used. DNA is more stable at a slightly basic pH and will dissolve faster in a buffer. This is true even for DNA pellets. Water tends to have a low pH, as low as 4-5 and high molecular weight DNA may not completely rehydrate in the short time used for elution. Elution of DNA can be maximized by allowing the buffer to sit in the membrane for a few minutes before centrifugation.
RNA, on the other hand, is fine at a slightly acidic pH and so water is the preferred diluent. RNA dissolves readily in water.
What other things can go wrong:
Low yields: If you experience yields lower than you expected for a sample, there are many factors to think about. Usually it is a lysis problem. Incomplete lysis is a major cause of low yields. It could also be caused by incorrect binding conditions. Make sure to use fresh high quality ethanol (100% 200 proof) to dilute buffers or for adding to the binding step. Low quality ethanol or old stocks may have taken on water and not be the correct concentration. If the wash buffer is not made correctly, you may be washing off your DNA or RNA.
Low Purity: If the sample is contaminated with protein (low 260/280) then maybe you started with too much sample and the protein was not completely removed or dissolved. If the sample has poor 260/230 ratio the issue is usually salt from the bind or the wash buffer. Make sure that the highest quality ethanol was used to prepare wash buffers and if the problem continues, give the colun an additional wash.
Some samples have a lot more inhibitors compared to others. Environmental samples are especially prone to purity issues because humic substances are solubilized during extraction. Humics behave similarly to DNA and are difficult to remove from the silica column. For this type of sample, specialized techniques exist to remove the protein and humics prior to the column step.
Degradation: This is more of a concern for RNA preps and an article that gives specific advice is here. Mainly with RNA, degradation occurs from inproper storage of the sample or an inefficient lysis, assuming of course that you eluted with RNase-free water. For DNA preps, degradation is not a huge problem because for PCR, the DNA can be sheared and it works fine. But if you were hoping to not have so much sheared DNA, then you may have used too strong a lysis method.
PCR Clean-up Special Considerations: The easiest of all the techniques, because it is simply adding a high concentration of binding salts (typically between 3-5 volumes of salt per volumes of PCR reaction) and centrifugation through the column. So when PCR Clean-up kits fail, it can be particularly frustrating. The first question I ask people is “did you check the results of the PCR on a gel?” because you cannot UV check a PCR reaction and have an accurate reading. There is way too much in a PCR reaction absorbing UV at 260: nucleotides, detergents, salts, and primers. In my experience, a failure of a PCR clean-up kit to work frequently is caused by a PCR reaction that has failed and so there was nothing to clean up. But if you know you had a strong PCR product, the best approach is to just save your flow-through fraction after binding. If the DNA doesn’t bind, that’s where it is. You can always rescue it and then clean it up again. And then call tech support and ask for a replacement kit.
