[Dominik]

Hi all,

I’m trying to modify some of your protocols to more closely fit my needs (soil, feeces, whole specimen and so on). I succeeed in doing so for extractions from homogenized insects (TNES + Prot K. Lysis, followed by your cleanup protocol). However I’m struggling with some of the protocols especially when it comes to TNA. Would you mind sharing some of the mechanisms behind the protocols or supply a source where I can look these up?
For example: What final concentration of GITC is needed to bind TNA to the silica, what concentration of Isopropranol is needed (and can this also be substituted by Ethanol?), do the concentrations change when using Gu-HCl and so on.

I planing to extract DNA/RNA from Soil, whole specimen, homogenized specimen (mostly insects) and nitrocellulose filters and probably protists / bacteria / viruses on leafs. I guess this all can be done with your protocols with Soil probably being the most challenging.

€: Any recommendations on washing buffers (instead of 80% EtOH that has to be prepared freshly). I will have to process a few thousand samples and preparing fresh EtOH before each extraction is not to practical.

Best, Dominik

[admin]

Ahoy,

For information on Ethanol/Iso. I recommend this:

DNA Precipitation: Ethanol vs. Isopropanol

The gist is that both work fine but you only need about 35% Iso whereas you need roughly 75% ethanol. Since you already have the lysis buffer and bead storage buffer in your tube you would have to add an inconvenient amount of ethanol. If your vessel is big enough there is no reason not to use it.

From my understanding the GITC is mostly responsible for lysis of cells and proteins as well as inactivation of DNAse/RNAse. A GU-HCL buffer should work just as well and you will find plenty of good recipes out there.

The actual binding is a result of the precipitated TNA being negatively charged in slightly basic pH and so is the Bead surface of silica beads, cations such as Na+ are needed as a bridge to allow for binding.
Most of the buffer is meant to lyse, keep the TNA safe and denature proteins. Normal table salt works fine for the binding aspect, it just cant do any of the other things (or at least not as well).
https://en.wikipedia.org/wiki/DNA_separation_by_silica_adsorption

The EtOH does not have to be fresh, not at all. The only issue is that if it goes below 70% it will prematurely re-elute your TNA. And since EtOH is hygroscopic and readily evaporates it has a strong tendency to loose concentration over time.
If it is in a small vessel (little surface area) in a reasonably cool spot (RT is fine just not in the sun) I would feel perfectly confident using it for a week or even more.  You could wash with Isoprop for all of the reasons I wrote above but this will not wash out salts very well.

I hope that clears things up. If you need more information, simply search for articles on DNA – silica/silica membrane binding. The chemistry/physics behind silica spin column binding and silica beads are essentially the same.

Cheers
Tim M

 

 

[Dominik]

Hi Tim,

thank you for your prompt and detailed answer, yet there remain some uncertainties, which I hope can be resolved. As far as I’m concerned, after whichever lysis, the binding of the DNA/RNA/TNA to the silica (beads / column) is the critical step. After I got my NA to bind to the silica I’m somewhat safe. Therefore I’m trying to figure out the main drivers of the binding:

1. Salt concentration
2. Ethanol / Isopropanol concentration
3. pH?

1. I looked through the protocols and found different final concentrations of salt in the binding procedure. For 4.1 it ranges from 2M – 0.6M while at 0.6M we loose some of the short fragments, which is intended in this case. In protocol 6.7 and 8.2 the binding happens at 1.3M – 1.6M of GITC, for the RNA binding after the DNase treatment it’s 0.8M. I’d conclude that the concentration is not that important as long as it is high enough (probably above 0.8 to bind the full size range), am I right with that?

2. Ethanol / Isopropanol seems to be clear. EtOH above 75%, Isopropanol above 35%. What still confuses me is, that in protocol 4.1 you mix 1 volume of samples with 2 volumes of binding buffer, so you get a final concentration of ~60% EtOH, which should exclude smaller fragments. I’m using this protocol for DNA extractions from tissue, will I loose any of the DNA by doing so? Getting great results so far.

3. pH: This is where I’m most puzzled. You wrote that the binding happens at slightly basic pH, which is certainly not true for the 4.1 and the RNA binding step in protocol 8.2 (which both have a pH below 7). Is there any explanation for this or is pH not the main driving factor? In most spin-column kit (MN, Quiagen) it is even advised to keep the pH below 6.5 indicated by the yellow color of the binding buffer (which I’m guessing is phenol-red?)

Final question:

The reason I’m asking is that I’m trying to modify some of the protocols. For some samples it is easier for me the go with a TNES/Prot. K lysis and bind the DNA from the lysate to the beads with your protocol 4.1. However this does not work well for soil and water samples, I’m getting some nasty inhibitors also indicated by a brown color of the finale DNA elution. When I’m trying to lyse this samples with GITC it gets even worse: The lysate somehow takes away the magnetic properties of the beads (my guess is, that something is co-precipitated and also binds to the beads as well). I found some protocols to remove inhibitors (some proteins, humic substances) from lysates and want to try these. However this will change the salt concentrations in the lysate if I’m doing the GITC lysis. For inhibitor removal, I’m going to use the Flocculant Solution from this protocol (https://www.protocols.io/view/mu-dna-a-modular-universal-dna-extraction-method-a-qn9dvh6/guidelines). Since we are mainly interested in DNA I think I’m fine doing this with TNES lysed samples, but will need to adjust for the GITC lysis if I also need RNA.

Any advice on that or some more insight in how you came up with the protocols?

best Dominik

[admin]

Ahoy,

I have recently performed a vast amount of empirical trials on silica binding conditions and I can tell you with complete confidence that you will not be able to fully theorize the best conditions.
I tried different batches of homemade Beads, different Commercial beads, different lysis/binding buffers, different alcohols and percentages and very importantly different concentrations of Beads and sample input.

I have completely failed to find a ratio that gives the best possible result.
Any variation favors one aspect over another, be it more or less RNA/DNA, more of both, a shift towards certain lengths, or clearly working better with one type or even batch of beads than another.

1: Yes, I believe so.

2:On this I have to refer to my rant above. Just yesterday I tried a binding/lysis buffer mainly consisting of 6M Urea, 1.7M NaCl and a small amount of LiCl (sorry, but I don’t want to fully disclose at this stage). I then tested how A) No-alcohol B) 36% EtOH C) 36% Isoprop. would influence yield.  I did most definitely not expect the Ethanol at this low percentage to give me the best DNA yield, the Isoprop to give me the best TNA yield with a strong bias towards RNA and the no alcohol to work at al. The no alcohol yield was bad and mostly DNA but still even though there was only about 30sec for the purely salt induced precipitation I got some DNA out.
You will also find that the EtOH and Iso. percentages mentioned on bitsizebio and other pages always state that that is the required concentration in presence of a mere 0.5M salt…

3: Me too. This one is indeed puzzling me so much I am tempted to say nothing.
All I can do is throw you a few bits of information that may or may not be involved in the true explanation.
RNA and DNA have pKA, meaning that they will be charged/not charged at different buffer pH values.  This is crucial for binding but more is not necessarily more.
Silica seems to show different binding affinities for TNA at different pH.
Proteins also bind to silica at certain pH. (https://link.springer.com/article/10.1007/s00396-015-3754-x)
Tooo many factors…. If you find something that works just go with it. The BOMB protocols are largely based on pre existing protocols. We tried not to reinvent the wheel and did not test, alter adjust, every aspect.
Some more factors:
Beads, especially the homemade ones, can have rough surfaces that may snag proteins even when no actual static binding occurs.
DNA and RNA extraction seems to be heavily influenced by NT concentration. Somehow both are less likely to precipitate when at very low concentration. Especially for RNA extractions it is quite common to add carrier molecules such as glycogen, NTs from non related species or linear polyacrylamide.

From personal experience, performing GITC TNA extractions of Proteinase K digested Zebrafish samples (even. with fancy commercial beads) I end up with a completely grey elution.
I am pretty sure that I am somehow extracting eumelanine with my TNA. I really really do not like this, but so far it has not interfered with qPCR or next gen sequencing in any way.

Salt concentration wise you can use a 1.5x GITC buffer  when TNES+Prot.K digesting first, at least that is what I do.
I adjust Sample and Prot. K volume to 200ul with TNES and then add 400ul 1.5x GITC post digestion.  However I have also done this with 1x GITC on days where the lab was cold and my 1.5x GITC crystalized and it worked just as well. This procedure is unfortunately prone to a really really annoying phenomenon where  everything turns into white slurry once you add the 1.5x GITC (performing everything at ~50 ˚C helps).

Final question: Mostly answered above but also I have to say that I personally do not know. I used to be just the site admin … Big team, others came up with the protocols.

I have a suggestion for soil that is close to something I  am currently trying. No guarantees but not much work either. (You would need LiCl for it.):
Lyse soil in 200ul GITC (5min) > quick spin down in a normal table centrifuge > transfer supernatant, discard debris > add 400ul of 6M LiCl and beads* store in fridge over night > shake, magnet and discard supernatant.
* ideally from the 50:50 stock so that you do not lower LiCl molarity.
LiCl does not precipitate proteins. LiCl precipitates RNA and to some extend DNA without the need for alcohol. If you get too little DNA out you can add Ethanol after the 24h incubation. If you choose the right amount of EtOH it should give you a pretty  good TNA extraction where proteins stayed in solution at all times thus being less likely carried along.

I know none of this represents the clear answers you were hoping for but this is all I got to offer.

Cheers

Tim M

 

 

 

 

[Author]

Posts