Saliva preservation substrate - lysis, antisepsis, and buffer?

[trooogdooor]

Hello, noble nerds,

 

I ask for your advice on a high-stakes problem.

 

I'm going to be collecting a large number of saliva samples, adding a preservation liquid to them, and then transporting them by mail at room temperature. The liquid basically needs to do three things:

 

1) lyse the saliva cells

2) provide anti-microbial protection (gram positive, gram negative, fungal)

3) buffer the DNA for preservation.

 

There are already some commercial vendors of such a preservation liquid. But they're extremely expensive. 15+ USD for a single sample! And I need to transport thousands, this is just beyond my budget.

 

 

*

So, I've stitched together the below lysis solution. It's meant to store the DNA for ca. 3 months.

 

 

• 50 mM Tris Hcl, pH 8.0 (buffer)
• 1 mM EDTA (buffer)
• 1% (v/v) Triton X-100 (lysis)
• 1% (w/v) sodium deoxycholate (lysis)
• 100 mM NaCl (lysis)
• 10 mM DTT (lysis)
• 0.1% sodium azide (biocide)

 

 

 

 

What I'm not sure about is, will this be antimicrobial enough? Sodium Azide kills gram-negatives, but what about gram-positives and fungals, should I add something more for those?

Am I missing some obvious and elementary reason why this wouldn't work?

Of course, I intend to do testing before I use it on scale, but there's no sense in wasting time if I'm doing something obviously stupid.

 

 

Thanks, guys.

Edited April 4, 2012 by trooogdooor

[CharonY]

It really depends how clean you can work. Your buffer seems a bit like a mix of everything which to me does not make terribly lot of sense. What you want to get rid of are DNAase activities. Bacteria are less of a problem when a) working clean and b) using a buffer that does not promote growth. My take is that freezing your saliva samples would be your best bet if it is only for a few month. Within the cells the DNA is nicely protected. If you need pure DNA you have to make additional extraction steps (e.g. with phenol) anyway, which nicely kills bacteria and also removes DNAse (and other proteins, of course). Simple Tris-EDTA buffer works well after that.

However, without removal of proteins, your DNA is going to degrade within your cell lysate.

 

Just FYI, the low amount of NaCl is not going to pertain to cell lysis, nor is DTT (a reducing agent). Mixtures of detergent are usually not helpful, it is easier to stick to one. EDTA is not a buffer, but functions as chelator. It reduces DNAse activity (by binding necessary divalent ions), but in a cell lysate there is likely too much for it to do enough.

[trooogdooor]

@CharonT: You make some solid points. I've removed DTT and NaCl, and added a new biocide, aimed at virus:

 

 

• 50 mM Tris Hcl, pH 8.0 (buffer)
• 1 mM EDTA (chelation buffer)
• 1% (v/v) Triton X-100 (lysis)
• 1% (w/v) sodium deoxycholate (lysis)
• 0.1% sodium azide (biocide)
• 1% sodium dodecyl sulphide (biocide, antiviral)

 

 

The process is like this:

1) Spit kit gets sent out to donor.

2) Donor spits into tube, adds preservative, sends it back.

3) Lab gets the kit and processes it, possibly 2+ months later.

So, freezing just can't be done, unfortunately.

*

Apparently, there are many commercial services who have this problem cracked. Check this, for example:

http://www.norgenbio...duct.php?ID=385

SALIVA DNA PRESERVATIVE

Norgen's Saliva DNA Preservative is an aqueous storage buffer designed for rapid cellular lysis and subsequent preservation of DNA from fresh specimens. The buffer prevents the growth of Gram-negative and Gram-positive bacteria and fungi, and also inactivates viruses allowing the resulting non-infectious samples to be handled and shipped safely. In addition, the buffer eliminates the need to immediately process or freeze samples and allows the samples to be shipped to centralized testing facilities at ambient temperature. The components of the buffer allow samples to be stored for more than 2 years without any detectable DNA degradation.

 

They've released papers showing virtually no DNA degradation after several years in this preservative. So, it's definitely doable, the question is only how?

Edited April 4, 2012 by trooogdooor

[CharonY]

My guess is that they provide a powerful protease inhibitor to the mix. I am not sure whether there are actually other companies selling competing kits. Others like Qiagen have storage solution for samples (not specific to saliva), but again, the formulation is proprietary. In one case it is a high-salt formulation.

We generally process frozen samples (for other biomolecules though) and I am not sure how many labs are using the kit. It sounds nifty for DNA, though.

Also, I am not sure that there is no degradation. Not massive maybe, but if all you do is PCR that is still fine (different story if you want whole genome sequencing).

[trooogdooor]

You guess correctly, it's for whole genome sequencing and genotyping, unfortunately we can't PCR it for that.

 

They do show very slight degradation in their DNA over a period of years, right. For our purposes, we plan to store the DNA for ca. 3 months, though, that's really all we need.

 

 

Reading up on this, I do agree that the addition of protease inhibitors would be wise for long term storage. All the solutions with PI's I read up on basically say "protease inhibitor cocktail", or something equivalent, and it seems that the default cocktail is something like: 1 mM4-(2-aminoethyl)-benzesulfonyl fluoride–HCl, 2 μg/ml aprotinin, 100 μM leupeptin, 1 μg/ml cystatin, and 1 mM benzamidine .

 

 

So I guess this makes my preservative:

 

 

 

• 1% (v/v) Triton X-100 (lysis)
• 1% (w/v) sodium deoxycholate (lysis)
• 0.1% sodium azide (biocide)
• 1% sodium dodecyl sulphide (biocide, antiviral)
• 50 mM Tris Hcl, pH 8.0 (buffer)
• 5 mM EDTA (chelation buffer, PI)
• 1 mM AEBSF, also known as 4-(2-aminoethyl)-benzesulfonyl fluoride–HCl (PI)
• 2 μg/ml aprotinin (PI)
• 100 μM leupeptin (PI)
• 1 μg/ml cystatin (PI)
• 1 mM benzamidine (PI).

 

Sensible?

 

I've also seen mention of Bestatin, E64, and Pepstatin A. Do you think that adding more PI's is the safer course of action when in doubt?

Edited April 5, 2012 by trooogdooor

[Arete]

What sort of WGS? Most platforms (Illumina Hiseq, Roche454, PacBio, etc) you're looking at a bare minimum of 10 μg of template - which will dictate the type of collection you use.

 

Options I've used for collecting parastie infected bodily fluids have been FTA cards - but the yield of parasite DNA is usually too low for WGS, but adequate for microarrays.

http://www.fishersci...2329_29104_-1_0

 

We've also used RNAlater.

http://www.qiagen.co...bilization.aspx

 

And liquid N2 Dewars.

Edited April 5, 2012 by Arete

[CharonY]

Considering that especially for 454 the DNA quality is a huge issue I would be kind of surprised if it works out. RNAlater was the high-salt formulation I was referring to. In that case one would store the tissue sample in it, but even so at 4° usually it is not more stable than a month.

 

With 454 around 3-5 µg could be enough, so from 1-2 ml saliva, freshly prepared, it may be enough. However, purity and integrity could be an issue. I doubt that you could get enough with sufficient purity and integrity out of it after prolonged storage at RT. In fact, I would actually check whether someone did next-gen sequencing, or at least built and checked libraries using that kit. PCR is a rather weak quality control and I would not trust a standard agarose gel to give me sufficient info on quality.

[trooogdooor]

We'll be using Illumina Hiseq, right.

 

We'll actually be aiming for 4-5ml of saliva, rather than the industry standard 2ml, for the reasons you cite.

 

I know for sure that next-gen sequencing is done using these kits, it's the norm for some sequencing studies. These RT saliva kits are also used by all the major genotypers (23andme etc), which of course use an order of magnitude less DNA; but the studies claim essentially unchanged DNA levels for years.

 

I'm curious, what do you think that Norgen and Oragen do to get this result, if not something similar to the preservative I posted above?

They seem to have published papers demonstrating stability at RT for years, how do you think they achieve this?

[CharonY]

Maybe I should say that I was thinking in terms of genome sequencing (as you mentioned whole-genome sequencing), however it appears you are more interested in genotyping (e.g. SNP analysis etc.). For the former, having a very intact and clean chromosomal DNA is important to be able to do decent assembly. What I have seen from the respective manufacturers were not information regarding the quality of the genome except PCR and a simple gel. Both of which show that you do not have totally degraded chromosomal DNA but do not give sufficient information regarding the integrity. It is possible that I overlooked that part, but human whole genome sequencings are done rather rarely.

 

 

But again, if you do not want to assemble the whole genome, which is highly unlikely considering the number of samples you want to run, it is much more feasible. In the end, I have not idea how the mixture is going to interact (with each other and the cells) and I would resign to a testing run. E.g. preserve it for a few months, then run e.g. PFGE or even a shotgun run with a sample and check the results.

 

23andme are doing microarrays, IIRC, so again different needs as opposed to whole-genome sequencing. And in fact, all commercial labs are just doing genotyping. The postanalysis for whole genome sequencing is still just too time and money-consuming to have a high throughput, especially with the next-gens.

[trooogdooor]

Hi Charon,

 

We will be doing some genotyping (SNP analysis, etc), but we will also be doing WGS, and the WGS part is crucial! You're right that it's a large number, we'll be doing hundreds, and thousands in a few years - that's also why I started out saying that my question is "high-stakes".

 

The WGS aspect is the reason I am concerned about DNA integrity.

Certainly, the main purveyors of the technology, like Oragene and Norgen, have used their RT saliva kits for WGS:

 

 

http://media.completegenomics.com/documents/Saliva_Application_Note.pdf

http://www.dnagenotek.com/ROW/pdf/MK-00014.pdf

 

 

 

I think your answer is pretty much the same one as I would give, though - IE, go forth and have a test run with the preservative, as devised. It's what needs to be done no matter what, I just want to make sure that my proposed list of chemicals isn't crazy.

 

• 1% (v/v) Triton X-100 (lysis)
• 1% (w/v) sodium deoxycholate (lysis)
• 0.1% sodium azide (biocide)
• 1% sodium dodecyl sulphide (biocide, antiviral)
• 50 mM Tris Hcl, pH 8.0 (buffer)
• 5 mM EDTA (chelation buffer, PI)
• 1 mM AEBSF, also known as 4-(2-aminoethyl)-benzesulfonyl fluoride–HCl (PI)
• 2 μg/ml aprotinin (PI)
• 100 μM leupeptin (PI)
• 1 μg/ml cystatin (PI)
• 1 mM benzamidine (PI).

 

I'll be doing that in a week or two. And maybe call you Chiron from now on, thanks for your advice.

Edited April 9, 2012 by trooogdooor

[CharonY]

I see. I have not found paper with regards to WGS, but I stand corrected. However, after a quick view over the notes it appears to me that they do not address whether they were able to assemble properly. I do not have time to delve into the lit. right now, but it appears to me that they just blasted sequences against the human database and assigned whether it was a hit or not. Again, more typical for genotyping workflows. Note that WGS is also often used in this context, i.e. do one (or several runs) and then utilize the data for searches without too much of an assembly and let gaps be gaps (I would call that high-throughput genotyping). However, since I am an "omics" guy I specifically refer to it in terms of reconstructing the complete genome (just so that there are no misunderstandings).

 

You could check the associated papers (if any) whether and how well the assembly went, or maybe ask the technical support of these company for that data. If they think they can sell something they are often willing to tell you stuff (short of the actual formulation that is).

 

And maybe call you Chiron from now on, thanks for your advice.

Huh? Is a horse growing out of my lower back? I should better check...

 

Edit: I was curious and quickly peeked into the Lam et al. paper cited in the notes and there they processed saliva directly. Also note that with next-gens a single run is usually insufficient to get enough coverage to assemble the whole genome.

Edited April 9, 2012 by CharonY