Methods for mtDNA purification suggested on the Evoldir mailing list

Subject: Other: mtDNA extractions/kits--answers
Date: Mon, 9 Jun 2003 02:19:51 -0400 (EDT)

I recently posted a query regarding isolation of 'pure' mtDNA from tissue,
blood, etc., for use in PCR-based methods. I am grateful for all your
responses-- a big thank you. It is a pleasure to correspond with members of
this group. The topic generated a lot of interest, and I have included a
summary of the responses below. I apologise for any cross-postings.
Thank you!
Gabriela Ibarguchi

Department of Biology, Queen's University
Kingston, Ontario, Canada

=======================================

Responses follow:

I used to use an old sucrose gradient or "pad" method (swinging bucket
ultracentrifuge required) that produced very clean mtDNA preps for bird
tissues (blood was not really doable). It required relatively fresh
tissues, or ones put into a sucrose/mannitol buffer, then frozen. The
mitochondria were isolated and could be run through 2-3 times to really
clean it up, followed by DNA isolation.

Robert C. Fleischer
_______________________________

We used the Wizard miniprep kits (Promega). Other such kits (I believe
Qiagen
also makes them) work well too. Cesium chloride is a huge pain to work
with; although this way is more expensive, it is not nearly so
cumbersome or toxic. MF Smith (UC Berkeley) was picking up a nuclear
duplication of her mitochondrial gene, and she published a paper on
that. I believe that paper included a more detailed description of her
strategy (which is what we used) for excluding nuclear DNA, so that she
could retrieve her true mtDNA gene copies without the nuclear copy.

Tina Hambuch (protocol by Peg Smith)
______________________________________________

WAKO Chemicals has a kit that uses a modified differential
centrifugation procedure to obtain "enriched" mtDNA. What you get is
not as pure as by using CsCl, but it is a lot faster. I have had rare
cases of "presumed" pseudogenes when working with WAKO extractions,
though.

Francesco Nardi
_____________________________________________
Reference below includes a quick simple method. You need a Beckman TL-100
ultracentrifuge.

S. M. Carr and O. M. Griffith (1987). Rapid isolation of animal
mitochondrial DNA in a small fixed-angle rotor at ultrahigh speed.
"Biochemical Genetics," 25:385-390.

Any of the kit-based methods will let in nuclear contamination, the above
method gives good separation of mt and nuc bands. We're using it to purify
reference individuals for mtDNA clones. Never thought I'd be doing CsCl
gradient stuff again.

Steve Carr
________________________

I think you will have a best chance to amplify the mtDNA in two long PCR
fragements. You could use primers in very conservative regions like 12S and
16S and obtain your mtDNA.

Gustavo Ybazeta
______________________________________

I noticed that numt are more common in blood tissues. I had that problem
with some of my blood samples (from yellow warblers) but it never
happened with my quill samples. If you can't find a protocol that
guaranties that the samples are free of nuclear DNA, I would suggest you
use quills and extract the DNA with the SIGMA or QIAGEN extraction kits.
When I did not have any quill duplicates of my blood samples, I used a
different set of primers that amplified a larger fragment that contained
the sequence I was interested in.

Marylene Boulet

_________________________________________
I have used a sucrose gradient centrifugation in the past to isolate whole
mitochondria, from which you can obtain a "mtDNA enriched" extract. This is
often done as a preliminary step before a CsCl protocol, but you can stop
after the sucrose gradient and have a reasonably good mtDNA prep. The
following references about numts may be helpful - there are some other
approaches besides purifying mtDNA.

Sorenson, M.D. & T.W. Quinn. 1998. Numts: A challenge for avian systematics
and population biology. The Auk 115: 214-221.

Sorenson, M.D. & R.C. Fleischer. 1996. Multiple independent transpositions
of mitochondrial DNA control region sequences to the nucleus. Proceedings of
the National Academy of Science USA 93:15239-15243.

Michael Sorenson
_______________________________________________

Back before PCR was well-established, a number of fish genetics labs used a
technique developed by Bob Chapman and Denny Powers to carry out
mitochondrially-enriched extractions. It involved gently homogenizing
several grams of liver tissue in a TEKS buffer (TE with KCl and sucrose),
going through a series of low- and high-speed centrifugations to remove
nuclei and pellet mitochondria, then lysing mitochondrial membranes with a
nonionic detergent (NonIDet). The resultant DNA wasn't pure mitochondrial
DNA, but it was highly enriched for mtDNA. Blood should work very well, and
potentially much better.

The reference for this method is:

Chapman RW, Powers DA (1984) A method for the rapid isolation of
mitochondrial DNA from fishes. Tech. Rep. No. UM-SG-TS-84-05. Maryland Sea
Grant Program, Chapel Hill, MD.

Although their method didn't do this, you might want to keep the pellet from
the first low-speed centrifugation in order to not lose most of the nuclear
DNA.

Another approach that I've read about but never tried was to lyse all
organelle membranes, then remove most of the nuclear DNA by alkaline
denaturation followed by abruptly reversing the pH, causing cross-linking of
nuclear DNA strands. The 'rafts' of nuclear DNA were then pelleted by
salt+centrifugation, whereas the mtDNA largely stayed in solution. I don't
have a reference for this; I came across it in a student thesis on European
arctic charr. You might be able to find something using a web search (?).

The extraction has reduced nuclear DNA content, but that some will probably
still be present. One gram of liver should still be adequate - I typically
used two to three grams of tissue for getting large amounts of mtDNA for
restriction digests, but Roy Danzmann at Guelph dropped down to roughly a
gram so he could process more samples in the big Sorvall centrifuge (pre-PCR
days!).

Chris Wilson
______________________________________________

When I worked with scallops, I used to purify mitochondria on a sucrose
gradient
before lysing to release the mtDNA (following a citation in: Wilding CS,
Beaumont AR, Latchford JW
Mitochondrial DNA variation in the scallop Pecten maximus (L) assessed by a
PCR-RFLP method
HEREDITY 79: 178-189 Part 2 AUG 1997).The one caveat is that I was working
with large quantities of
mitchondria rich muscles and therefore could see the mitochondria easily for
extraction. If you only have small quantities this may be more difficult.
During
my time working on this I recall a separate protocol for isolating
mitochondria
using mini sucrose gradients in microcentrifuge tubes. I can't remember
whether
I was sent the protocol by the investigator or whether he published this -
sorry! Anyway, It was John D.G. Boom who, at the time worked either at the
Bamfield Labs on Vancouver Island or at UBC. Try looking at this paper, it
might
have the methods in:

MITOCHONDRIAL-DNA VARIATION IN INTRODUCED POPULATIONS OF PACIFIC OYSTER,
CRASSOSTREA-GIGAS, IN BRITISH-COLUMBIA
BOOM JDC, BOULDING EG, BECKENBACH AT
CANADIAN JOURNAL OF FISHERIES AND AQUATIC SCIENCES 51 (7): 1608-1614 JUL
1994

Craig Wilding
___________________________________

I have used the following quick & easy protocol (below)
for the same purpose (getting mtDNA relatively free of nuclear). See
Sunnucks & Hales (1996) - you can download a PDF from my website (address
below) under 'aphids'.
This protocol does not make mtDNA as clean as CsCl, but CsCl is a pain in
the neck. You can usually dilute out any remaining nuclear DNA by finding
the most dilute PCR template solution that will give useable product. We
actually find that this often works even without the mtDNA purification -
general 'genomic' DNA (mt + nuc) can be diluted until the nuclear copies no
longer amplify but the mtDNA is still going strong. We have experienced
nuclear tranposed copies of mtDNA in getting on for a dozen species (of
invertebrates) now...!

Paul Sunnucks

Protocol: Tamura and Aotsuka (1988) mtDNA rapid (alkaline lysis) extraction

for 200ml
buffer A 0.25 M sucrose 17.1 g
10mM EDTA 4ml of 0.5M
30mM tris/HCl pH7.5 6 ml of 1M

buffer B 0.15M NaCl 6 ml of 5M
10mM EDTA 4ml of 0.5M

Put some eppendorfs in the freezer to chill (put microfuge in cold room,
or get rotor very cold). Don't leave microfuge in coldroom for any longer
than necessary.

1/ Get freshest material available. If frozen, material should be at -80
if possible.
Keep on ice while preparing.

2/ homogenize tissue in cold buffer A, in cold glass homogenizer; enough to
disperse, but do not over-do - makes available the whole mitochondria.
If you want to do lots, do lots of separate tubes rather than packing
loads of tissue into a few tubes. Eg for 1g aphids, I would do 6-8 tubes.
(Although 1-5g will give a lot of mtDNA, small amounts will be useable for
PCR: e.g. 50 Drosophila are supposed to give mtDNA visible even after
restriction, on agarose with ethidium bromide - this amount should be
enough for thousands of PCRs).

3/ Transfer to chilled eppendorfs by pouring carefully.

4/ Spin in cold microfuge (prechilled, in cold room) 2000 rpm for 2 minutes.

5/ Take off supernatant; repeat 4/-5/ until liquid is clear - this spins
out the cell debris, and leaving mitochondria in suspension.
(Can keep the mush - it contains a lot of nuclear DNA which can be salted
out).

6/ Spin 14000 rpm for 10 minutes to pellet mitochondria.

(OPTIONAL 7/ Resuspend in buffer A and respin to clean more, keep pellet)

8/ Resuspend pellet in 50µl of buffer B.

9/ Add 100 µl of 0.18M NAOH/1% SDS made fresh, gently vortex inverted, and
put on ice for 5 minutes. Lyses mitochondria and liberates closed circular
and supercoiled mtDNA (for many inverts, also some plasmids from
associated organisms)

(for 1.5 ml, 75µl of 4M NaOH, 60µl 25% SDS, make up to 1.5 ml with water -
put the 865µl water in first)

10/ Add 75µl 5M KoAc (pH5 potassium acetate, see Maniatis for recipe,
special way of making it), gently vortex inverted, and put on ice for 5
minutes.

11/ Spin 14000rpm 5 minutes.

12/ Standard phenol chloroform extraction, and DNA precipitation.
______________________________________________________________________

I have tried several "plasmid prep" kits but found lots of nuclear DNA in
the resulting
product. Apparently, the effectiveness of plasmid preps in removing
chromosomal DNA has to do with the chromosome being associated with or
attached to the cell wall in bacteria so it precipitates with (or can be
filtered out with) all the cellular debris.

Richard Broughton
_____________________________

The following reference can be found as a pdf in the Promega website:
http://promega.com/pnotes/43/43p10/43page10.html

Beckman K. B., Smith M., and Orrego C. (1993) Purification of Mitochondrial
DNA with WizardTM Minipreps DNA Purification System. Promega Notes Magazine
43: 10-15

Cheers,
Gabriela.