Rolling Circle Amplification of Skin Swab Specimens
Laboratory of Cellular Oncology, NCI, NIH
Bethesda, Maryland, USA
Citation: Schowalter et al (2010) “Merkel Cell Polyomavirus and Two Previously Unknown Polyomaviruses Are Chronically Shed from Human Skin”
Cell Host & Microbe7:509, PMID: 20542254
rev: 23 August 2012
Introduction: Rolling circle amplification (RCA) is a powerful technique for specific detection of circular DNA molecules – most notably, viruses with circular DNA genomes. The application of RCA to the study of circular DNA viruses has recently been reviewed by Reimar Johne and colleagues (PMID: 19375325). The technique relies on the extremely high processivity and strand-displacement activities of the mesophilic DNA polymerase of bacteriophage phi29. Once the polymerase has begun generating a new DNA strand, it essentially never drops off the template, allowing it to extend around a circular template for multiple passes, thus producing a long series of tandem repeats of the original circular template.
In this protocol, we incorporate an upstream processing step in which unwanted linear fragments of DNA in a starting sample are destroyed, while circular DNA molecules are spared. This enrichment for circular templates appears to be important for robust amplification of low copy number circular templates in complex samples. The protocol was originally developed for analysis of circular DNA viruses present in swabs wiped across the surface of healthy human skin. The starting swab material is, of course, heavily contaminated with uninteresting fragments of human DNA.
The DNA extraction and processing steps are performed more rapidly than some other protocols and utilize relatively gentle conditions in an effort to avoid DNA damage. The appearance of even a single nick or damaged base renders an entire template strand unsuitable for RCA. It is possible that some forms of DNA damage could be repaired using NEB “PreCR” treatment (NEB #M0309S). However, in a pilot experiment we found that DNA prepared using the methods below did not seem to benefit from treatment with the repair mixture – suggesting that the harvest procedure causes only minimal damage to the circular DNA of interest.
Although this protocol removes a large majority of linear fragments of human cellular DNA, it cannot remove circularly permuted human DNA (see PMID: 22403181), plasmids present in skin bacteria, or circular DNA bacteriophages. In some swab specimens, these unwanted DNA types dramatically outnumber HPV or HPyV DNA. The problem can be addressed by purifying virions by density gradient ultracentrifugation (see protocol).
Harvest:
•We have traditionally used wooden cotton-tipped swabs (Greiner Bio-One #421180). More recently, we have shifted to polyester swabs (Puritan #25-3316-U).
•Swabbing human subjects requires Institutional Review Board approval. Informed consent must be obtained from potential study volunteers.
•The sample is collected by asking the volunteer to wipe the swab back and forth across the skin surface of interest as vigorously as they’re comfortable with. It helps if the volunteer holds the swab close to the tip to avoid breaking the swab shaft. We try to sample later in the day to increase the amount of time during which shed material can re-accumulate after morning skin washing. It is helpful to gradually rotate the swab to saturate each surface with shed material. As far as we can tell, it doesn’t seem to matter whether the volunteer is wearing makeup.
1) Cut the end of the swab into a 1.5 ml microcentrifuge tube.
2) Add 150 µl of Digestion Buffer composed of:
25 mM Tris pH 8.0
25 mM EDTA
10 mM DTT (make fresh, Pierce #20291)
1% SDS
0.5% proteinase K (Qiagen #19131)
3) Incubate the suffused swab tips at 50ºC for 15 minutes
4) Add 900 µl of PB Buffer (Qiagen #28106). Place the tubes on a rocker for 10 minutes then centrifuge at 16,000 x g for 1 minute. Note: it is important to be sure that the low salt buffer in the swab tip has been displaced by the surrounding high salt PB buffer. Rocking and centrifuging seem to assist in this goal.
5) Transfer the swab tip into a purple Qiagen spin column (from kit #28106). Spin to extract residual fluid from the swab tip. This step, which can be technically challenging and might increase the risk of sequence contamination, is probably optional. In some pilot experiments, it seemed that a great majority of the DNA in the swabs escaped into the free PB buffer, such that the small amount of fluid retained in the cotton only contains a fraction of the total harvest.
6) Using a vacuum manifold (Qiagen #19413), feed the remaining PB fluid through the purple spin column. Try not to get fluid on the ledge at the top of the spin column. Use fresh Vac Connectors (Qiagen# 19407) to protect the column tip from contamination with lab plasmids.
7) Using the vacuum manifold, wash the spin column 1x with 500µl of PB buffer. Wash the spin column 2x with 750µl of PE buffer. Try not to let the resin remain dry for very long. Note that columns may have different flow rates, depending on factors such as how much makeup the subject was wearing.
8) Spin the columns for 5 minutes at room temperature to achieve complete dryness. Note: over-drying can result in reduced sample yield due to irreversible binding of DNA to the silica resin.
9) Transfer the spin columns into fresh microcentrifuge tubes. Add 75 µl of 0.2x TE (2 mM Tris pH 8, 0.2 mM EDTA). Let the tube sit for five minutes then spin to collect the eluate.
Sample Pre-Digestion:
•In the preprocessing step, the extracted DNA is digested with exonuclease V (trade name Plasmid Safe, Epicentre Biotechnologies #E3103K). The enzyme progressively degrades DNA ends. Circular dsDNA (no ends) is spared.
•Optional: the digestion of very long fragments of DNA is facilitated by addition of restriction enzymes that are unlikely to cut polyomaviruses or papillomaviruses. Both virus families are relatively deficient in CpG dinucleotides, perhaps as a mechanism for evading regulation by CpG methylation. Even in instances where CpG dinucleotides are present in the viral genome, they may be methylated and therefore protected from restriction digestion. Restriction enzymes such as AfeI, NotI and SalI have been used successfully. The new “HF” (high-fidelity) versions of NotI and SalI (www.NEB.comR3138S and R3189S) are preferable because they are less likely to damage the viral DNA and they are designed to work in NEB Buffer 4, which is compatible with Plasmid Safe. Addition of a restriction enzyme is likely to be helpful, but does make a dangerous untested assumption about as-yet-undiscovered viral sequences (i.e., that they do not contain unprotected cut sites for the restriction enzymes being used. The use of restriction enzyme digestion should be considered optional. An alternative would be to omit the restriction enzyme and extend the digestion period to give Plasmid Safe more time to digest very long linear DNA fragments. In general we perform separate digestions with AfeI and SalI-HF for each sample. AfeI has the useful feature of digesting human mitochondrial DNA, which might be a problem in samples involving live cells. Skin swab DNA seems not to have detectable amounts of intact circular mtDNA.
1) Transfer 25 µl of purified DNA sample (above) into an 0.5 ml microcentrifuge tube.
2) Add 50µl of digestion mix:
1.5x NEB Buffer 4
1.5 mM rATP
0.2% Plasmid Safe
0.1% restriction enzyme (optional)
3) Digest 30 minutes at 37ºC. Optional: Add EDTA to a final concentration of 20mM. Heat-kill the Plasmid Safe and restriction enzyme by heating 65ºC 10 minutes.
Note: In the next few steps, the DNA is ethanol precipitated. Efficient recovery of very small amounts of DNA is tricky but not impossible. It’s essential to use an optimized method presented in an old paper by Crouse and Amorese in the “Focus” newsletter Life Technologies used to publish. The paper is a great illustration of the idea that time-honored “standard” laboratory methods are sometimes not the product of extensive initial optimization. The Crouse & Amorese (1987) Focus9(2):3paper is recommended reading.
4) Add 38 µl of 7.5 M ammonium acetate. Mix by flicking the tube. Note: ammonium acetate powder is very hygroscopic. Old bottles may therefore contain a substantial mass of absorbed water. Thus, if ammonium acetate stock is made from old powder, it’s important to add extra to account for the extra water. Buying prepared 7.5 M stock solution from Sigma (Cat #A2706) is an easier approach.
5) Add 300 µl of 95% (190 proof) ethanol. Do not use absolute (200 proof = 100%) ethanol for this step. For unknown reasons the 200 proof ethanol our lab buys doesn’t work as well.
6) Invert the tube several times. Let stand at room temperature 1 hour then transfer to 4ºC overnight.
7) Precipitate the DNA for at least 16 hours. DNA is very stable in ethanol and the samples can be stored for up to two weeks at this step.
8) After precipitation, return the tubes to room temperature. It is very important that the precipitation be warmed back to room temperature. Crouse and Amorese speculate that the increased viscosity of cold ethanol solutions interferes with the migration of the precipitated DNA to the bottom of the tube during centrifugation.
9) Centrifuge the tubes at maximum speed (~16,000 x g) for 1 hour at room temperature.
10) Rapidly aspirate away the supe, carefully avoiding the side of the tube with the DNA pellet. The DNA pellet won’t be visible to the naked eye. Don’t worry if there’s still residual fluid clinging to the walls of the tube.
11) Add ~400µl of 70% ethanol. Close the lid and let the tube stand ~10 minutes to give the salts a little time to dissociate from the DNA pellet.
12) Centrifuge 10 minutes. Rapidly aspirate away the 70% ethanol wash then briefly re-centrifuge the tube to collect residual ethanol clinging to the walls. Carefully remove the residual ethanol with a 10 µl pipettor.
13) Allow the tube to air dry for 10 minutes. Occasional fanning of the tubes may be helpful for expelling residual ethanol vapor from the bottom of the tube.
14) Add 5µl* of TempliPhi (GE Lifesciences Sigma #25-6400-10) sample buffer. Flick the tube to distribute the buffer around the lower walls of the tube then centrifuge to consolidate. Flick again, then allow the DNA to dissolve by incubating ~10 minutes at room temperature.
*Note: Some protocols have suggested that RCA reactions on small amounts of template DNA run more efficiently in extremely small volumes. However, this approach must be balanced against the problems of evaporation effects and recovering only tiny amounts of amplified DNA for analysis. In our hands, the TempliPhi reaction can successfully be scaled up by a factor of 2-5x (i.e., total reaction volumes of 10-50µl, as opposed to the recommended total reaction volume of 10µl).
15) Centrifuge the tube briefly, then heat at 95ºC for 3 minutes. Return the tube to room temperature. According to the TempliPhi protocol, the rate of cooling doesn’t matter.
16) Prepare the RCA enzyme mixture, as instructed by the Templiphi kit, by mixing 5µl of reaction buffer + 0.2µl of enzyme per sample. Add 5µl of the enzyme mixture to each sample. Mix and reconsolidate. Incubate for 24 hours at 30ºC. It’s best to use a 30ºC incubator, such that the entire tube is the same temperature. In a water bath, the lid may be cooler than the tube walls, resulting in condensation of water vapor on the lid of the tube.
17) Heat inactivate 65º 10’.
18) Digest half (or all) of the RCA reaction with a restriction enzyme (or enzymes) of interest. For initial analysis, it’s good to use a frequent cutter like BstYI, which leaves a BamHI-compatible overhang. ApoI, which leaves an EcoRI-compatible overhang, is another frequent cutter enzyme to consider. Use of less promiscuous enzymes, such as EcoR1 or BamH1, could miss pathogens that happen not to have cut sites for the less common cutters.
19) Analyze the digest on a 2% agarose/TAE gel precast with GelGreen stain (www.PhenixResearch.com#RGB-4105). Image the gel on a blue light transilluminator (e.g., www.ClareChemical.com) to avoid UV-induced DNA damage. Excise bands and subject to gel extraction (for example, Qiagen# 28704).
20) Clone purified bands into an appropriate cloning plasmid using standard T4 DNA Ligase. One possibility would be to use the BamHI site of pZero-2 (Invitrogen). The plasmid has a ccdB death gene that is interrupted by successful insertion, thus reducing the background of re-ligated plasmids with no insert.
A potential pitfall of plasmids such as pZero-2, where fragment insertion occurs within a transcriptional unit, is that some viral sequences are highly toxic when transcribed bacteria. It may be helpful to use a cloning plasmid in which the cloning site is protected by bacterial transcription terminators. We sometimes use the BamHI site of a home-grown plasmid called pAsylum (available from http://www.addgene.org/). For pAsylum, re-ligation can be effectively prevented by aggressive dephosphorylation. An enzyme called APex from www.Epibio.comseems to be a bit more effective than the traditional CIP. Both CIP and Apex work in NEB buffers 2, 3 and 4, so it can be directly spiked into the pAsylum digest at 0.5 units/µg of DNA. CIP has greater activity at 50ºC, though it gradually dies at this temperature. A compromise is to incubate with CIP at 37ºC for 30’, then shift to 50ºC for 15’. Then add fresh CIP and repeat the 37º/50ºC incubations.
Another interesting cloning alternative, which we haven’t yet tried ourselves, is offered by a company called www.Lucigen.com. They sell transcription-protected “CloneSmart” plasmids that come with pre-dephosphorylated blunt ends. Note that use of the CloneSmart plasmid would require digesting the RCA material with a restriction enzyme that leaves blunt ends.
20) The following sequencing primers work well in our hands:
Zero159R: CCCAGGCTTTACACTTTATGCTTC
Zero481L: TGGCGAAAGGGGGATGTGC
Asylum-R: CTGAAAGAGGAACTTGGTTAGG
Asylum-L: ATGTTTCAGGTTCAGGGGGAGG
21) Align results using Blast or Blast-X searches. Happy hunting!
Overview
Content Tools
ThemeBuilder