Purification of Non-Enveloped Viruses from Skin or Tissue Samples
Chris Buck and Diana Pastrana
Laboratory of Cellular Oncology, NCI, NIH
Bethesda, Maryland, USA
Citations
•Schowalter, et al. (2010) “Merkel Cell Polyomavirus and Two Previously Unknown Polyomaviruses Are Chronically Shed from Human Skin” Cell Host & Microbe7:509
•Pastrana et al. (2013) “A Divergent Variant of the Eleventh Human Polyomavirus Species, Saint Louis Polyomavirus” Genome Announcements 1:e00812
•Peretti et al. (2015) “Hamburger Polyomaviruses” Journal of General Virology 96:833
Introduction
Healthy human skin is home to a rich diversity of microbes. The bacterial constituents of the skin microbiome are increasingly well cataloged based on their ribosomal RNA and other gene sequences. In contrast, new human papillomavirus (HPV) and human polyomaviruses (HPyV) species are still being discovered at a rapid rate. Many of these discoveries have used a technique known as random-primed rolling circle amplification (RCA), a method that indiscriminately amplifies circular DNA molecules, such as the circular dsDNA genomes of HPVs and HPyVs. One trouble with using RCA to detect HPV and HPyV genomes is that skin-dwelling bacteria harbor plasmids and circular DNA bacteriophages that may dramatically outnumber the viral genomes our lab hunts. Our initial solution to this problem was to digest the sample with restriction enzymes that are unlikely to cut polyomavirus genomes then treat the sample with exonuclease V (Plasmid Safe). This approach has been remarkably successful for skin surface samples.
More recently, we have been working to adapt the methods to allow analysis of more complex tissue samples, including excised skin warts, blood, or minced muscle. In contrast to skin surface samples, in which host cell genomic DNA has been heavily fragmented during the process of desquamation, tissue samples contain very long fragments of host cell DNA. The long fragments of host cell DNA are an excellent substrate for phi29 polymerase. Pilot experiments with human whole blood suggested that the restriction enzyme/Plasmid Safe approach is overwhelmed by the large excess of human DNA in typical samples.
Our current approach to complex tissue specimens is to digest non-encapsidated DNA with a generic endonuclease (Benzonase), then use density gradient ultracentrifugation through an Optiprep gradient to separate HPV and HPyV virions from unwanted bacterial or human DNA fragments. Virion purification also seems to significantly improve detection sensitivity for skin swab samples.
Extracting virions from skin swabs
•Swabbing human subjects requires Institutional Review Board approval. Informed consent must be obtained from potential study volunteers.
•We have traditionally used wooden cotton-tipped swabs (Greiner Bio-One #421180). More recently, we have shifted to polyester swabs (Puritan #25-3316-U). It is possible to pack a larger number of the polyester swabs into the extraction column.
1) Perform the swabbing as late in the day as possible. In theory, this gives virions a chance to re-accumulate after morning skin washing. Collect a total of five swabs from each skin or mucosal surface of interest.
2) Optional: pre-wet the swab by dipping it in normal saline. This may be particularly helpful for sampling dry and crusty skin lesions.
3) Hold the swab at a point close to the cotton tip. Holding the swab at the end away from the cotton often results in the wooden shaft of the swab breaking in the middle. If the volunteer is self-sampling and some viruses from the volunteer's fingertips happen to get on the cotton, the more the merrier.
4) Vigorously rub the swab back and forth on the skin. Apply as much pressure as comfort allows. Ideally, volunteers should apply enough pressure for the skin to be slightly reddened (erythematous) afterward. The swab should be gradually rotated as it is moved across different areas of skin. This will help saturate all surfaces of the swab. The swab should end up at least a little discolored and greasy-looking on all sides. As far as we can tell, the presence of makeup doesn't matter.
5) Put the swab back into its collection tube, store at room temp (or 4ºC or -80ºC, if available). Do not attempt to use any sort of preservative fluid for storage. Dessicated HPV and HPyV virions seem to be very stable and dry swabs can be stored for up to a week. We don’t yet have any experience with longer-term storage of swabs that have been pre-wetted with normal saline.
6) Add 150µl of nuclease cocktail (DPBS with 10mM MgCl2, 0.5% Triton and 0.1% Benzonase) to each swab. Liquid should be completely absorbed into the swab. Incubate 15’ room temperature.
7) Break off the cotton ends of the swabs and pack them together into a Pierce #89896centrifuge column. Add 0.5 ml of DPBS with 0.8M NaCl and 0.5% Triton to the top of the column. Centrifuge 3,000 x g 15’. Remove eluted material to a siliconized 2 ml tube. The first eluate should be cloudy due to emulsified lipids.
8) Rearrange and mash up the swab pile using a 1000 µl pipette tip. Extract the swabs two more times with 0.5 ml of DPBS/NaCl/Triton. Mash up the swabs between washes to prevent the formation of channels around the cotton matrix. The final wash should be pretty much transparent.
9) Load extracted fluid onto a 27-33-39% Optiprep gradient with ~1 ml steps. See standard protocolfor instructions on ultracentrifugation and fractionation of the gradient.
Extracting virions from warts
•Excised warts should not be placed in liquid. Theoretically, virions might dissolve out of the wart and into the storage liquid. If the wart is provided in liquid, use the liquid for step 4 below. It seems very unlikely that this protocol would work for fixed specimens.
•Dry warts can be shipped in microcentrifuge tubes at ambient temperature. HPV virions are very stable and refrigeration or freezing seem to be unnecessary.
1) Place the wart material into a plastic dish, such as a 5 cm Petri dish or a single well of a 6-well cell culture plate.
2) Use a pipette tip to press the wart onto the plastic. Use a scalpel to mince the wart.
3) Carefully transfer as many wart fragments as possible into a screw-cap 1.5 ml microcentrifuge tube. Ideally, the centrifuge tube should be siliconized. If you are uncomfortable using the scalpel for the transfer, then use an autoclaved steel spatula.
4) Using a 1000 µl pipettor, rinse the dish with 250 µl of Dulbecco’s PBS (with calcium and magnesium, without glucose) supplemented with 1% Triton X-100. Recover as many wart bits as possible. If necessary, repeat the rinse once or twice (i.e., total collection volume of 500 µl or 750 µl).
5) Add 2 µl of Benzonase (Sigma) endonuclease. Incubate at 37ºC 20 minutes, inverting the tube occasionally.
6) Add a few milligrams of collagenase H (Sigma C8051). We’ve never attempted to weigh the collagenase - just sprinkled in a little bit of powder that, by eyeball, looks like a few mg.
Note: The recommended collagenase H preparation has some tryptic activity that we worry might chew up the Benzonase. We therefore give the Benzonase some time to begin digesting non-encapsidated DNA prior to adding collagenase.
7) Briefly vortex the sample a few times. Put at 4ºC overnight.
8) Adjust the NaCl concentration of the suspension to 0.85 M by adding 0.17 volume of 5 M NaCl. Put tube on inverter for 15 minutes.
9) Centrifuge the tube 5 minutes at 5,000 x g. Transfer the supernatant to a fresh siliconized tube.
10) Resuspend the pelleted material in 0.5 ml of DPBS with 0.8 M NaCl. Sonicate the suspension using a cup sonicator until the wart pieces are mostly broken up and/or friable-looking.
Note: We suspect that a great majority of the virions come out in the initial supernatant in step 9. However, some additional virions may be liberated by the sonication. Since sonication can fragment DNA, we refrain from sonicating until after the first supernatant is removed. That way it’s likely that a majority of the virions won’t be exposed to potentially damaging sonication, but any entrapped virions that need sonication will have a chance to be liberated and recovered.
11) Clarify the sonicated lysate, combine second supe with first supe.
12) Re-clarify the pooled supernatant and run over an Optiprep gradient according to standard protocol.
Larger samples
We’ve recently become interested in the viral content of human foodstuffs. Protocol modifications needed for processing larger amounts of high-complexity samples from the local supermarket should be published soon. Meantime, please don’t hesitate tocontact usfor advice on processing larger samples.
Collection of gradient fractions
For wart specimens, it’s possible to detect fractions containing purified virions using Picogreen DNA stain. Picogreen-positive fractions can be pooled and RCA’d together.
Skin swabs often do not contain enough viral DNA to detect by Picogreen. In this case, the safest approach is to subject the core of the gradient (fractions 3-8) to RCA amplification reactions. In some cases, we’ve fractionated the entire Optiprep gradient and RCA’d all fractions. So far, there has not seemed to be much of anything interesting outside the core (fractions 3-8) of the gradient. We extract and RCA each individual fraction, with the theory that we don’t want things that are abundant in one fraction overwhelm less abundant things that have been usefully separated into different fractions. For MiSeq analysis, we typically pool together the separate RCA reactions for any give gradient (or, in some instances for any given patient).
Since the Optiprep gradients remove nearly all unwanted non-viral DNA, it is not necessary to perform Plasmid Safe (exonuclease V) digestion prior to RCA. In our experience, Optiprep material can be subjected directly to proteinase K digestion and ethanol precipitation (see RCA protocol). In other words, there seems to be no need for purifying Optiprep fractions over Qiagen PCR Purification purple spin columns. RCA reagents can be added directly to the DNA that has been ethanol precipitated out of the Optiprep fractions.
Overview
Content Tools
ThemeBuilder