Minipurification for plant virus diagnosis

Objective: Rapid, small scale plant virus purification permitting electrophoretic detection of capsid proteins.

- 22ml - 0.2M Na citrate, pH 6.5 (see table)
- 0.15ml - 0.5M Sodium diethyldithiocarbamate (polyphenol oxidase inhibitor)
- 0.05ml - 0.5M Iodoacetamide (IAm, thiol protease inhibitor), usually not essential
- 1g - PVP 10K for tannin containing tissues only - for tissues that foam add a drop of antifoam (antifoam A, Sigma)


1- Grind 1g plant tissue in 22ml cold buffer with additives in small Waring blendor cup.

2- Express through wet muslin into polycarbonate centrifuge tube, centrifuge 12min at 35K in Beckman 50.2Ti rotor. Perform all centrifugations at room temperature.

3- Pour supernatant into polycarbonate centrifuge tube (if supernatant has particulates (probably wax), pour through Miracloth or filter paper), add 4 drops 10% Triton X-100, mix.

4- Centrifuge 45min at 35K in 50.2Ti rotor; discard supernatant as soon as centrifuge stops; wash tube walls thoroughly with DW; drain tubes for about 15min; tap tubes dry.

5- Suspend pellets (50mM NaPO4, pH 7.0), typically 50-500Ál depending on pellet size

6- Analyze by gel electrophoresis (SDS gels with silver staining), EM, or serology. For silver staining, avoid thiols in sample (e.g. reduce and alkylate samples)


Minipurification is a partial purification. It helps identify and classify unknown viruses. It exploits size and density uniqueness of viruses, resolving power of gel electrophoresis and detection sensitivity of silver stains. One assigns unknowns to groups based on capsid protein size and yield. One can examine purified virus by electron microscopy and serology.

Include a healthy plant control. The same host infected with another virus is a better control. Common host contaminants include rubisco (55K, 12K subunits), phytoferritin (multiple subunits around 25K). Species contain characteristic contaminants; indeed, host protein "spectra" can be species specific. Most contaminants are not proteins, but carbohydrates (gums) which smear near the top of protein gels. Brief ultracentrifugation clarifies samples much better than low speed centrifugation.

Supernatants (step 3) of tannin rich tissues turn milky with Triton and give large, glassy pellets (step 4) and ultimately smeary patterns, or no bands at all. Soluble PVP removes tannins.

We run Laemmli gels and stain with a modified Morrissey silver stain.

This procedure is robust (tolerates variations) and works for many viruses. You can modify equipment and reagents to suit needs. Rhabdoviruses and tospoviruses yield nucleoproteins. Potential problem viruses are: caulimoviruses (occluded), luteoviruses and geminiviruses (low concentration, difficult to extract).

Optimize purifications by varying this procedure. Host species, tissue age, plant growth temperature, buffer composition, heat treatments, freezing and thawing, etc. can be easily compared. Citrate is cheap and soluble. It yields more virus and fewer contaminants than other buffers. Where citrate doesn't work I recommend trying other buffers over a range of buffer concentrations (vary by a factor of 2 or 3) and pHs.

Adjuncts to minipurification

1- Most virus-plant combinations give cleaner minipurifications at 0.4M citrate than at 0.1M citrate. Note that citrate is a stabilizing anion in the Hofmeister series. Some viruses and some hosts are better at 0.1M citrate. For new virus-host combinations I often minipurify with 0.1M, 0.2M and 0.4M citrate. 0.6M citrate is incompatible with Triton X-100.

2- Reducing extraction pH to 4.5 decreases recovery of rubisco. Viruses which are stable at pH 4.5 often give substantially cleaner (at least in terms of proteins) preparations. To determine purification and stability properties of new viruses I routinely minipurify at pHs 6.5,5.5 and 4.5. (see table)

3- Proteases cleave the N-terminus of rubisco L. Proteolysis cleaves or degrades some host proteins. Many viruses are resistant to proteolysis. Some viruses (e.g. potyviruses, potexviruses) are characteristically "trimmed" by proteases while others (e.g. tobamoviruses, tombusviruses) are generally inert to proteolysis. Proteolysis can reveal nepovirus capsid proteins comigrating with rubisco L. Dissolve trypsin and chymotrypsin (separately) to 10mg/ml in 1mM HCl (which prevents self digestion). Add 50Ál of protease to the sample before grinding. Proteolysis does not reduce recovery of rubisco. The degree of digestion of rubisco L subunit (appearance of faster migrating bands) is a useful measure of proteolysis.

4- For larger viruses (e.g. rhabdoviruses), 10min @ 35K may pellet too much virus. Lowering centrifugation speed may give higher yields, but more Triton will be required to solubilize membranes. If the final high speed pellet is green then the clarification supernatant requires more Triton.

5- For records I routinely place gels in plastic sheet protectors and scan at high resolution (roughly 10Mb files) with a HP Scanjet 5200C flatbed scanner. I import the images into Paint Shop Pro for aligning, cropping and labeling. I print labeled images with an HP Deskjet 932C.