Extraction of Rotavirus from Human Feces by Treatment

Transcription

1 APPLIED AND ENVIRONMENTAL MICROBIOLOGY, Jan. 1981, p /81/ $02.00/0 Vol. 41, No. 1 Extraction of Rotavirus from Human Feces by Treatment with Lithium Dodecyl Sulfate M. C. CROXSON' * AND A. R. BELLAMY2 Virus Laboratory, Auckland Public Hospital,' and Department of Cell Biology, University of Auckland,2 Auckland 1, New Zealand A procedure has been developed for the isolation of rotavirus from human fecal specimens based on the resistance of the virus to treatment with cold 1% lithium dodecyl sulfate at neutral ph. A single detergent treatment of fecal material followed by low- and high-speed centrifugations yielded a virus suspension of sufficient purity for viral ribonucleic acid to be analyzed directly by electrophoresis on polyacrylamide gels. Human rotaviruses, members of the family electrophoretic analysis of large numbers offecal Reoviridae, are responbile for acute enteric infection principally in young children (1, 2). Ro- specimens of limited size. taviruses are constructed of a double-layered MATERIALS AND METHODS protein shell, probably having skewed icosahedral symmetry of T = 13 (17), and are excreted was propagated in mouse L-cell fibroblasts at 320C, Cells and virus. Reovirus type 3 (Dearing strain) in the feces in large quantities (7). The viral according to the method of Smith et al. (20). Virus genome is composed of 11 segments of doublestranded ribonucleic acid (RNA) (9, 16, 18), and phosphate-deficient medium to which had been added labeled with [32P]phosphate was prepared by using 10 different strains uci of carrier-free of virus (electropherotypes) [32P]orthophosphate (The Radiochemical Centre, Amersham, England) per ml. Virus may be distinguished on the basis on differences was purified by the method of Smith et al. (20), in the migration rates of these genome segments including dialysis and two successive bandings on a when the RNA is subjected to electrophoresis CsCl gradient to ensure that all radioactivity present on polyacrylamide gels (6, 9, 15). in the virus preparation was virus associated. InI-labeled virus was purified in a similar manner after The ability to distinguish different strains of virus on the basis of the electrophoretic properties of the genome RNA is of particular utility Hunter reagent [N-succinimidyl 3-(4-hydroxy, 5- reaction of the purified virus with the Bolton and since the number of serotypes of human and ('25I)iodophenyl propionate] (4) obtained from The animal rotaviruses is yet to be Radiochemical Centre. firmly established Detergents and solvents. Freon trichlorotrifluoroethane solvent was a product of E.I. du Pont de (23, 24). Furtheremore, the sensitivity of the gel electrophoretic typing method and its ability to Nemours & Co., Wilmington, Del. Nonidet P-40 and detect small differences between otherwise apparently identical virus isolates enable molecu- Rohm and Haas, respectively. Sodium deoxycholate Triton X-100 were products of the Shell Oil Co. and lar epidemiological studies to be carried out on was a product of British Drug Houses, Poole, England. rotavirus infections (15). Lithium dodecyl sulfate (LDS) was prepared from the A major drawback to the application of gel sodium form (Sigma Chemical Co., St. Louis, Mo.) by electrophoresis for typing rotavirus isolates is passing a 10% solution through a column of Li'-form Dowex-50 the time involved in purifying virus from feces cation-exchange resin (13). Sodium N-lauryl sarcosine was a product of Sigma. and extracting the viral RNA. Typically, this Fecal samples. Samples of "normal" and rotavirus-positive feces were obtained from the Infectious has involved extraction of fecal material with fluorocarbon, cycles of high- and low-speed centrifugation to concentrate partially purified vi- Electron microscopy. Samples of virus or par- Diseases Ward, Auckland Public Hopsital. rus, sucrose zone sedimentation, and phenol extraction and ethanol precipitation to purify the cellulose-carbon-coated copper grids, blotted dry, tially purified fecal suspensions were applied to nitro- RNA (16). However, an ideal procedure whereby stained for 3 min with 0.10-saturated ammonium molybdate, and then examined in a Philips 300 transmis- large numbers of samples could be examined sion electron would eliminate solvent extraction and keep centrifugation to a minimum. We therefore report dure for the recovery of rotavirus RNA. The microscope. Fluorocarbon and phenol extraction proce- the results of a study aimed at simplifying the procedure for recovery of rotavirus RNA essentially extraction procedure for rotavirus and rotavirus followed those of Rodger et al. (16) and Croxon and RNA to develop a protocol suitable for the rapid Bellamy (6). A 20% fecal suspension was prepared in 255

2 256 CROXSON AND BELLAMY 0.1 M tris (hydroxymethyl) aminomethane - acetate buffer, ph 8, sometimes with sonication. The suspension was then extracted with 20% (vol/vol) fluorocarbon and clarified at 6,000 rpm for 30 min (MSE Mistral 6L). The resulting virus was concentrated from the supernatant by centrifugation at 28,000 rpm for 60 min in the type 30 rotor of a Beckman L2 ultracentrifuge. The pelleted material was resuspended in 1 ml of buffer, layered over a 3-ml 20% sucrose cushion, and further centrifuged at 45,000 rpm in an SW50.1 rotor for 1 h. This step concentrated relatively pure virus, whereas contaminating fecal material was largely retained in the sucrose cushion. Pellets of partially purified virus were resuspended in lx SSC (0.15 M NaCl, M sodium citrate, ph 7), and the suspension was brought to 0.5% sodium dodecyl sulfate (SDS)-0.1 M sodium acetate buffer (ph 5) and extracted three times with water-saturated phenol. Viral RNA was then precipitated from the final deproteinized aqueous supernatant by addition of 2 volumes of ethanol and incubation overnight at -20 C. The precipitated RNA (usually 5 to 50,ug) was recovered by ultracentrifugation (SW50.1 rotor, 45,000 rpm, 30 min). Acrylamide gel electrophoresis of viral RNA. Viral RNA was resolved by electrophoresis on 10% acrylamide slab gels, using the discontinuous buffer system described by Laemmli (10). A 3% stacking gel was used and the RNA was applied to the wells at a loading of 1 to 5,tg of RNA per channel in 0.01 M tris(hydroxymethyl)aminomethane (ph 7) M ethylenediaminetetraaacetate-0.1% SDS-20% sucrose. Electrophoresis was for 1 h at 50 V and 20 ma followed by 4 h at 150 V and 40 ma in an apparatus similar to that described by Studier (22). The gels were then stained with 5,ug of ethidium bromide per ml in water and photographed in ultraviolet light on a Blak-Ray ultraviolet transilluminator screen (Ultraviolet Products, San Gabriel, Calif.). Yields of rotavirus RNA derived by different extraction protocols were compared by direct visual inspection of the ethidium-stained bands. RESULTS Preliminary screening of detergents. Preliminary evaluation of the effect of sodium deoxycholate, Triton X-100, Nonidet P-40, Sarkosyl, and SDS on the physical properties of human fecal specimens was carried out by adding the compounds to samples of a "virus-negative" fecal suspension (1% final detergent concentration) and incubating at 0 C for 15 min. The suspensions were then clarified (4,500 x g, 30 min), and the supernatants were subjected to ultracentrifugation (SW50.1 rotor, 40,000 rpm, 60 min, 40C). The relative sizes of the resulting pellets were then taken as a measure of the efficacy with which the detergent treatent had been able to render particulate material nonsedimentable. Of the detergents examined, only SDS yielded a minimal pellet size, but precipitation of the detergent in the cold increased the bulk of the pellet considerably. Detergent precipitation was reduced substantially when LDS was substituted for SDS, a result anticipated on the basis of the well-known cold solubility of the lithium salt of this detergent (13). Stability of virus in LDS. Previous work (A. R. Bellamy, unpublished data) revealed that purified reovirus type 3 might be stable to treatment with SDS provided the ph was maintained near neutrality and the temperature was kept below 37 C. This observation suggested that the Reoviridae in general, including rotaviruses, might be stable to detergent treatment. To investigate more definitively the possible stability of the Reoviridae to treatment with LDS, experiments were carried out with radioactive reovirus as a model system. Purified "2'I-labeled reovirus, buffered at either ph 7.5 with 0.1 M tris(hydroxymethyl)aminomethane-chloride or ph 5 with 0.1 M lithium acetate, was subjected to treatment with cold 1% LDS at both 37 and 0 C (Table 1). Table 1 demonstrates that the virus was essentially stable to treatment at 00C regardless of ph but was unstable at 37 C at either ph 5 or 7.5. The anomalously low recovery of virus in the absence of detergent was found to be due to the marked tendency of virus to adsorb to the polypropylene centrifuge tube in the absence of detergent. Recovery of virus from feces after detergent treatment. A further model experiment (Table 2) in which radioactive virus was added to a buffered fecal suspension containing either 1% LDS or no detergent enabled the recovery of virus to be determined after either detergent treatment or the more traditional fluorocarbon extraction procedure (see above). TABLE 1. Stability of '251-labeled reovirus to treatment with LDS % Virus recoveredc Smla Temp Buffer/ Deter- P (OC) ph gentb Adsorbed Pellet to tube < < x SSC a APPL. ENVIRON. MICROBIOL x SSC All samples contained approximately 25,tg of '25Ilabeled reovirus, equivalent to 40,000 cpm. b LDS to 1% was added to samples 1, 2, 4, and 5. 'Proportion of total added radioactivity recovered after incubation at either 0 or 37 C for 30 min and then centrifugation for 1 h at 45,000 rpm in the SW50.1 rotor of a Beckman ultracentrifuge.

3 VOL. 41, 1981 It is clear from Table 2 that detergent treatment enhanced the recovery of reovirus (82%), whereas during fluorocarbon extraction (41% recovery), over half of the added virus was lost. Experiments not presented here revealed that these losses occurred as a result of adsorption to fecal solids, a process not blocked by the addition of bovine serum albumin to the fecal suspension but substantially reduced by the addition of detergent. As will become apparent below, losses could also be reduced by vigorous sonication, suggesting that some virus may be entrapped rather than adsorbed. Recovery of rotavirus from feces by detergent treatment. To investigate the efficacy of detergent treatment as a method for recovering human rotavirus from fecal suspensions, a series of known virus-positive specimens, selected on the basis of electron microscopic examination, were buffered to ph 8 with 0.1 M tris(hydroxymethyl)aminomethane-acetate and extracted with either cold 1% detergent or fluorocarbon. Virus was then concentrated by centrifugation from both samples, and the RNA was extracted (see above) and applied to acrylamide gels. Ethidium bromide staining of the gels after electrophoresis enabled the yield of rotavirus RNA to be estimated, which in turn provided a measure of the yield of rotavirus achieved by the two procedures (Fig. 1F and L). Figure 1 demonstrates that detergent extraction resulted in approximately the same or a better yield of virus (RNA) than that which was achieved by the fluorocarbon method. Sonication of samples (lanes 4, 5, and 6) improved the yield obtained by the fluorocarbon method, suggesting that some of the losses revealed in Table 2 may have resulted as much from physical entrapment as from adsorption of the virus to solid material. TABLE 2. Recovery of radioactive virus from feces after detergent or fluorocarbon extraction Samplea ph Deter- Fluoro- % Recov- Sample' ph tbentb carbonc ery (pel- Feces/virus Feces/virus Buffer/virus aradioactive 32P-labeled reovirus (approximately 1 mg, equivalent to 106 cpm) b LDS to 1%. e Trichlorotrifluoroethane to 20% (vol/vol) and centrifugation to separate phases. d Proportion of total added radioactivity recovered after incubation at 0 C for 30 min and clarification at 6,000 rpm, followed by centrifugation for 1 h at 40,000 rpm in the SW50.1 rotor of a Beckman ultracentrifuge. ROTAVIRUS PURIFICATION FROM FECES 257 Effect of detergent treatment on virus architecture. Since the rotaviruses are composed of a double-layered protein shell, the possibility exists that a purification procedure based on cold detergent treatment might disrupt the outer shell of the virus while leaving the inner shell intact. Radioactive human rotavirus is not readily available, and infectivity measurements are not possible. Samples of detergent-purified material therefore were examined in the electron microscope to determine the effect of detergent treatment on virus architecture. The integrity of detergent-purified virus was compared with control material subjected only to fluorocarbon extraction (Fig. 2). It is evident from Figure 2 that the virus survived cold detergent treatment satisfactorily, although removal of minor polypeptides essential for infectivity but not required for virion stability cannot be excluded at this stage. Simplified detergent extraction procedure for electropherotyping of rotavirus isolates. Since cold detergent treatment yields a virus pellet minimally contaminated with other material, it seemed probable that virion lysis and direct application of partially purified fecal material to an acrylamide gel might permit viral RNA to be examined without the normal requirement for sucrose gradients, phenol extraction, and ethanol precipitation. A rotavirus pellet derived from cold detergenttreated feces was resuspended in 0.01 M lithiumacetate buffer, ph 5. The sample was then adjusted to 1% LDS and heated at 80 C for 5 min to lyse the virus. A portion of the extract was then mixed with an equal volume of sample application buffer (see above) and applied directly to the gel. A second sample was clarified at full speed (30 lb/in2) in a Beckman Airfuge for 5 min before application. Both samples were then subjected to electrophoresis in the normal manner (Fig. 3). Figure 3 reveals that rotavirus RNA may be typed satisfactorily by direct application of the lysed fecal pellet derived by detergent treatment. Although substantial amounts of material could be removed after heating, this material did not appear to affect the quality of the gel electropherogram obtained by this simplified procedure, provided the total loading of material was not too high. Extraction with cold detergent clearly releases virus with high efficiency and yields a sample not grossly contaminated with other material. The method enables viral RNA derived from samples as small as 0.1 ml of fecal suspension to be analyzed by gel electrophoresis. If the Airfuge is used for centrifugation steps, the complete

4 258 CROXSON AND BELLAMY APPL. ENVIRON. MICROBIOL F L F L F L F L F L F L FIG. 1. Yield ofrotavirus RNA achieved by cold detergent extraction ofsix different fecal samples. Samples 4 to 6 were sonicated (MSE ultrasonicator) for 2 min at 0 C before treatment with LDS or fluorocarbon. L, LDS extracted; F, fluorocarbon extracted. FIG. 2. Effect ofdetergent treatment on rotavirus architecture. (A) Rotavirus recovered from fecal specimen, using detergent treatment method; (B) rotavirus recovered from fecal specimen, using the freon extraction procedure. Ammonium molybdate stain. x150,000.

5 VOL. 41, 1981 a b c d e f g FIG. 3. Electrophoretic analysis of rotavirus RNA derived from a detergent-treated fecal pellet. Lanes: (a) Reovirus RNA; (b, c, d) increased loadings of lysed virus corresponding to 0.1, 0.2, and 0.5 g of feces, respectively; (e, f, g) increasing loadings oflysed virus clarified at 100,000 x g for 5 min before application. electropherotyping may be achieved in 5 to 6 h, rendering the procedure suitable for routine application to large numbers of fecal specimens. DISCUSSION Mandel (11) was the first to demonstrate that poliovirus survived treatment with SDS provided the ph was maintained near neutrality. This observation was applied by Philips et al. (14) to the purification of poliovirus from infected HeLa cells, and this procedure and others based on the use of related detergents such as sodium N-lauryl sarcosine (12) are currently in use for the routine purification of poliovirus for biochemical studies. The work described here demonstrates that both reovirus and human rotavirus survive treatment with cold 1% LDS and that a simplified purification procedure based on detergent treatment can be applied to fecal material. The intractable nature of fecal material and the high ROTAVIRUS PURIFICATION FROM FECES 259 losses of virus which result from either entrapment or adsorption make the use of detergent an attractive proposition. Presumably cold detergent treatment does not lyse bacteria, the principal component of human feces (21), and leaves the virus particles intact. The relative cold solubility of the lithium salt of lauryl sulfate enables virus pellets to be prepared which are not heavily contaminated with detergent: subsequent heat lysis of the partially purified virus yields an extract of sufficient purity for direct application to acrylamide gels. The method developed here therefore should prove useful for the analysis of clinical specimens when time and centrifuge space are limiting factors. The detergent resistance of doubled-shelled icosahedral viruses such as reovirus and rotavirus is a somewhat surprising observation in view of the effects of other treatments such as heat (5), urea, or enzymatic digestion (19), which are able to remove the outer shell of reovirus. The outer shell of both reovirus and rotavirus is entirely proteinaceous, and being distinct from the inner nucleoprotein core of the virus (8), presumably is stabilized solely by protein-protein interactions. Boatman and Kaper -(3) have proposed that the sensitivity of many viruses to SDS results from disruption of the electrostatic protein-rna interactions which are responsible for stabilizing the virion structure. This conclusion was reached largely because the high dissociation efficiency of SDS seemed confined to viruses which are thought to derive most of their intrinsic stability from protein-rna interactions rather than protein-protein interactions. The double-layered nature of the Reoviridae coat protein may thus confer relative stability on the virus and perhaps be the cause of the notable resistance of both reovirus and rotavirus to detergent treatment. ACKNOWLEDGMENTS We acknowledge the able technical assistance of G. Neale. This work was supported by grants from the Medical Research Council of New Zealand and the National Children's Health Research Foundation of New Zealand. LITERATURE CITED 1. Bishop, R. F., G. P. Davidson, I. H. Holmes, and B. J. Ruck Virus particles in epithelial cells of duodenal mucosa from children with acute non-bacterial gastroenteritis. Lancet ii: Bishop, R. F., G. P. Davidson, L. H. Holmes, and B. J. Ruck Detection of a new virus by electron microscopy of fecal extracts from children with acute gastroenteritis. Lancet i: Boatman, S., and J. M. Kaper Molecular organisation and stabilising forces of simple RNA viruses. IV. Selective interference with protein-rna interactions by use of sodium dodecyl sulphate. Virology 70:1-16.

6 260 CROXSON AND BELLAMY APPL. ENVIRON. MICROBIOL. 4. Bolton, A. E., and W. M. Hunter The labeling of proteins to high specific radioactivities by conjugation to a "2'I-containing acylating agent. Biochem. J. 133: Borsa, J., and A. F. Graham Reovirus:RNA polymerase activity in purified virions. Biochem. Biophys. Res. Commun. 33: Croxson, M. C., and A. R. Bellamy Two strains of human rotavirus in Auckland. N.Z. Med. J 90: Holmes, I. H Viral gastroenteritis. Prog. Med. Virol. 25: Joklik, W. K Reproduction of reoviridae, p In H. Fraenkel Conrat and R. R. Wagner (ed.), Comprehensive virology, vol. 2. Plenum Press, New York. 9. Kalica, A. R., M. M. Sereno, R. G. Wyatt, C. A. Mebus, R. M. Chanock, and A. Z. Kapikian Comparison of human and animal rotavirus strains by gel electrophoresis of viral RNA. Virology 87: Laemmli, U. K Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature (London) 227: Mandel, B The extraction of ribonucleic acid from poliovirus by treatment with sodium dodecyl sulfate. Virology 22: Mapoles, J. E., J. W. Anderegg, and R. R. Rueckert Properties of poliovirus propagated in medium containing cesium chloride: implications for picornaviral structure. Virology 90: Noll, H., and E. Stutz The use of sodium and lithium dodecyl sulphate in nucleic acid isolation. Methods Enzymol. 12B: Philips, B. A., D. F. Summers, and J. V. Maizel, Jr In vitro assembly of poliovirus-related particles. Virology 35: Rodger, S. M., and L. H. Holmes Comparison of the genomes of simian, bovine, and human rotaviruses by gel electrophoresis and detection of genomic variation among bovine isolates. J. Virol. 30: Rodger, S. M., R. D. Schnagl, and I. H. Holmes Biochemical and biophysical characteristics of diarrhea viruses of human and calf origin. J. Virol. 16: Roseto, A., J. Escaig, E. Delain, J. Cohen, and R. Scherrer Structure of rotaviruses as studied by the freeze-drying techniques. Virology 98: Schnagl, R. D., and I. H. Holmes Characteristics of the genome of human infantile enteritis virus (rotavirus). J. Virol. 19: Shatkin, A. J., and J. D. Sipe RNA polymerase activity in purified reovirus. Proc. Natl. Acad. Sci. U.S.A. 61: Smith, R. E., H. J. Zweerink, and W. K. Joklik Polypeptide components of virions, top component and cores of reovirus type 3. Virology 39: Stephen, A. M., and J. H. Cummings Microbial contribution to human fecal mass. J. Med. Microbiol. 13: Studier, F. W Analysis of bacteriophage T7 early RNA and protein on slab gels. J. Mol. Biol. 79: Thouless, M. E., A. F. Bryden, and T. H. Flewett Serotypes of human rotavirus. Lancet i: Zissis, G., and J. P. Lambert Different serotypes of human rotavirus. Lancet i: Downloaded from on January 11, 2019 by guest