key: cord-0004919-nswbm9g3 authors: Collins, M. S.; Alexander, D. J.; Harkness, J. W. title: Heterogeneity of infectious bronchitis virus grown in eggs date: 1976 journal: Arch Virol DOI: 10.1007/bf01318001 sha: eecf968682082e196999faa0787c0492ac1d0c35 doc_id: 4919 cord_uid: nswbm9g3 Egg-grown infectious bronchitis virus, strain Beaudette, was concentrated and centrifuged on sucrose density gradients to separate the virus into five peaks with densities of 1.144, 1.160, 1.172, 1.191 and 1.218 g/cm(3). All peaks retained infectivity, complement fixation activity and were labelled with(3)H-uridine. Morphologically the densest peak consisted of very large virus particles and amorphous material, the other peaks consisted of mainly intact particles although small differences in size and pleomorphism were seen. Polyacrylamide gel electrophoresis of material from the density gradient peaks revealed four major polypeptides and at least 10 minor polypeptides. The proportions of the polypeptides were approximately similar for all peaks with the exception of the densest peak in which the major polypeptides were greatly reduced. The four major polypeptides had approximate molecular weights of 1. 52,000, 2. 45,000, 3. 34,000, 4. 32,000. The major polypeptides 1 and 4 were shown to be glycosylated as were two of the minor polypeptides. Avian infectious bronchitis virus (IBV) has been classified as a coronavirus (AL~EID~t et al., 1968) . Morphologically it exists as pleomorphic virus particles 80--120 nm in diameter which usually possess projections 20 nm in length and 9--11 nm in width at the outer end (MCINToSI-I et al., 1967 ; BEI~I~¥ and ALIVIEIDA, ]968). However, the presence or absence of the projections and their exact morphology has varied with different reports (Cu~' NINGZ4A~, 1970) . l~ecently HA~KNESS and BI~, AOEWELL (1974) have shown variations in the possession of a corona amongst 12 strains, and have related these to immunological variations. APOSTOLOV et al. (1970) examined IBV particles in ultra thin sections which revealed an envelope, similar in appearance to the "unit cell membrane", enclosing an internal thread-like structure 7--8 nm in diameter. Studies on the morphogenesis of IBV in chick embryo fibroblasts (NAZERIAN and CU~I~-G~AM, 1968), ehorioMlantoie membrane cells (BEeKER et aI., i967) or VEt0 cells (Cu~I~G~A~ et al., 1972) have produced similar results to those obtained with the Linder strain eoronavirus in human embryo lung cells (Os~mo et al., 1971) . This virus was shown to bud into the cisternae or vesicles of the endoplasmie retieulum but occasionally cytoplasmic inclusions of tubular forms were seen which suggest an alternative mode of virus reproduction. BE(JKER et al. (1967) report the complete IBV particle seen in morphogenesis studies as consisting of a double outer shell and a denser inner she]] surrounding a core of amorphous material. There have been few studies on the virion structure of IBV or eoronaviruses in general. HEIa~OLZ]~R et al. (1972) examined the structural polypeptides of the human eoronavirus OC43 and detected a minimum of six polypeptides with a molecular weight range of 15,000--191,000 of which 4 were glycopolypeptides. Estimations of the buoyant density of IBV have varied considerably: In isopyenic caesium cMoride density gradients the density of the Beaudette strain grown in eggs has been reported as 1.23 g/em 3 (TEv~.THIA and Cu~I~GI~, 1968), but in sucrose density gradients it was found to be 1. 19 (CtrN~JNGHAM, 1970) . ME~GELI~G and CORL~ (1972) demonstrated a peak of activity at either 1.198 or 1.205 for IBV on CsC1 density gradients but reported a distribution between 1.189--1.220 particularly if the virus had been concentrated by dialysis against polyethylene glycol. TANNOCK (1973) reported a wide distribution of virus in. sucrose density gradients with a peak of activity at 1.176. Better resolution was obtained on potassium tartrate density gradients with a peak at 1.16, but tartrate caused high losses of infectivity and disruption of virus particles. In the present study we have examined the morphological and structural properties of virus particles which differed in buoyant density on sucrose density gradients. The Beaudette strain of avian infeetiouos bronchitis virus (IBV) was grown in 9 or 10-day-old fertile chicken eggs by inoculating each egg with 100 ELDs0 (50 per cent egg lethM doses of virus) by the Mlantoic route. Infected amnio-allantoie fluids were harvested 40 hours after infection. Identical results were obtained with virus that had been cloned by three passages to limiting dilutions in eggs as well as uneloned virus. Virus from harvested fluids was concentrated by centrifugation at 30,000×g for 40 minutes and resuspension in 0.1 ~I Tris-ttCI buffer pH 7.2. Initial density gradient analysis was by centrifugation at 80,000 × g for 2 hours through either a 20 ml 20--55 per cent (w/w) continuous sucrose density gradient or a discontinuous gradient consisting of sucrose solutions of densities: 1.134 (4.5 ml), 1.150 (4.5 ml), 1.164 (4.5 ml), 1.180 (4.5 ml) and 1.26 (2 ml). Four bands from the interfaces of the discontinuous gradient were collected separately and again centrifuged at 80,000 × g for 2 hours through the appropriate discontinuous gradient prepared from sucrose solutions with the following densities: a) 1.134 (9 ml), 1.150 (9 ml); b) 1.150 (9ml), 1.164 (9rnl); c) 1.164 (9ml), 1.180 (9ml) and d) 1.180 (8ml), 1.26 (10ml). Final purification and separation was o~ a continous sucrose density gradient at 80,000× g for 16 hours. The material from each peak was reeoncentrated and resuspended in 0.1 ~ Tris-HC1 buffer pH 7.2 for protein estimations, infectivity and complement fixation testing or in 0.01 5I phosphate buffer pH 7.2 for polyacrylamide gel eleetrophoresis. Infectivity was estimated by inoculating 0.1 ml volumes of a ten-fold dilution series of virus suspension into the allantoic cavity of 9 or 10-day-old fertile chicken eggs. Seven eggs were used for each dilution. The eggs were candled daily and the end point expressed as ELDs0 calculated by the Spearman-Karber method. Virus labelled with aH-uridine was obtained by inoculating eggs with 50 BCi of SH-uridine immediately after infection. 3It (5)-uridine (20--30 Ci/mM) was obtained from the gadiochemical Centre, Amersham, U.K. The complement fixation test was basicalIy that of BRACEWEIm (1973) but was modified so that virus and not antiserum was titrated. A standard serum dilution containing two complement fixation units (CFU) was added to each two-fold dilution of antigen. Chicken antisera specific for IBV strain Beaudette was supplied by C.D. Bracewell, Central Veterinary Laboratory, U.K. Polypeptides were analysed by polyacrylamide gel electrophoresis in the presence of sodium dodecylsulphate (SDS-PAGE) as described (ALEXANDER,, 1974). Protein was estimated by the method of LowaY et al. (1951) using bovine serum albumin as a standard. Centrifugation of concentrated virus on linear sucrose density gradients produced a broad spread of material detected by measurement of extinction at 254 nm, ranging in density from 1.t2--t.22 g/cm 8 and consisting of up to three peaks ( Fig. 1) . To examine the possibility of structural or morphological differences of virus particles with different buoyant densities, a discontinuous sucrose density gradient was used to separate the virus into four distinct peaks (Fig. 2) . Fractions containing the four peaks were taken and recentrifuged on discontinuous gradients to give clean preparations of each peak. The densest peak could be separated into two distinct peaks at this stage (Fig. 3) . Although this was a consistent finding, the amount of material involved in the denser of the two peaks (V) was related to the length of time the virus was stored at temperatures other than --60 ° C. Material from each of the five separated peaks was centrifuged on a continuous sucrose density gradient for 16 hours at 80,000 X g. The results obtained are shown in Figure 4 (a--e). The densities recorded for each pee~k were: I t.144, II 1.160, III 1.172, IV 1.191, V 1.218. Peak V was always more heterogeneous in density than the other peaks. i // ? , 5 I II ttl IV rV I I I I 5 10 15 20 FRACTION NUMBER Fig. 2 . Diseontinuous sucrose density gradient anMysis Concentrated virus was applied to a discontinuous suerose gradient as described in methods section and centrifuged for 2 hours at 80,000 × g ./" . Density gradient separation of peaks IV a~ld V Peak IV/V from a discontinuous density gradient similar to that shown in Figure 2 was centrifuged for 2 hours at 80,000 × g on a discontinuous sucrose gradient Any manipulation of IBV preparations resulted in a large drop in infectivity titre. Although it could be shown that infectivity, complement fixing (CF) activity and 31t-uridine were associated with all five peaks, titres of infectivity and CF activity were very low. Much higher levels were obtMned if the second discontinuous density gradient step in the separation of the peaks was omitted although this meant that peaks IV and V were not separated. Typical results are shown in Table t. Some differences in the levels of the activities associated with each peak were detected, in particular I and IV/V showed relatively low levels of CF activity per mg protein but high levels of associated SH-uridine. Comparison of CF aetivity/mg protein in allantoic fluid with peaks II and III from the sucrose density gradients suggest that the virus has been purified more than 400-fold relative to protein. Negative-stain electron microscopy reveMed several differences in the morphology of the material in each peak (Fig. 5) . Peak I consisted of pleomorphic virus particles which were frequently seen as elongated filamentous forms. Filamentous virus particles also formed a small proportion of the virus populations seen in eleetronmierographs of peak III but were rarely seen in peak II, which consisted mainly of regular spherical virus particles. Peak IV consisted of particles similar in morphology to those of peak III but with a much larger size range, occasionally very large particles were seen in this peak. Peak V consisted mainly of amorphous material although clumps of large particles were also present. The amorphous material was usually in large aggregates but occasionally individual pieces were seen which, although lacking marked structural definition, were distinctive in appearance (Fig. 5e inset) . The mean diameters of the virus particles in each peak were: I: l13:k23nm, I I : 97:J:19nm, I I I : 1 1 2~2 4 n m , IV: 1 2 2 ± 2 2 n m , V: 1 8 2 ± 7 0 nm. In each peak particles that possessed projections could be seen (Fig. 5a--e) . These projections were unlike those usually associated with coronaviruses and rarely formed a complete corona. However, the presence of projections is in contrast with the work of HA~KN~SS and BRACEW~LL (1974) who reported the absence of projections on virions of this strain of IBV. The possibility existed that the originM seed virus may have been a heterogeneous population of genetically different viruses which were merely separated from each other by sucrose gradient eentrifugation. This possibility was eliminated The virus pools produced were cormentrated and subjected to sucrose gradient analysis. The virus grown from each of the peaks was also heterogeneous and could be separated into four peaks on discontinuous density gradients. Material from each of the five density gradient peaks was disrupted with SDS and dithiothreitol (DTT) and anMysed by SDS-PAGE. Staining with amidoblack revealed a similar pattern for each of the five peaks (~'igs. 6 and 7). In gels of materiM from peaks I --I V four major po]ypeptides were evident (1--4) which accounted for more than 70 per cent of the tofM protein, 40--50 per cent of the total protein was present as polypeptide 1 ( Table 2) . Polypeptide 3 appeared as a shoulder on polypeptide 4 in gels of materiM from peaks L I V but could be seen as a separate polypeptide in gels of peak V where polypeptide 4 was greatly reduced. Arch, V~roI. 50/1--2 5 Fig. 6 . Separation of the polypeptides from density gradient peaks. Material from each of the sucrose density gradient peaks was disrupted with 1 per eeng SDS and 2 per cent DTT and separated by polyaerylamide gel eleetrophoresis. Gels were s~ained with amidoblaek. The high molecular weight bands seen nearer the origin than band a were not always apparent in stained gels The approximate molecular weights of the major polypeptides were: 1.52,000; 2.45,000; 3.34,000; 4. 32,000 (Table 2) . Eight or nine minor polypeptides (a--i) were also detectable in gels of all five peaks (Fig. 7) . The proportions of the polyI)eptides were similar for material from all the density gradient peaks with the exception of peak V. In gels of peak V material much higher levels of the minor polypeptides were detected (Fig. 7 , Table 2 ). Occasionally two additional minor bands could be seen in stained gels. These bands had approximate molecular weights of 130,000 and 150,000 but were not usually detected by densitometer scanning. Polyaerylamide gel electrophoresis of material from the density gradient peaks and staining with Schiff's reagent revealed four bands, A --D (Fig. 8) . Band B was present as a minor band which could not always be detected in different preparations of the peaks. The apparent molecular weights of the four bands were: A, 94,000; B, 78,000; C, 52,000; and D, 32,000 (Table 3 ). Bands C and D corresponded in migration to the major polypeptides 1 and 4 and B corresponded to the minor polypeptide e, this indicates that these three are glyeopolypeptides. Band A did not correspond to any amidoblaek-stained polypeptide and staining ) 1V V III Relative Migration Fig. 7 . Polypeptide analysis MateriM from each of the sucrose density gradients was treated as described in Figure 6 . C~ls were stained with amidoblaek and scanned at 620 nm. The origin, is on the left and the anode on the right Fig. 8 . Gtycopotypeptide analysis Eleetrophoresis conditions were identical to those for polypeptide analysis (Fig. 7) . Gels were stained with Schiff's reagent and scanned at 540 nm with Schiff's reagent followed by amidoblack staining showed that although bands B, C and D were superimposed by bands e, I, and 4 respectively, band A migrated between polypeptides b and c, as suggested by the molecular weights. Glyeopol)Teptides which can. be stained with Sehiff's reagent but not with polypeptide stain have been reported by FA:IlaBANKS et al. (1971) in a study of the polypeptides of the human erythrocyte membrane. Our results show that egg-grown IBV, strain Beaudette, produced a broad density distribution of virus particles on continuous sucrose gradients. Use of discontinuous density gradients resolved the virus material into five distinct peaks, all retaining infectivity and CF activity. HOSAKA st g/. (1966) obtained similar results with Sendai virus, but were able to show, on a morphological basis, that the variations in density were directly related to the RNA content of the different particles. If the 3H-uridine activity associated with the density gradient peaks was representative of virus I~NA, our results suggest that the differences in density of IBV particles are not related to differences in R]XA content. Although the densest peak IV/V had four times the amount of SH-uridine cpm per CF unit compared to peaks II and III, peak I, the least dense, had an equally high epm/CF units ratio. It, seems possible, due to the presence of amorphous material and high nucleic acid to protein and CF activity ratios, that peak V, the densest peak, consists of internal virus material from disrupted particles plus very large intact virus particles--these particles could contain more than one genome. In contrast, peaks I~V consisted of apparently whole infectious virus particles. Additional evidence that the amorphous material may be ribonucleoprotein or similar virus material is the morphological resemblence to the ribonucleoprotein of some oncornaviruses (BonoG~SI et al., 1973) . The possibility exists that not all contaminating cellular material was removed by the sucrose density gradient centrifugation procedures employed in this study. However, identical PAGE patterns were consistently produced using different preparations of virus; the PAGE patterns of each peak (with the exception of peal: V) were very similar; examination of the peaks by electron microscopy revealed relatively homogeneous populations with no obvious cellular material present.; PAGE in the absence of SDS and DTT produced no bands for any of the peaks; all of which suggest that if cellular material was present, it was bound to or incorporated into the virus particles. Any assessment of freedom from contaminating cellular material is difficult with a virus as labile as IBV. Populations of coronaviruses have physical to infective particle ratios which may be as high as 104 (J. D. AL~EIDA, quoted by BRADBUR~E, 1972) , for this reason we have used CF units rather than infectivity to give a more realistic comparison of the virus purity. The four hundred-fold purification recorded for peaks II and III compares favourably to the levels of purity obtained using Newcastle disease virus from allantoic fluid (RnEvE and ALS, XA~DER, 1970) . Studies on the morphogenesis of eoronaviruses have revealed two possible mechanisms for release: budding from either the membranes of the endoplasmic reticulmn or cytoplasmic vesicles, but particles have not been seen to bud from the plasma membrane (HAMaE et al., 1967; However some modification of the cell surface may take place since syncytial formation occurs in cells infected with IBV (AxERs and CUNNINGHAM, 1968). BEm~Y and ALMEDIA (1968) demonstrated by electron microscopic examination of virus-antibody-complement interactions that the envelopes of IBV particles were derived from the host cell. It is probable that the corona.viruses, like other envelope viruses have a lipid composition resembling that of the host membrane from which they bud (KLENK and CHOPPIN, 1969, RE~'~ONEX et al., 1971 ) and therefore like Newcastle disease virus (STEa-]~CK and ])UI~AND t963) would be expected to exhibit host specific differences in buoyant density. If this is the ease then two possibilities may exist. Firstly, IBV virions budding from the membranes of different organetles may possess the quMitative and quantitative differences in lipid composition seen in these membranes (gocsnn et al., 1968) . Secondly, IBV grown in the embryonated egg may be produced by budding from the endoplasmie retieulum of cells from more than one organ or tissue. Analytical studies of the tipids from the endoplasmic retieulum of different organs have clearly demonstrated marked variations in composition (I~ovsE~ et al., 1968) . In this respect it is interesting to note that the density distribution of the eoronavirus OC43 which had been adapted to a single organ, the suckling mouse brain, extended only over the range 1.18--1.20 g/cm 3 on continuous sucrose gradients (KA¥]s et al., 1970) . Our results suggest that four major and ten minor polypeptides are associated with IBV virions. This compares favourably with the 16 polypeptides detected in preparations of this virus by BI~I[A~ (personal communication). If both the major and minor polypeptides represent the structural polypeptidcs of the virus, then IBV is unlike coronavirus 0C43 which has been reported to contain only 6 or 7 polypeptides (HIERItOLZER et al., 1972) . Although the apparent IBV ~structural proteins are within the approximate coding capacity of the large genome suggested for IBV by W~TKINS et al. (1975) , it seems more probable that, unlike the paramyxoviruses, IBV virions incorporate host polypeptides as well as lipids [the work of BEan~ and AZ~EID~ (1968) suggests that IBV has a host-derived lipoprotein membrane], or host material becomes closely bound to the virus particles. Either case may account for some, or all, of the minor polypeptides seen in ~his study. We thank Mr. N. J. Chettle for his excellent assistance. Comparison of the structural polypeptides of four avian paramyxoviruses The biology of large I~NA viruses Replication and Cytopathogenicity of avian infectious bronchitis virus in chick embryo kidney cetIs Morphogenesis of avian infectious bronchitis virus and a related human virus (strain 229 E) The morphological and biological effects of various antisera on avian infectious bronchitis virus Localization of RNA turnout virus polypeptides. 1. Isolation of further virus substructures A direc~ complement fixation test for infectious bronchitis virus using heat inactivated chicken sera An investigation of the replication of coronaviruses in suspension cultures of L 132 cells Diseases of swine Avian infectious bronchitis Replication of avian infectious bronchitis virus in African green monkey kidney cell line VEI~O An electron microscope study of the development of a mouse hepatitis virus in tissue culture cells Eleetrophoretie analysis of the major polypeptides of the human erythrocyte membrane Growth and intra-eellular development of a new respiratory virus Morphological variation among avian infectious bronchitis virus strMns Protein composition of corona,virus O C43 Studies on the pleomorphism of HVJ virions Purification and further characterisation of an '~IBV-Iike" virus (coronavii~as) Lipids of plasma membranes of monkey and hamster kidney cells and of parainfluenza virus grown in these cells Protein measurement with the Folin-phenol reagent Recovery in tracheal organ cultures of novel viruses from patients with respiratory disease Buoyant density of hemagglutinating encephalomyelitis virus of swine; comparison with avian infectious bronchitis virus Morphogenesis of avian infectious bronchitis in chicken embryo fibroblasts Electron microscopic studies of coronavirus Isopyenie separation of Newcastle disease virus on sucrose gradients using zonaI ultracentrifuge rotors The lipid class comparison of Semliki forest virus and of plasma, membranes of the host cells Lipid composition of animal cell membranes organe]les Electron microscopy of the hepatocellular and kupffer cell lesions of mouse hepatitis with particular reference to the effect of cortisone Host influence on the density of Newcastle disease virus The nucleic acid of infectious bronchitis virus. Arch. ges. Virusforsch Antigenic eharacterisation of infectious bronchitis virus The morphological development of the virus of transmissible gastroenteritis as compared to other viruses The ribonucleic acid of infectious bronchitis virus Authors' address: Dr. I). J. ALEXANDEI~, Central Veterinary Laboratory, New Haw, Weybridge, Surrey KT 15 3NB, Great Britain.1Received April 24, i975