key: cord-0004861-gxp04dsu
authors: Bingham, R. W.
title: The polypeptide composition of avian infectious bronchitis virus
date: 1975
journal: Arch Virol
DOI: 10.1007/bf01317539
sha: 12ce8410d1ca0fa6f438c4f22bd92af6f8b78f04
doc_id: 4861
cord_uid: gxp04dsu

Avian infectious bronchitis virus grownin ovo was purified by differential centrifugation and isopycnic sedimentation in density gradients. The purified virus was analysed by SDS polyacrylamide gel electrophoresis and found to comprise up to sixteen polypeptides, four of which were glycopeptides. Bromelain treatment of the particles removed three polypeptides and two glycopeptides.

The eoronaviruses have been classified as a separate virus group mainly or~ the basis of their distinctive morphology (12) . There have been many studies [reviewed by McI~TosH (7) ] demonstrating that the growth of these viruses is insensitive to inhibitors of DNA metabolism or DNA dependent RNA synthesis. For this reason the nature of the eoronavirus genome has been assumed to be RNA. However, there have been few reports of the results of direct chemical analyses of the virion. There have been two reports (10, ll) of extraction and analysis of nucleic acid from purified virions of avian infectious bronchitis virus (IBV) and these indicated that the virus contains I~NA. It has also been deduced from the appearance of the particles, the sensitivity of infectivity to ether and chloroform, and the mode of development of particles observed in thin sections of infected cells, that they contain lipid derived from the host as an integrM part of the outer shell. However, there are no reports on extraction and anMysis of lipids from purified particles. Although there are at least seven different, species in the eoronavirns genus (7) , only one t33oe, the human eoronavirus OCI3, has been subjected to electrophoretic analysis of the polypeptides (5) . This report describes preliminary biochemical investigations of another coronavirus, avian infectious bronchitis virus. (a Massachusetts serotype) from the commercially available live vaccine, Bronvirin (Glaxo Ltd.). Each of these stocks had a long history of passage in embryonated cMcken eggs before their receipt in this laboratory.

Primary chicken kidney cell cultures were used for production or infectivity assays of IBV. The kidneys from 3-7 week old specific-pathogen-free chickens (Wickham Laboratories Ltd., ~Vickham, Hampshire) were dispersed by serial 5 minutes incubations at 37 ° C in 0.05 per cent trypsin (Difeo Laboratories) in phosphate buffered saline (PBSA: Dulbecco's type A, Oxoid Ltd.). The cells obtained were strained through nylon mesh and dispensed either in Eagle's Medium Basal (BME :'Wellcome Reagents Ltd. ) or in L-15 medium (Gibco), both media being supplemented with 8 per cent foetal calf serum, 4 per cent tryptose phosphate broth, 100 units/ml penicillin and 100 ~g/ml streptomycin. The BME was used for cells grown in vented plastic dishes in a 5 per cent CO2 atmosphere, the L-15 used for growth of cells in closed glass bottles or tubes. For the growth of virus, confluent cell cultures were changed into maintenance medium, which was similar in composition to the growth medium but without the foetal calf serum. Estimations of virus infectivity were made by inoculating 0.2 ml aliquots of appropriate dilutions of virus into rolling tube cultures already containing 1 ml of maintenance medium. Cytopathic effects on the cells were observed at 48 hours (strain Beaudette) or 72 hours (strain Massachusetts 4t and Connecticut) after inoculation, and infecti~dty measurements experssed as TCDs0/0.2 ml. IBV-H t20 caused no distinct cytopathic effect and was not used in experiments requiring infectivity measurements. IBV-Beaudette could also be estimated by plaque formation assays, but this method was not routinely used. All cell culture and virus growth was performed at 37 o C.

Usually 104~105 TCD~0 of virus was inoculated (in 0.05--0.2 ml) into the allantoic cavity of 10-day old embryonated chicken eggs (Orchards Farm, Gt. Missenden, Bucks) and incubated at 37 ° C for 24 hours. After chilling at 4 ° C overnight the allantoic fluid was harvested and immediately clarified by centrifugation at 16,000 g for 20 minutes. This and all subsequent operations were carried out at 00--4 ° C. The virus was pelleted at 75,000g for 1 hour in a MSE No. 59595 rotor and resuspended in PBSA with the aid of a glass Dounee homogeniser. The resuspended virus was overlaid on a linear 25 to 55 per cent sucrose (w/w) in PBSA gradient and centrifuged overnight (12--16 hours) in a MSE No. 59590 rotor at 90,000g. The virus bands were collected, diluted 8-fold in PBSA and repelleted at 75,000g for 1 hour. The pellets were resuspended in the same manner as before and layered on either (a) a linear 25--55 per cent sucrose (w/w) in PBSA gradient, or (b) a linear 25--55 per cent potassium sodium tartrate (w/w) in PBSA gradient and centrifuged overnight at 60,000g in a MSE No. 59108 rotor.

Purified virus collected from the second gradient step was diluted in PBSA, deposited at 75,000g for 1 hour and dissolved in a small volume of 5 per cent sodium dodeeyl sulphate (SDS) at 37 ° C for 30 minutes. 2-mercaptoethanol was added to a final concentration of 2 per cent and the solution incubated in a boiling water bath for 2 minutes. The dissolved virus was dialysed overnight at room temperature against 5 m~ Tris, 38 mM glyeine buffer (pH 8.3) containing 1.0 per cent SDS, 0.I per cent 2-mercaptoethanol, 3.3 ~ Urea and 5 per cent sucrose. After dialysis a trace amount of bromophenol blue was added and the sample reheated at 100 ° C for 1 minute immediately prior to eleetrophoresis.

Electrophoresis was carried out on cylindrical potyacrylarnide gels 12 cm × 0.65 cm diameter, cast in perspex tubes. The acrylamide concentration was 10, 7.5 or 5 per cent (w/v), the acrylamide: bis-acrylamide ratio being constant at 37.5:1 by weight. The gels also contained 0.1 per cent SDS, 0.5 ~'I urea, 0.03 per cent N, N, N', N'--tetramethy]ethylenediamine, 0.07 per cent ammonium persulphate and 0.375 5I Tris--I-IC1 buffer, pH 8.9. Gels were pre-electrophoresed at 100V for 1 ~--2 hours, the electrophoresis buffer consisting of 25 m~ Tris, 192 m~ glycine, 0.1 per cent SDS, 0.5 ~ urea., 0.1 per cent 2-mereaptoacetic acid, pH 7.8. If non-reducing conditions were required, the 2-mercaptoacetic acid was omitted----the buffer then being pH 8.3-and reductant omitted during preparation of the sample. After pre-electrophoresis the sample was layered directly on the top of the gel in not more than 200 B1 voiume, and eleetrophoresis carried out at 100V for 4--8 hours. In some experiments the system described by ~VEBE~ and Osso~ (t5), with phosphate buffer pI~ 7.0, was used. After eleetrophoresis gels were stained for protein (9) or carbohydrate (16) . The molecular weights of viraI peptides were determined by reference to the mobilities of fi-galactosidase, phosphorylaae A, transferrin, bovine serum albumin, pyruvate kinase, ovalburain, alcohol dehydrogenase (yeast), myoglobin and lysozyme in control gels.

Fractions from gradients were exhaustively dialysed against distilled water to remove the sucrose or tart.rate gradient material. Concomitant changes in volume during dialysis were noted. Estimations were carried out for RNA (8) , protein (6), I)NA (2) and carbohydrate (4) . Alternatively protein and t{NA were estimated from the extinctions at 260 and 280 nm (E~6o and E2so) (13) .

IBV from the first sucrose gradient stage of the purification was incubated with bromelain (Sigma Chemical Co.) (3) . After incubation, the mixture of virus and enzyme was diluted in ice-cold PBSA, pelleted, resuspended and centrifuged on a second sucrose gradient as described above.

On the first sucrose gradient stage the virus formed two distinct bands, Table 1 indicates the buoyant densities of the different strains. The two bands were put on separate gradients for the second stage of the purification, and they retained their original density as a single band. The various bands from both gradient stages were negatively stained in sodium phosphotungstate and examined with the electron microscope by Dr. J. D. Almeida, Wellcome Research Laboratories, and were found to contain intact virus ; there was no detectable difference between the light and heavy forms. A small proportion of the virus particles became degraded on tartrate gradients, so in subsequent purifications the tartrate was replaced by a secor~d sucrose gradient.. The infectivity of the purified virus fractions was measured, and a typical result is presented in Table 2 . It, can be seen that both the light and heavy forms of the virus were i.niectious, although the heavy band was significantly more infectious with respect to protein and nucleic acid. The second gradient step appears to have caused a loss of infectivity of the heavy form, and a reduced RNA: protein ratio of that band, suggesting a partiM loss of nucleic acid.

It was thought, that the virus might be contaminated by mycoplasmas, which are often present in hens eggs and which may be of similar size and density (if). The infected allantoic fluids were therefore tested for viable mycoplasmas by 1~. W. BI~G~A~:

Miss Patricia M. Furr of this Institute but the results were negative. Purified IBV preparations were tested for avian leukosis virus (ALV) by Mr. T. Webster, Wellcome Research Laboratories, using the complement fixation test (COFAL) and these were also negative. No particles resembling mycoplasmas, ALV or any other virus were ever seen in any of the electron microscopic examinations of purified virus. (13) . e Material from 1st sucrose gradient after rebanding on tartratc gradient.

It was not possible to obtain sufficient, purified virus to make a complete analysis of nucleic acid, protein, carbohydrate and lipid content and relate this to the dry weight of virus material. However, it was possible to make some chemical determinations and relate these to the protein content of the particles. The RNA: protein ratio ranged from 0.71--0.84 (5 batches of virus) when measured by chemical means (8, 6)--on measurement by extinctions, the ratios ranged from 0.036---0.069 (9 batches of virus), the average being 0.054. The amounts of RNA determined by the two methods were in very close agreement. The discrepancy in the RNA: protein ratio arises from the very different results obtained from the two methods for protein estimation. As the method of LOWRY et at. (6) is rather sensitive to interference by a variety of compounds, including sugars, it is possible that the viral nucleic acid, carbohydrate and lipid have caused sufficient interference to render the result from this method invalid. The carbo-hydrate: protein ratio ranged from 0.11--0.22 (3 batches of virus)--the protein being estimated by E260/E2s0 measurements. The carbohydrate values were corrected for the RNA content of the preparations. Attempts were also made to detect the presence of I)NA in purified virus. Although trace levels were detected, these levels were within the calculated levels of interference in the diphenylamine test by the viral I~NA present. I t was eoneluded that, there was probably no genuine I)NA in the preparations.

These results are the first direct chemical confirmation that the virion nucleic acid is indeed RNA, and also that the particles contain carbohydrate in addition to the nucleic acid and protein. Other experiments in progress in this laboratory (S. JOH~S0:N and R. W. BrNo~AM--unpublished results) have confirmed the presence of various lipids in purified preparations of the virus.

Initial experiments using the Tris-glycine buffered electrophoresis system gave much better resolution of the IBV polypeptides than that achieved with the phosphate system (Fig. 1 )--some of the bands visible in the former did not appear Fig. 1 . PAGE of IBV. A. phosphate buffer system, B. tHs-glyeine buffer system, 10 per cent aerylamide m both systems, identical samples loaded on all four gels. P. stained for protein, C. stained for carbohydrate. As the photographic process did not record the red bands as e/early as the blue, the positions of the carbohydrate bands seen with the naked eye have been marked alongside the appropriate gels It. W. BINc~nA~: at all in the latter. Therefore all subsequent experiments were done by the former method. Figure 2 illustrates typical densitometer tracings (made with a Joyce-Loebl Chromoscan) of gels which have been stained for either protein or carbohydrate. There are up to sixteen peptides present, three or four of which appear to be glycosylated. Although the profiles of the light and heavy forms appear to be dissimilar (Fig. 3a, b) , it may be seen that the same peptides are present in both, and only the relative proportions differ. Examinations of many such profiles showed that this was a reproducible phenomenon. No significant difference between strains was detected. Numbers were assigned to the peptidcs on the basis of their molecular weights to assist in comparison of profiles. The molecular weights determined from twelve experiments are listed in Table 3 . When the separated light, and heavy forms of the virus were treated with bromelain and recentrifnged on sucrose gradients, it was found that both forms yielded identical single bands of virus at a new lighter density of 1.15. These treated particles were examined by electron microscopy (by Dr. S. Patterson of this Institute) and were almost totally lacking in projections. PAGE profiles of the virus before and after treatment are illustrated in Figure 3 . I t can be seen that the bromelain removed for carbohydrate. Gels run for 5 ~/2 hours. Only those bands consistently found in every preparation have been labelled VP or VGP. 0eeasionally VP 70 showed slight staining with PAS as seen here, but this was not reprodueible two of the glycopeptides (VGP180 and VGP83) and substantially reduced the proportions of VP 130, VP 106 and VP70. As VP53 and VP49 usually tend to run as an overlapping pair, it is not possible to say whether any VP53 remains. The relative proportions of the lower molecular weight peptides have considerably increased in relation to the prominent VP 49. Three new peptides, designated XP 27, XP20 and XP18 (27,000, 20,000 and 18,000 daltons respectively) have also appeared in the profile. It seems likely that these are cleavage products of some of the larger peptides, rather than contamination with the bromelain; when brome- heavy form gave identical pattern). Gels run for 6 hours lain was electrophoresed in a parallel gel it gave several peaks, none of which coincided with the new peptides. Also it was ealculated that about 90 per cent of the added bromelain would have had to remain adhering to the virus through the purification steps after incubation in order to yield the amounts detected of the new peptides. These results suggest that VGP180, VP130, VP106, VGP83 and VP70 are accessible on the surface of the vMon, the remaining peptides presumably being internal. Newcastle Disease virus (strain Hefts) and influenza virus (strain X31) were grown in ovo, purified by exactly the same techniques and electrophoresed in parallel to the IBV preparations. Only the typical peptide bands of those viruses were detected, and there were no bands corresponding to any of the bands located in the IBV samples. This was evidence that the purification scheme used was adequate for isolating pure specimens of other enveloped viruses of similar size, density and origin to IBV without host component contamination. 

The prime question which arises when considering the observed polypeptide composition of any virus :is that of the purity of the preparation. This is especially important when there are apparently a large number of peptides present, as in this ease. The four main contaminants that may arise in purified preparations of most animal viruses are host cellular components, endogenous virus from those cells, mycoplasmas and bacteria. The second and third possibilities have been tested for and excluded in this ease, and the fourth would have been very obvious in electron microscopic examination. Also any significant level of bacterial or mycoplasmM contamination would have resulted in readily detectable quantities of DNA in the preparations. This leaves the problem of possible contamination by host cellular peptides. It is difficult to differentiate between host material that is an integral part of the virion and that which may have remained associated with the preparation through the various stages of the purification. One approach to the problem would be a comparison of virus grown in different cell types on the assumption that the peptides contributed by the host would differ in the various preparations. However, this would not resolve the question of whether it was an integral part of, or merely associated with, the virion. This is not possible with IBV anyway, due to the very restricted selection of cells in which it may be grown--the strains used here were only capable of culture in ovo or in primary chicken kidney cells. In the samples of purified virus used here, there were no major contaminants visible by electron microscopy. Although there were occasional pieces of membrane, some clearly deriving from disintegrating virions, the relative amounts of virus particles and possible host components were such that any contribution of the latter to the analysis would have been negligible. In some preparations examined there was a "contamination" by detached viral projections, mainly in the heavy band--as the amounts of these present varied from sample to sample, this could account for the difference in the relative proportions of the polypeptides observed in the light and heavy forms. This leaves the possibility that host polypeptides may be an integral part of the virion. It has been shown (1) that when IBV was treated with unheated rabbit antiserum raised against normal chicken cell membranes, holes appeared in the virus envelope, similar to those produced using rabbit antiserum raised against the virus itself. Use of chicken antiserum did not cause these holes. This suggested that host components were an essential part of the virus envelope.

A recent report (14) describes the I~NA content of the virus as being in one ribonuclease sensitive piece, of molecular weight 9 × 106 daltons. If this represents the size of the IBV genome, then the virus would contain adequate coding capacity to account for all the peptides observed, assuming they are all independently coded and some are not cleavage products of others, eliminating the necessity to postulate the presence of peptides of cellular origin.

It is of interest to compare these results with the report on the polypeptides of the human coronavirus 0C43 (5) . Although those authors only claim the presence of six or possibly seven polypeptides, examination of their gel tracings shows the presence of several more minor bands. Also, they used a phosphate buffered pH 7.2 PAGE system very similar to that which gave comparatively poor resolution with IBV. Thus the avian and human coronaviruses may not be as dissimilar as is at first apparent.

The morphological and biological effects of various antisera on avian infectious bronchitis virus

A study of the conditions and mechanisms of the diphenylamine reaction for the eolorimetric estimation of deoxyribonucleic acid

Influenza virus proteins. I. Analysis of polypeptides of the virion and identification of spike glycoproteins

Colorimetric method for determination of sugars and related substances

Protein composition of coronavirus OC43

Protein measurement with the Folin phenol reagent

Coronaviruses: a comparative review

The determination of small amounts of pentose, especially in adrenal acid derivates. Hoppe-Seyler's

Sendal virus structural proteins: analysis of polypeptides linked with disulphide bonds

The nucleic acid of infectious bronchitis virus. Arch. ges. Virusforsch

Isolierung und Kristallisation des G/~rungsferrnents Enolase

The ribonucleic acid of infectious bronchitis virus

The reliability of molecular weight determinations by dodecyl sulphate--opolyacrylamide gel electrophoresis

Glycoprotein staining following eleetrophoresis on aerylamide gels

Department of Veterinary Pathology, l~oyat (Dick) School of Veterinary Studies