key: cord-0912955-id5z8rtv
authors: Descôteaux, J. -P.; Lussier, G.; Berthiaume, L.; Alain, R.; Seguin, C.; Trudel, M.
title: An enteric coronavirus of the rabbit: Detection by immunoelectron microscopy and identification of structural polypeptides
date: 1985
journal: Arch Virol
DOI: 10.1007/bf01378976
sha: ae15dc29747fd52191111c58ddd721c93e6a4bb7
doc_id: 912955
cord_uid: id5z8rtv

The immunoelectron microscopy (IEM) technique has been used for the detection of a rabbit enteric coronavirus (RECV). Immune serum was prepared in guinea pigs; the viral antigen used for the immunization procedure was obtained from the caecum of a sick rabbit, concentrated by centrifugation and purified on Percoll gradient. In order to identify the viral particles used in the immunization procedure, the protein pattern of the particles was determined by electrophoresis and compared with the pattern of other known coronaviruses. Analysis of structural polypeptides of the purified viral particles revealed a pattern similar to that reported for other coronaviruses. These polypeptides cross reacted with two other coronavirus specific immune sera (IBV and TGE). IEM assay of fecal samples collected from healthy and sick rabbits showed the presence of immune aggregates in specimens from both sick and healthy rabbits. Those aggregates contained viral particules sharing morphological characteristics with other coronaviruses. Furthermore, IEM assay was shown to be more sensitive than a direct EM procedure to detect coronavirus particles in rabbit feces. This assay also allowed the detection of a larger number of chronic carriers.

The immunoelectron microscopy (IEM) technique has been used for the detection of a rabbit enteric coronavirus (RECV). Immune serum was prepared in guinea pigs; the viral antigen used for the immunization procedure was obtained from the caecum of a sick rabbit, concentrated by centrifugation and purified on Percoll gradient. In order to identify the viral particles used in the immunization procedure, the protein pattern of the particles was determined by electrophoresis and compared with the pattern of other known coronaviruses. Analysis of structural polypeptides of the purified viral particles revealed a pattern similar to that reported for other coronaviruses. These polypeptides cross reacted with two other coronavirus specific immune sera (IBV and TGE). IEM assay of fecal samples collected from healthy and sick rabbits showed the presence of immune aggregates in specimens from both sick and healthy rabbits. Those aggregates contained viral particules sharing morphological characteristics with other coronaviruses. Furthermore, IEM assay was shown to be more sensitive than a direct EM procedure to detect coronavirus particles in rabbit feces. This assay also allowed the detection of a larger number of chronic carriers.

The electron microscopy (E.M.) technique is commonly used to detect the presence of coronavirus particles in fecal samples collected from different animal species (3, 8, ll, 14, 18) including man (13) . However, in suspensions of fecal materials, numerous particles do not always exhibit the typical coronavirus morphology and consequently are often refered to as viruslike or coronavirus-like particles (5, 7) . This is particularly the case when rabbit fecal material is evaluated for the presence of corona~-iruses. Particles have been observed in fecal material collected from young rabbits with clinical signs of enteritis (6, 11, 12) . However, very tittle is known about these particles. They share some morphological characteristics with other members of the family Coronaviridae but most often do not show typical surface projections. Those structures are either shorter than the reported length or not clearly defined and appear as a fuzzy coat. Furthermore, according to our experience, coronavirus particles are often present in low titer in rabbit feces. Proper identification is then difficult to achieve and it has become necessary to develop a more sensitive assay. For that purpose, we characterized the structural polypeptides of the viral particles in rabbit feces and we adapted an immunoelectron microscopy (IEM) technique to evaluate the presence of rabbit enteric coronavirns (RECV) in rabbit feces.

Healthy rabbits were obtained from a commercial colony with a tow incidence of enteritis. Sick animals were randomly selected among the specimens submitted to The Laboratory Animal Diagnostic Service of the Armand-Frappier Institute.

A total of 30 fecal samples were examined. Seventeen were collected from healthy animals and 13 were obtained from young rabbits showing clinical signs of enteritis. The samples were collected from the caecum diluted in phosphate buffered saline as a 20 per cent suspension, centrifuged at 3000 × g for 30 minutes in a refrigerated centrifuge and the supernatant frozen at --70 ° C until used.

Fecal material collected from the caecum of a sick rabbit was prepared as above, purified and characterized.

For purification procedures, the materiM was first diluted 1:5 in NTE buffer (0.15 ~ NaCI, 0.05 M Tris HC1, 0.001 M EDTA, pH 7.4) and examined by E. M. for the presence of viral particles. Aliquots (100 ml) were clarified by low speed centrifugation and banded isopycnically oa Percoll TM self generating isoosmotic gradients (Pharmacia, Uppsala, Sweden) adjusted at 1.1 g/cc. Centrifugation was carried out in a type 30 rotor (Beckman, Palo, Alto. Co.) at, 30,000 rpm for 30 minutes at 4 ° C. Sucrose gradient centrifugation was carried out on 20--50 per cent gradient to determine particle density, using an SW 40 rotor. The HA activity was used to monitor the presence of viral particles in the fractions. The assay was done in U-type microplates using rabbit erythrocytes as previously described (1 t). Peak fractions were stored at --70 ° C until further use. An aliquot of peak fractions showing an elevated I-IA activity and con-taining R E C V as observed by E M was used to determine the protein pattern of the viral particles.

Electrophoresis under dissocating conditions was carried out using the discontinuous buffer system described b y LAEMMLI (1970). Stacking and resolving gels consisted respectively of 3 per cent acrylamide (0.08 per cent bis-aerylamide) and 10 per cent acrylamide (0.32 per cent bis-acrylamide). Electrophoresis was carried out at 65 mA. Separated proteins were then transfered eleetrophoretieally on nitrocellulose sheets (20) . Blotting was performed at 50 v for 2 hours in buffer. Nitrocellulose sheets were then prepared for immunodeteetion by washing with buffer and incubating with guinea pig i m m u n e serum specific for R E C V prepared in our laboratory. Bound specific antibodies were detected using a fluorescent goat anti-guinea pig IgG serum (Miles Laboratory, CA, U.S.A.). Bands were visualised by direct illumination by short wave U V light (Chromato-vue cabinet; U V P Inc. CA) and photographed with a Polaroid t y p e 107 film using a 480 filter.

Anti I B V (avian infectious bronchitis), and anti T G E (porcine transmissible gastroenteritis sera were obtained from the Centre de recherche en mddecine v6tgrinaire, I n s t i t u t Armand-Frappier. Fluorescent antisera to guinea pig, chicken and porcine IgG were obtained commercially (Miles Laboratories, CA, U.S.A.).

purified R E C V as observed by E M were used to immunize two guinea pigs. The animals were first bled from the venous orbital sinus to obtain pre-immunization serum and then immunized with 0.5 ml of the H A fraction mixed with an equal a m o u n t of complete F r e u n d adjuvant. The preparation was inoculated subcutaneously and the animals were bled six weeks after to determine the presence of specific hernagglutination inhibiting (I-II) antibody titers (11) . The sera obtained were frozen at --7 0 ° C. Whole unheated serum was used throughout the evaluation procedure.

Aliquots (25 ~l) of fecal suspensions and of peak fractions used for immunization procedure were deposited onto the surface of an agar mierowell to concentrate the samples and a c a r b o n -F o r m v a r coated copper grid was placed face downward on top of the drop. After complete extraction of the aqueous phase into the agar phase, the grid was picked up, stained with 3 per cent phosphotungstic acid (PTA) pH 6.8 and examined with a Philips EM 300 electron microscope. Each specimen was examined for I0 minutes or less if viral particles were observed.

H A fractions from the Percoll gradient used for immunization and fecal samples collected from healthy and sick rabbits were evaluated for the presence of R E C V . The specimens were first filtered through a 0.22 n m filter to eliminate natural aggregates. T w e n t y five ~1 of the filtrate obtained were then m i x e d with an equal q u a n t i t y of specific antiserum diluted 1/40 and deposited on an agar well as described above. A carbon formvar-coated grid was overlaid on top of the drop and after almost complete hydroextraetion, the grid was stained with P T A and examined by EM. Control grids were prepared using pre.immunization serum sample. Each specimen were examined for 10 minutes or less if i m m u n e aggregates were observed.

The exact Fisher's test (t) was used to analyse the results obtained.

RECV banded isopycnieally at a density of 1.06 g/co on Percoll gradients (Fig. 1) . The presence of complete viral particles in the peak fractions {9---12) were identified by HA assay (Fig. 1) and confirmed by EM examination (Fig. 2 a) . On sucrose gradients, the viral particles banded at a density of 1.18. I  I  I I I  I  I  I I  I  ~ I  I  I  I  I I  I |  I  I  2 Coronavirus particles (Fig. 2 a. ) appeared as spherical enveloped particles with a diameter ranging between 40 and 50 nm without peplomers. An inner tongue shape and a double shell envelope were also seen in a large number of particles. Peplomers between 10--12 nm in length were observed but were often broken. Fractions 13 to 22 contained mostly viral fragments or aggregated hemagglutinin.

HI antibody titers were found in guinea pig serum following immunization procedure. No detectable titers were observed in pre-immune serum. When an aliquot of diluted immune serum was reacted with the viral HA fraction used for immunization, immune aggregates were observed by IEM (Fig. 2b) . The immune aggregates formed showed the presence of numerous coronavirus particles of variable sizes surrounded by antibodies.

Electrophoresis pattern reacted with guinea pig anti-RECV revealed 8 structural proteins ranging from 100,000 to 34,000 molecular weight: 7 were associated with the envelope (100,000, 82,000, 81,000, 76,000, Table t) . I m m u n o b l o t t i n g of the same protein profile with a chicken anti-I B V serum, permit the identification, by crossed reaction, of only 6 of the polypeptides: 5 associated with the envelope (100,000, 82,000, 76,000, 71,000 and 34,000) and the 54,000 nucleoprotein ( Fig. 3B and Table 1 ). Significantly different (p=0.004) b Significantly different (p = 0.023) observed by EM, the viral particles were difficult to identify. They were in limited number not well defined and the surface projections were generally damaged (Fig. 2c) . However when the IEM assay was used, immune aggregates were easily detected (Fig. 2d) . These aggregates were formed b y particles of various sizes surrounded by immunoglobulins. Although the surface projections were short and most often covered b y antibodies, they were distinguishable in some particles. The results of the statistical analysis indicate a significant difference (p=0.023) between the number of positive specimens collected from sick animals. A significant difference was also obtained (p=0.004) between positive specimens collected from healthy animals.

For some time now, a coronavirus has been implicated in the etiology of rabbit enteritis (11) . Morphological evidence based on EM studies of fecal samples had lead to this hypothesis. Data are now presented to support our preliminary report of a coronavirus enteric infection in rabbits. Viral particles purified from fecal material collected from the caecum of a rabbit showing clinical signs of enteritis were purified on Percoll gradient. Their morphology and antigenicity were well preserved. Particles banded 17" at the same density reported for other enveloped viruses (17) . In sucrose gradients, virus banded at a density of 1.18, as reported for other coronaviruses (16) . Furthermore, a specific antiserum was prepared in guinea, pigs and was used to identify the structural proteins of the RECV by immunedetection of virus protein blots on nitrocellulose and to develop an immuneelectron microscopy assay for the rapid diagnosis of this specific virus in rabbit feces.

The eleetrophoretic profile of the structural proteins confirms that these virus particles had the same structural protein profile as member of the family Coronaviridae. The pattern of structural proteins should be exemplified by the group prototype, IBV which has been shown by Cavanagh (1981) to contain three membrane proteins of 94,000, 84,000 and 30,000 and a nueleoprotein of 54,000 molecular weight. The pattern of enveloped protein of RECV is more complex consisting of five proteins 100,000, 82,000, 81,000, 76,000, 71,000, 65,000 while good agreement is reached for tile nucteoprotein at 54,000 and the matrix protein at 34,000. Since we already known, by EM studies, that the viral peplomers are damaged, it is not unexpected that we find multiple bands in the 82,000 to 65,000 molecular weight which could be broken peplomers.

Our results with the two other coronavirus specific sera (IBV and TGE) is even more interesting. Since both these sera recognized the 34,000 matrix protein, the 54,000 nuc]eoprotein and the 100,000, 82,000 and 71,000 molecular weight envelope protein. Both the t00,000 and 82,000 are recognised peplomer proteins (4, 16) and a protein between 60,000 to 70,000 has been described for mouse hepatitis, bovine coronavirus and porcine haemegglutinating encephalitis virus. Our results are in good agreement with the typical coronavirus profile. The 76,000 molecular weight protein reeognised by the chicken anti-IBV is somewhat of a puzzle: it could be a higher molecular weight protein that has been partially degraded, since EM observation confirms that the spike protein is broken, but it is difficult to understand why the porcine anti-TGE does not recognize it. However, the data obtained support the h)xpothesis that this is indeed a coronavirus, and that there are cross reactions with two other coronaviruses. It is to be noted that by blotting we are identifying antibodies that recognise mostly primary and secondary structure epitopes which are likely to be more conserved that tertiary and quaternary epitopes. Those at the surface of the viral particle could be recognized by blocking biological activity such as infectivity or by IEM. This specific virus has not yet been cultured in vitro. However, its presence can still be easily detected by IEM assay. The results of this study indicate that this procedure is a useful method for the detection of I~ECV. The observation of immune aggregates not found in controls, the diameter of the particles observed and the presence of short peplomers indicate that these particles are coronaviruses. The electro-phoresis s t u d y performed in the viral particles used for the immunization procedure confirm their identification. When these particles are observed by EM alone, t h e y are found in limited n u m b e r and are often difficult to identify. The use of specific antiserum contributes then to facilitate the diagnosis of R E C V b y eliminating the ambiguities on the identification of this specific virus and b y reducing the observation time requested for its identification. The coronavirus particles are readily found in aggregates surrounded b y antibodies and the observer does not have to rely only on a specific morphology for identification. Furthermore, the results of statistical analysis confirm, as expected, the superiority of IEM over EM. The I E M assay is used in the diagnostic field to evaluate the presence of virus in specimens from various sources (9, 15) . In our laboratory it is used for the detection of h u m a n viruses in fecal material submitted for diagnosis purposes (2) . I t allows the identification of specific viruses and also the detection of viral particles present in limited numbers. The results of the present s t u d y indicate t h a t this assay is also a valuable procedure for detection of R E C V in rabbit feces.

H e a l t h y carriers have also been detected in the rabbit population. Positive specimens in healthy animals were detected with both procedures used. However I.E.M. seems to detect more healthy carriers t h a n E.M. I t is not surprising to find healthy carriers since theyse have already been reported in different species (3, 8, 19) including man (13) . However it is the first time t h a t healthy carriers of coronavirus are reported in the rabbit species. These carriers are quite likely responsible for the persistence of the infection in certain colonies where mortalities due to enteritis are high. The I E M then becomes a useful tool to recognise the infection and to p r e v e n t the introduction of infected stock in a colony. The results of the screening of all new animals introduced could contribute to the control of enteritis due to coronavirus and thus considerably decrease the economical loss due to high mortality rate.

Statistical Methods in Medical Research

Rapid detection of human viruses in feaces by a simple and routine electron microscopy technique

Coronavirus-like particles present in simian faeces

Structural po]ypeptides of coronavirus IBV

Virus-like particles 45 to 65 rim, in intestinal contents of neonatal calves

Preliminary observations on enteritis associated with a eoronaviruslike agent in rabbits

Winter dysentery: a coronavirus-like agent in the faeces of beef and dairy cattle with diarrhea

Corona-like particles in the feces of normal cats

Detection of coronavirus strain of 692 by immune electron microscopy

Cleavage of structural proteins during the assembly of the head of bacteriophage T4

Preliminary report on the observation of a coronavirus in the intestine of the laboratory rabbit

VAN: Coronavirus-like particles in laboratory rabbits with different syndromes in the Netherlands

Association d'infeetion ~ coronavirus avee l'ent6rocolite h6morragique du nouveau-n6

A new coronavirusdike particle associated with diarrhea in swine

Electron microscopy of hogcholera virus and its antigen-antibody complex

The Biology of Coronavirus

Separation of Herpes simplex virus virions and nucleocapsids on Percoll gradients

Coronavirus infection in a litter of pups

The biology and pathogenesis of eoronaviruses

Electrophoretic transfer of proteins from polyaerylamide gels to nitrocellulose sheets: Procedure and some applications

The authors would like to thank Drs. Gr4goire Marsolais and Enrico Di Franco for providing anti IBV and TGE sera and Ginette Godin and Pierrette Lessard for their technical assistance.

Authors' address: Dr. J.-P. DESCSTEAUX, Centre de Recherche en Virologic, Institut Armand-Frappier, 531, Blvd. des Prairies, C.P. I00, Laval, Quebec, Canada H7N4Z3.Received June 4, 1984