|
Archives of Virology "Runde" Virus, a Coronavirus-Like Agent Assoeiated With Seabirds and Tieks
Abstract
From 206 1. uriae collected in the seabird colonies at Runde, Norway, two identical virus strains demonstrating no antigenic relationships to major arbovirus groups were isolated. The new strains demonstrated a corona-virus like morphology, haemagglutinated chicken red cells and were sensitive to sodium desoxycholate. Multiplication with CPE was demonstrated in BHK 21/c13 and BSCd cells, and without CPE in Vero and GMK cell cultures. The mouse pathogenicity was relative]y low. In gel precipitation three to five specific lines were seen. Precipitating antibodies have been found in seabird species commonly infested by I. uriae.
The ecological circumstances of the isolates indicate an earlier unrecognized arbovirus circulating between seabirds and I. uriae. This corona-like virus has been tentatively termed Runde virus.
In 1973 investigations were undertaken to evaluate the extent, distribution and ecological circumstances of arbovirus loci in Norway. Until then., no arbovirus isolates had been reported from this country, although ecological and cli-nicaI/epidemiological considerations (3, 24, 26) and a limited serological survey on bovine sera (28) indicated the existence of Central-European tick-borne encephalitis virus fool.
The early phase of the project concerns the study of the importance of ticks as vectors; later, mosquito-borne agents will be investigated.
In Norway five Ixodes species are abundant: _[. ricinus (23) , I. tria.nguticeps (24) , I. hexagonus (17) , I. tividus (16) and 1. uriae (15, 23) . The latter species is present at great numbers in the vast Norwegian seabird colonies. Its ability to attack man and mammMs has been questionable, but this has been recently documented (18) .
This paper describes studies with Ixodes uriae as a possible vector. These studies were undertaken since:
a) The tick is part of rather confined ecosystem which might be suitable for studies on virus-vector-host interrelationships.
b) Virus-transmission by this tick might have mainly zoonoticM but also some public health importance. c) Reports of arbovirus isolates from seabird colonies have been presented earlier (9, 10, 12, 32, 33) .
The seabird colonies at Runde were chosen from a great number of possibilities due to relative ease of access, and because an unusually high and unexplainable chick mortality had recently been reported.
From 206 I. uriae ticks collected at Runde in late September 1973, three virus strains have been isolated. One of these isolates belongs to the Uukuniemi group (30, 31) .
The additional two strains are serologically closely related or identical. Some characteristics and ecological circumstances of the latter viruses are reported in the present article.
Description o/ the Biotope The island Runde is situated at 62°25'N ' 5°38'E, approx. 30 kin from the city Alesund (Fig. 1) . The total area is about 6.4 km 2 and the circumference 20 kin. There are two small settlements on the island, l~unde in southeast and Goksoyr in northeast :
Most of the island consists of an uneven mountain platea.u, and has altitudes varying between 100 and 330 metres above sea level. In the southwest part, the plateau is terminated by cliffs bordering the sea. Most seabird colonies are situated in this area. Approaching sea-level there is naked stone with niches and clefts inhabited mainly by Rissa tridactyla, Sula bassana, Alca torda and Uria aalge. Further up towards the plateau there are rockfalls covered by thin layers of soil supporting various gramnivor plants. These parts are inhabited by the Common puffin (Fratercula arctica).
We visited the island between September 25--27, 1973 , and collected Ixodes uriae in the puffin rockfalls. According to local ornithologists, the puffins had migrated South about 3 weeks earlier. We found the ticks under and between the rocks. Signs of engorgement were never seen, and they seemed to have entered into diapause. Most of the ticks were to varying degrees surrounded by desiccated skinflaps indicating that they had newly emerged to their present stage. Some had not completed their development from nymph to adult, and were totally surrounded by nymph cutis. The ticks were kept on dram-vials with moist plaster of Paris, and transported alive to the laboratory. We collected a totM of 257 I. uriae, 206 of which were processed for virus-isolation (Table 1) .
Isolation Procedures The ticks were divided according to stage and sex, pools of imagos consisting of 5 and nymph pools of 20 individuals. After rinsing in saline, the ticks were ground in a mortar in 0.4 per cent bovine albumin in PBS pI-I 7.4 (APBS) with mycostatin, streptomycin and penicillin. Suspensions were clarified by centrifugation at 1500 × g and inoculated i.c. into baby mouse litters, Born: NMRI (SPE), aged one to three days. Suspensions not inoculated the same day as processed were kept at --70°C (l~evco UltrMow). Mice were observed during 3 weeks for signs of illness. Diseased mice were killed, brains removed aseptically and homogenated in APBS with antibiotics, centrifuged and passed into new baby mice. Litters which remained healthy were killed after 3 weeks, brains processed and passed as described. Two blind passages were performed routinely. Calculations of 50 per cent end point, titers, LDs0 baby mice (BMLDa0) were performed according to IR,~D and Mu~NCI4 (20) , based on ten-fold titrations in mice.
Virus-multiplication has been tested in BHK21/c t3, IteLa Bristol, BSC-1, RK-13 and Vero celt lines and also in primary GMK cells.
BI-IK21/c I3 cells were grown in the medium described by MAePHE~aSOS" and STO~E~ (it). All other cultures were grown in Eagle's medium with 2 per cent inactivated calf serum. Tubes were seeded with 105 cells in 1 ml medium. Cultures were used when confluent monolayers were present. Cells were washed with saline, virus was diluted ~enfold from 10 -1 to t0 -6 in the medium, A volume of 0.2 ml of each dilution was inoculated into three tubes and allowed to adsorb for 1 hour at room temperature before washing with saline and addition, of new medium, Culture tubes were incubated for 8 days at 37 ° C and inspected daily for a Cytopathie effect (CPE). After 8 days culture fluids were inoculated i.c. in baby mice litters, 0.0t ml per mouse. Cell controls were treated in parallel.
BHK21/e 13 cultures in Roux bottles were established by seeding 6 × 10a cells/ml. Efforts to plaque the isolates were performed in BItK 21/c 13 cells under carboxymethylcellulose overlay by a method recently described (22) .
One to three hundred times concentration of cell culture fluids were performed by precipitation with polyethylene glycol (PEG) 6000 and NaC1 as described by McS~ARRY and BENZINGEE (14) . In some instances the precipitates were sonicated at 100W for 3 × 15 seconds in a Branson B 12 sonifier,
Sensitivity to sodium-desoxycholate (SI)C) was determined in baby mouse litters by the method of T~EI~a (25) .
For haemagglutination (HA), Immunoelectroosmophoresis (IEOP) and closed hexagon immundiffusion (CHI) experiments, the following antigen preparations have been employed : a) Crude suckling mouse brain (SMB) preparations: a 20 per cent infected suspension in APBS which has been centrifuged for 15 minutes at 10,000 rpm. b) Sucrose-aceton (SA) extracted infected mouse brains: Prepared according to C~AR~:E and CASALS (4), but omitting the final lyophilization step. c) Cell culture antigens: Culture fluids from infected BHK21/c13 Roux bottle cultures were precipitated with 6 per cent PEG 6000 (Macrogolum, Norsk Medisinaldepot) and 2.2 per cent NaCI (14) . The 100--300 times concentrated virus-precipitates were sonieated.
Antibodies were produced in adult, white mice inoculated with infectious suckling mouse brain preparations.
Initially each mouse received a virus dose of approximately 2.5 x 104-5 BMLD~0 in 0.25 ml brain suspension mixed thoroughly with 0.25 mI Freund's complete adjuvant (Difco). Subsequently 3 weekly injections with approximately 5× 104-5 BMLD~0 in 0.5 ml suspension were given. The last injection, identical to the initial one, was performed one week thereafter. Another 7--8 days later paraeentesis and bleeding from the retro-orbital sinus were performed. Arltibody preparations for Tahyna virus were produced in the same way.
A fowl antiserum to avian infectious bronchitis virus (AIB) with a NI of 6.1, was kindly provided by the Institute of Veterinary Medicine, Oslo. Reference mouse antibody preparations to TBE, Tribec, EEE, WEE and Uukuniemi (S 23) were supplied by the Yale Arbovirus Research Urlit.
An ornithological expedition to tternyken, Rost (approx. 67 ° 28°N, 12°E (Fig. 1) , collected 19 seabird sera in early May 1975. The composition of this material is cited in Table 6 .
Serological Methods tIaemagglutination (HA) and haemagglutination inhibition (HAI) tests were performed according to CLARKE and CASALS (4), modified for microtitration equipment (Cooke Eng. Co.). HA activity was tested within the pH range 5.6--7.2 at 4 °, 22 ° and 37 ° C. Erythrocytes from a variety of species have been investigated. These results will be reported separately, and all HA and HAI titers in the present paper refer to the employment of 0.5 per cent chicken erythrocytes.
(IEOP) Twelve ml gel was poured onto precoated 8 × 8 cm lantern slides. For antigen detection I per cent agarose (L'Industr. Biol. Franc.) and for antibody detection a mixture of 0.6 per cent agarose and 0.4~ per cent Difco Baeto agar was used (29) . Three rows of paired wells with 3 mm diameter were punched out. The well interdistance was 3 ram. In the I-lepascreen eleetrophoresis apparatus (Spectra Biologicals, Oxnard, Calif., 9303) 2 slides were run simultaneously for I--2 hours. In screenings, this allows testing of 58 unknown samples in one run.
Gel Precipitation This was performed by a sensitive modification of the micro Ouehterlony technique termed closed hexagon immunodiffusion (CHI) (27) , and also with varying patterns on lantern slides.
To obtain the maximum number and intensity of the precipitation lines, the antigens were applieated several times (2--5) during the 16 24 hours prior to antiserum application. Due to the lower diffusion rate of antigens, an interval of 2--4 hours between the last antigen and the antiserum application was adopted. The first lines then could be seen 4--5 hours after antiserum application. The gels were incubated at room temperature overnight, and then several days at -~4 ° C before staining and final reading.
IEOP and CHI gels were stained by 4 per cent tannic acid as previously described (8) .
Negative Contrast EIV[ Infected culture fluids from BgK21/cl3 cells grown in Roux bottles were concentrated 300 times by PEG6000/NaC1 precipitation (14) , resuspended in borate saline pit 9 or PBS ptI 7.4 and sonicated. Then 0.3 ml suspension was diluted 1/10 in PBS and centrifuged for 1 hour in a SorvM1 RC 2-B with rotor 55--34 at 20,000 rpm. The pellet was suspended in a few drops of distilled water and negatively stained with 2 per cent phosphotungstie acid pI-I 7.0 or 0.5 per cent uranyloxalate pI-I 6.0. One drop virus-stain mixture was placed on a formvar-earbon coated grid and excess fluid was withdrawn with filterpaper. The grid was examined in a JEM 100B Electron Miscroseope at a magnification of 50,000 ×.
Thin Section EM Infected and control BHK 21/c 13 cultures were harvested 4 days p.i. by means of a rubber policeman. The ceils were fixed in 3 per cent glutaraldehyde for 1 hour at -~ 4 ° C, washed three times in eacodylate buffer, postfixed in 1 per cent osmiumtetroxyde for 1 hour at room temperature and centrifuged for 3 minutes at 2500 rpm. The cell pellet was resuspended in a few drops of caeodylate buffer and centrifuged in micro capillary tubes in a hematocrit eentrifuge for 3 minutes at 12,500 rpm. The pellets were removed from the capillary tubes, dehydrated in acetone and embedded in Spurt by a rapid procedure (6) . Ultrathin sections (silver) were cut on a U 2-Reichert Ultramierotome and doubly stained by saturated uranylacetate in 50 per cent ethanol and Reynolds lead citrate (21) . Electron microscopy was carried out in a JEM 100B at a magnification of 50,000 ×.
Two pools, both processed from 5 unengorged female I. urine, termed E81 and E85 were inoculated into mouse litters 2 and 3 days old respectively. The mice seemed unaffected during the 3 weeks observation period. However, in the first passage (M1) the mice were moribund after 1.4 days for E81 and 13 days for E85. In the second passage (M2), incubation periods were reduced to i0 and 9 days respectively and in the subsequent passages to 5--7 days for both strains. The viruses were reisolated twice from the original tick pools during the following year. These results are summarized in Table 2 . The original tick pools were diluted 1/10 in BHK-medimn and inoculated on five BHK21/c 13 tube cultures. The tubes were harvested after five days. No CPE was recorded. The culture fluids were incubated into three baby mouse litters each. The mice were moribund after 8--10 days. The viruses were shown to be serologically identical to the strains isolated in mice by HAI, CHI and IEOP.
The Ru E8t tick pool was diluted 1/100 in BHK-medium, and three BHK cultures in Roux bottles were infected. Tile cultures were harvested after four days. Infected culture fluid was inoculated i.c. into 5 baby mouse litters, which were paralysed at day 10 p.i. The rest of the culture fluids were concentrated 100 times by PEG 6000/NaCI, and used as antigen. The isolates were shown to be serologiealty identical, and also identical to the mouse-passaged strains by HAI, CHI and IEOP.
A 20 per cent brain suspension from the third suckling mouse brain passage (M3) titered 105-s BMLDs0/ml in 1 day old mice. The virus suspension was titrated in parallel in 7 and 1.4 days old mice. As shown in Table 3, 1 week old mice were nearly as sensitive as the newborn, while there was a drop in infectivity of 1.2 logs and a prolongation of average survival time from 6.3 to 10.5 days for 2 week old mice. During production of antibody preparations it was demon-strafed that adult mice were refractary to intraperitoneally injections with both E81 and E85. In HeLa Bristol and RK-13 cultures no signs of virus multiplication were detected during primary inoculation and 3 consecutive blind passages. No CPE was present, and culture media concentrated 1--300 times by PEG 6000/NaC1 precipitation did not affect 1--3 days old suckling mice. As mentioned before, virus multiplication but no CPE was recorded after infection of BHK 21/c13 cultures with the original tick pools.
Plaquing in BHK 21/e 13 cells was not successful by the method used.
Both I~unde ES1 and E85 virus demonstrated a very marked sensitivity to treatment with sodiumdesoxycholate. The titers were reduced from 5.5 to 3.4 in the case of Ru E81 and 5.8 to 3.6 in the case of l~u E36.
l~unde virus produces antigens in suckling mouse brains as well as in BHK 21 l c 13. The antigens demonstrate specific reactions with mouse antisera and immune aseitie fluids in HAI, CHI and IEOP. The antigenic preparations have been kept at, --20 ° C for 11 months without loss in reactivity, Itaemagglutinating (HA) antigens can be produced ~om infected SMB by the SA extraction method of CL~R~:E and CASALS (4), hut Mso suspensions of crude infected SMB and PEG 6000/NaC1 precipitated cell culture fluids contain haemagglutinating activity. The ability to agglutinate chicken erythrocytes is relatively independent of pit and temperature. The highest titers and most clearcut endpoints have however been attained by pit 6.4--6.8 at 4 ° C.
In HAI no serologic differences between RUE81 and E85 could be demonstrated. The mouse antisera and immune ascitie fluids were inhibitory to 4I-IA units of antigen to dilutions of 1/80--640. Normal mouse sera were not inhibitory, neither was a fowl antiserum to AIB virus or antisera to TBE, Tribec, EEE, WEE and Tahyna.
Precipitating antigens, as revealed by CHI and I E O P , also were present in the same preparations as cited above. I n CItI 3 --5 specific lines have been noted (Fig. 2) , while in I E O P double lines were seen. No antigenic difference between the two strains was detected. Virus from mouse brains and cell culture demonstrated total i d e n t i t y b y these methods. No reaction was observed with a n antiserum aga, inst avian infectious bronchitis virus.
The 19 seabird sera were screened by I E O P , and the specificity secured by CHI. Foul" out of 19 birds had precipitating antibodies to R u E 8 t . The composition of the material and the results are sho~nt in Table 4 .
Negative Contrast EM A series of electron micrographs of negatively stained preparations are presented in Figs. 3, 4. Fig. 3 shows corona-virus-like structures in a preparation from l~u E 81 infected cell culture supernate. Similar structures were never seen in uninfected controls. Most of the particles have an approximate total diameter of 170 nm, but in other preparations particles with diameters varying from l l 0 to 220 n m have been seen. Fig. 4 A demonstrates an intuct particle at a greater magnification. I t seems to contain an inner structure 80 nm in diameter. Fig. 4 B and 4C demonstrate partly and completely disrupted particles with loss of knobs and release of inner helical material. Staining with 0.5 per cent uranyloxalate (Fig. 4D ) produces a sort of 3-dimensional picture of the virus-like particle.
Thin Section EM Thin sections of R u ES1 infected B H K 21/e13 cultures 4 days p.i., demonstrate virus-like particles in the cytoplasm and extraeellularly between broken and intact cells (Fig. 5) . The particles have a diameter of 100--110 nm. Some of them look empty, while others are penetrated by stain and expose inner structures. The particles seem to have a double outer membrane, but no knobs are visible by thin sections and positive staining.
When the Runde virus strairts were first isolated, it had to be kept in mind that they might originate from the staff handling the tick pools, or from the baby mice used for isolation. The reisolations from the original tick suspensions (Table 2) , a~d the properties differing from known human, rodent and also avian eoronaviruses (2, 19) , could not entirely rule out these possibilities. But the fact that antibodies to the virus were demonstrated irt seabirds and the isolation of virus also in cell cultures seem to settle this question. Circumstances of tickcollection strongly support that Runde virus is an arbovirus in the ecological sense, and not a mechanical pick-up :
1. I t is highly unlike that a relatively labile virus should survive for 3 weeks in I. uriae without active multiplication.
2. The lack of engorgement or pregnancy of the female I. uriae, and the fact that they were in diapause and surrounded by desiccated skinlaps, strongly indicate that t~unde virus m a y have passed interstadially in the virus-carrying ticks.
Still, the final inclusion of t~unde virus amortg the arboviruses must await the demonstratiorl of true biological transmission by .I. uriae.
Although the morphology of gunde virus strikingly resembles that of the eoronavirus group, some characteristics of the virus differ considerably from 3* those of known coronaviruses. This holds true for the pathogenicity to mice (2, 13) , the ability to grow in cell cultures (1, 2, 5) and the haemagglutinating properties (2, 7) . Besides, the size of the virion seems larger than for other members of the coronavirus group (19) .
The discrepancy in size between negative contrast and thin sections may be partly due to shrinkage during fixation and dehydration prior to embedding and thin sectioning, and partly to collapse of the particles by negative staining adding to their actual size. The most reliable staining procedure for size calculations probably is negative staining by uranyloxalate which also stabilizes the structure of the particles.
Detailed taxonomic considerations for Runde virus must await further characterization of biological, serological and molecular properties. But Runde virus is probably a "new" coronavirus on the grounds of: its morphology, the lack of cross reaction with avian infectious bronchitis virus and the lack of serological reactions with control mouse sera probably eliminating the possibility of MI-IV isolates. Also, our Department of experimental animals have no indications of mouse hepatitis in the mouse colony which have been used in these studies.
The demonstration of this hitherto unrecognized virus associated with seabirds and ticks also necessitates further efforts to clarify-the ecological and possible epizootologieal implications. As alre~ody mentioned, an unusually high chick mortality ill the seabird colonies at Runde has been reported during the Ias~ few years. Bacteriological and toxocological research has not been able to reveal any definite etiological factor, although pesticide residues have been demonstrated to some extent. This must of course bring viruses into consideration. A synergistic action between pesticides and virus seems one possible working hypothesis in this connection. Since Norwegian seabird colonies in general are rather frequently visited by scientists, students, holiday travellers and the local public, the recent documentation of the ability of I. uriae to attack man (18) raises the question of public health implications. Isolations from I. urine of Uukuniemi viruses (30, 31, 33) orbiviruses (12, 32) and flaviviruses (9, 33) ill addition to Runde virus further stresses this point.
At present nothing is known concerning the distribution and general infection rates of Runde virus. Some indications that it might be widespread in Norwegian seabird eolonies can be given, however. Firstly, the isolation of two virus strains from only 206 I. urine collected at Runde indicates a rather high infection rate of the vectors in this location. Secondly, the demonstration of a high rate of seropositive birds at tlernyken, Rest, proves that virus-circulation is not restricted to Runde.
Further field work and experiments which may throw more light upon th(~ ecological interrelationships of Runde virus as well as upon the characteristics of the virion are in progress.
We are indebted to Eva Mood Petterson, Einar Brunvold and Hatlgrim Sog£rd for their skilfuI technical assistance, and to Tone Lea for her valuable help with the preparation of figures.
|
