key: cord-265201-ab67pnct
authors: Sugiyama, K.; Amano, Y.
title: Hemagglutination and structural polypeptides of a new coronavirus associated with diarrhea in infant mice
date: 1980
journal: Arch Virol
DOI: 10.1007/bf01314978
sha: 
doc_id: 265201
cord_uid: ab67pnct

The hemagglutination (HA) and receptor destroying enzyme (RDE) activities of a newly isolated mouse enteric coronavirus (designated as DVIM) are described. DVIM agglutinates mouse or rat red blood cells (RBC) at 4° C. At 37° C the agglutination was rapidly reversed. The optimal pH for HA and for RDE activities using mouse red cells were shown to be 6.5 and 7.3 respectively. Hemagglutination by DVIM was not inhibited by pretreatment of RBCs withVibrio cholerae filtrate or by pretreatment with Influenza-A neuraminidase. Therefore, the DVIM receptors on RBCs differ from the receptors of Influenza-A, and the RDE activity of DVIM acts specifically on this receptor. In addition, an analysis of the DVIM polypeptides showed that the virions contain five major, VP1 (M.W. 139,000), VP2 (68,000), VP3 (53,000), VP4 (38,000), VP5 (22,000) and two minor, VP1a (110,000), VP1b (100,000) polypeptides. VP1 and VP1b were digested by bromelain, suggesting that they constitute the surface glycoproteins.

Coronaviridae are a family of lipid-containing RNA viruses which exhibit a unique morphology. The virions have widely-spaced surface projections which form a radiating "corona" around the particles as shown by electron microscopy (18, 26) . The diarrhea virus of infant mice (DVIM), a newly-isolated mouse enteric coronavirus, is antigenically related to other coronaviruses by complement fixation assay but clearly distinguishable by neutralization assay (21) .

The existence of a receptor destroying enzyme (RDE)-like activity has not been reported for any of the coronaviruses. We report here a RDE activity associated with DVIM virions which differs from that of ortho-or paramyxo-7* 0304-8608/80/0066/0095/$ 02.20 viruses. T h e s t r u c t u r a l p o l y p e p t i d e s of D V I M were a n a l y s e d b y p o l y a e r y l a m i d e gel eleetrophoresis.

The mouse enteric coronavirus (DVIM) was kindly provided by Dr. Kozaburo Sate (Central Laboratory of Shionogi Pharm. Co., Osaka, Japan). D V I M was passaged in BALB/c-3 T 3 cell cultures in Eagle's minimal essential medium containing 10 percent fetal calf serum. Virus inoculation was carried out with 107 TCID~0 per l~oux bottle. A t 24 hours post-infection virus was harvested by three freeze.thaw cycles and stored at --70 ° C. Influenza virus strain A/RI/5, A/Victoria/3/75, A / N J / 8 / 7 6 and Hemagglutinating virus of J a p a n ( H V J or Sendal virus) strain Z were grown in the allantoic cavity of 10-day-old embryonated ehieken eggs as previously described (24) . The same purification processes were employed for each virus at 4 ° C. Alter clarification of culture fluid b y centrifugation at 8,600 × g for 30 minutes, virions were concentrated by eentrifugation at 70,000 × g for 90 minutes, resuspended in 0.1 M N T E buffer (0.05 ~ Tris-C1, 0.I ~c NaC1, 0.001 M EDTA), p i t 7.3, layered onto a discontinuous gradient of sucrose (15:30:50 percent w/w), and centrifuged at 64,000 × g for 2 hours. Virions were collected from the interphase between 30 and 50 percent sucrose solutions and were dialysed against N T E buffer. For polypeptide analysis, virions were repurified by linear sucrose gradient (20 to 50 pereent w/w) eentrifugation at 77,000 × g for 90 minutes.

Routine H A tests were carried out as previously described (23), using 0.5 percent mouse RBCs in Dulbeeeo's PBS, containing 0.3 percent bovine serum albumin. Alternatively, 0.25 percent avian RBCs, or 0.5 percent other m a m m a l i a n RBCs were used. The titer was recorded as the reciprocal of the highest virus dilution causing a detectable HA.

Hemadsorption tests were done on DVIM-infected B A L B / c -3 T 3 cell monolayers grown on glass cover slips. A 0.5 percent suspension of mouse RBCs were adsorbed for 3 hours at 4 ° C. Cover slips were washed and prepared for analysis by light microscopy. The experiment for elution kinetics was performed on identical eell-monolayers at various incubation periods. To test for R D E -l i k e activity, cells were incubated at indicated temperatures, prepared for microscopy and photographed. The elution ratio was expressed as the percentage of remaining RBCs enumerated per photographic field.

Sodium dodecyl sulfate (SDS)-PAGE was performed on 7.5 percent polyacrylamide gels by the method of MAIZE~ et al. (17) , using 6 m m diameter tubular gels 60 m m long.

Purified virus was solubilized in 1 percent SDS, 1 percent 2-mereaptoethanol, 20 percent glycerol, and 0.006 percent bromphenol blue, for t.5 minutes at t00 ° C. After preeleetrophoresis for one hour at 5 mA/gel, samples were applied and eleetrophoresis was carried out at 5 mA/gel in 0.i ~ sodium phosphate buffer, p H 7.2, for 6 hours, at room temperature.

Gels were fixed overnight with 20 percent sulfosalieylie acid and stained with 0.25 percent Coomassie brilliant blue in 50 percent methanol. The gels were destained with several changes of 15 percent methanol in 7 percent acetic acid. Gels were stained for carbohydrate and lipid with Schiff's reagent and oiLred-O, respectively, following the methods of HIEtgHOLZEP~ et al. (8) . Stained gels were scanned with a densitometer (Joko gel scanner) at 600 n m for peptides, 550 n m for carbohydrates, and 450 n m for lipids, respeetively.

The approximate molecular weights of the polypeptides were determined by the m e t h o d of St~APIaO et al. (22) . Percentage composition of each strueturM polypeptide was determined from the photometric scans using a Joko scanning planimeter.

R e d blood cells of seven different species were tested for agglutinability with DVIM, and the results are summarized in Table 1 . Only mouse and r a t I~BCs were a g g l u t i n a t e d at 4 ° C, indicating a restricted receptor range for the H A activity.

Mouse I~BCs agglutinated by D V I M at 4 ° C, could be liberated from agglutination by incubation a t 20 ° C ( Table 2 ). The D V I M sample had a titer of 4,096 H A U at 4 ° C, which decreased to 8 H A U after 60 minutes of incubation at 20 ° C. This indicates t h a t in addition to its H A activity, ;[)VIM virions appear to possess a l~DE-like a c t i v i t y which differs from the I~DE a c t i v i t y shown for p a r a m y x oand myxoviruses, which is n o t active at 20 ° C. 

Avian red cells were prepared as 0.25 percent suspensioss in PBS (pH 7.3), mammalian red cells were prepared as 0.5 percent in the same PBS Hemagglutination test were carried out at 4 ° C as usual, then the microplate was incubated at 20 ° C, for 60 minutes

The H A titer was stable at 4 ° C. However, H A appeared to be reversable only under some p H conditions, when incubated at 37 ° C, as shown in Table 3 . The optimal p H for H A a c t i v i t y is relatively wide, whereas R D E -l i k e a c t i v i t y appeared optimal at a p H of about 7.4 and exhibited no a c t i v i t y at p H 6.5, at 37 ° C. Tile H A a c t i v i t y was stable at 37 ° C, up to 24 hours, b u t was i n a c t i v a t e d at 56 ° C, within 20 minutes (data n o t shown).

K. SUGI¥~A and Y. AI,~ANO: Table 4 . Bromelain, a proteolytic enzyme which removes t h e club-shaped projections from the surface of other coronaviruses (4% 8, 25), affected both H A a c t i v i t y and infectivity of DVIM. Trypsin did not affect H A a c t i v i t y or infectivity, while pepsin affected only H A activity. The H A titer was reduced slightly by NP-40 t r e a t m e n t , and infectivity was destroyed. E t h y l ether destroyed both virus H A a c t i v i t y and infectivity. HA-titrations were carried out with 0.5 percent mouse red blood cells in indicated pH, at 4 ° C, thereafter the microplate was incubated at 37 ° C for 60 minutes 

The d a t a in Tables 2 and 3 The infected cells exhibit virus-specific enzyme aeti~dties associated with these syncytia. Fig. 1 shows the elution kinetics of red blood cells from the syncytia, induced by DVIM 10 hours after infection, at 20 ° and 37°C respectively. Approximately 70 to 80 per cent of RBCs were liberated within 30 minutes at 37 ° C while the liberation at 20°C progressed slowly. Therefore, the cell surfaces of DVIM-induced syncytia exhibit the same H A and RDE-like activities associated with virions. If liberated I~BCs are reexposed to DVIM-infeeted monolayers, no hemadsorption is observed. The same syneytia could again hemadsorb freshly prepared RBCs (Fig. 2) . This implies that hemadsorption of DVIM does not correspond to "pseudo-hemadsorption" described by BUCKNALL et al. (3) . Furthermore, the elution of red blood cells from syneytia appears to be a result of enzymatic digestion of RBC receptors by a gDE-like activity situated on the syneytia surface. This activity was dependent upon incubation temperature, similar to RDE-like activity of DVIM virions.

To confirm the receptor destroying activity of DVIM, the effect of the purified virus on red blood cells was examined. Mouse RBCs treated with DVIM were reagglutinated b y Influenza-A viruses or HVJ, but could not be reagglutinated by ])VIM virions (Table 5) . I n contrast, R D E treated mouse RBCs could still be fully agglutinated by DVIM (Table 6 ). Influenza-A hemagglutination of I~BCs was reduced in proportion to the concentration of RDE. i 0 0 K. SUGIYAMA a n d Y. AMANO: ~fouse r e d blood cells were t r e a t e d w i t h R D E a t indicated concentration, dissolved in 0.01 M Ca-Borate buffer (pH 7.2), at 37 ° C for 60 minutes, after which t h e cells were w a s h e d a n d resuspended as a 0.5 p e r c e n t suspensions i~ P B S a n d used for H A titration. I I D E was o b t a i n e d from Vibrio eholerae v a r i a n t of 558 s t r a i n filtrate (Takeda P h a r m . Co. Osaka, J a p a n ) . Control t~BCs were n o t t r e a t e d with t h e R D E Fig. 2 . R e a d s o r p t i o n of freshly p r e p a r e d I~BC to pre~dously exposed syncytia. H e madsorbed s y n e y t i u m which were i n c u b a t e d at 37 ° C to release I~BCs for 60 minutes, were exposed to freshly p r e p a r e d RBCs a n d p h o t o g r a p h e d Above results strongly suggest that DVIM possesses an enzymatic activity which reacts with surface components of RBCs. However, this enzymatic action appears to differ from the bacterial neuraminidase and that of Influenza-A virions. 

Since DVIM is a newly isolated coronavirus, its structural polypeptides have not been characterized. Therefore we analysed the virion polypeptides by SDS-PAGE. Electrophorescd polypeptides, when stained with Coomassie brilliant blue, revealed a mimmum of seven bands. A densitometer scan of stained polypeptides of DVIM and HVJ are shown in Figs. 3 A and 3 B, respectively. The five major polypeptides of DVIM were designated as VP 1 to VP5 in decreasing order of electrophoretic mobility, as shown in Fig. 3A . Two additional minor bands V P l a and V P l b were observed between VP1 and VP2. The approximate molecular weight and molar ratios of the structural polypeptidcs, are presented in Table 7 . Preliminary analyses indicated that three polypeptides, VP 1, VP 1 b and VP2 are glycoproteins, while VP2 reacted with oil-red-0, suggesting it may be a lipid-containing protein. 

A densitometer scan of a stained polyacrylamide gel of the polypeptides of virus particles treated with bromelain (1.0 mg/ml at 37°C for 15 minutes) is shown in Fig. 4 . As can be seen from this figure, VP 1 and VP i b were digested by the bromelain treatment, suggesting that these polypeptides are found on the surface of the virion. Polypeptides V P l a and VP2 were also reduced, and only trace amounts of them were still observed, suggesting these polypeptides may also be surface polypeptides. VP3 was unaffected by bromelain treatment, suggesting that it represents an "inner core" protein.

Direct HA (i. e. without pretreatment of the virus) and hemadsorption have been described for several eoronaviruses, 0C-43 (11, 12) , HEV (6, 20) , IBV (1), and bovine coronavirus (2) . The hemagglutinating ability of 0C-43 required several passages in suckling mouse brain (12) , and Massachusetts strain of IBV (1) required both sucrose gradient purification and digestion with phospholipase C for activation. The HA of DVIM, reported here, was detectable directly in the culture fluid. The ratio of infectivity to HA varied between 1.2 and 3.1 × 104 TCID50/ 0.1 ml/HA unit. DVIM hemagglutinated the RBCs of only two species tested, while other coronaviruses, 0C-43, HEV and IBV, were less specific. Accordingly, HA of DVIM was expressed only using two types of red blood cell species, mouse and rat, and was also detectable at low temperatures in a stable pattern.

The formation of "prozone" described by BING~AM et al. (1) was not observed for DVIM. The instability of HA and hemadsorption at 20 ° C suggests that a viral enzyme may be responsible for uncoupling the virus from the red blood cell receptor. The possible enzymatic activity of DVIM was demonstrated on homologous virus receptors of mouse RBCs without any effect on receptors for heterologous viruses, Influenza-A and HVJ. Furthermore, it was also shown that bacterial neuraminidase (RDE) had no effect at all on the receptor for DVIM; nevertheless, the receptors for influenza-A viruses were destroyed by the same treatment. Additionally, even after purification through a sucrose density gradient, DVIM retains its receptor-destroying activity, thereby showing that this activity is viral-specific and associated with virions. Our preliminary results strongly suggest the existence of some type of receptor-destroying enzyme such as that of Influenza C virus (10, 13).

The optimal pH of DVIM HA activity was distinct from that of the receptordestroying activity. It has not been determined whether these activities may reside on a single polypeptide or on two different polypeptides. Because the surface projections of DVIM are somewhat different morphologically compared to other coronaviruses, it will be of interest to study the exact nature of the hemagglutinating and receptor-destroying activity of DVIM. The receptor-destroying activity of DVIM appears to be unique, when compared to that of ortho-and para-myxoviruses.

Purified DVIM contains seven species of polypeptides, five major and possibly two minor polypeptides. These results compare favorably to those obtained for human coronavirus OC 43 (8), IBV (14) , bovine coronavirus LY-138 (7) and other murine eoronaviruses ?¢II-IV-3 (16) and J H M (27) . The classification of the polypeptides of D V I M as glyco-or glycolipoprotein m u s t be considered as tentative at this point since the chemical nature of these complexes cannot be precisely defined by differential staining techniques. The number and size of gtyeoproteins of eoronaviruses which have been reported previously b y m a n y investigators, were shown to vary. I t appears t h a t our V P I (MW 139,000) corresponds to a large polypeptide (MW 150,000--200,000) reported for other coronaviruses (4, 8, 9, 15, 16, 19, 24, 25, 27) , because it is glyeosylated and appears to be situated on the surface of the virion. Other D V I M virion polypeptides also appear to be surface glyeoproteins. Furthermore, some of the investigators agree t h a t coronaviruses contain a polypeptide having an a p p r o x i m a t e molecular weight of 50,000, corresponding to our VP3, which is not glyeosylated and tends to comprise a large proportion of the virus protein (4, 15, 19, 25) . Polypeptide VP3 represents 26 percent of virus protein and is not affected b y bromelain digestion at all. This suggests t h a t V P 3 is also located in the "inner core" (5, 19, 25) of the virion. Although a considerable a m o u n t of V P 4 was not observed in Fig. 4 , the possibilityt h a t the V P 4 could be an artifact produced b y different conditions of reduction (24) , still remains unexplained.

I-Iemag'glutinat%n by avian infectious bronchitis v i r u s -°a eoronavirus

Replication of an enter'it bovine coronavirus in intestinal organ cultul°eS

Intraeellular development and mechanism of hemadsorption of human eoronavimls

The polypeptide structure of transmissible gastroenteritis virus

Isolation of subviral components from transmissible gastroenteritis virus

A hemaggIutinating virus producing encephMomyelitis in baby pigs

Structural polypeptides of the enteropathogenic bovine eoronavirus strain LY-138

Protein composition of coronavirus 0C43

Purification and biophysical properties of human coronavirus 229E

The relationships of the receptors of a new strain of virus to those of the mumpus-NDV-influenza group

Hemadsorption by coronavirus strain OC-43

Some characteristics of hemagglutination of certain strains of "IBV-like" virus

A comparison of "influenza C" with prototype myxoviruses

Receptor-destroying activity (neuraminidase) and structural polypeptides

The polypeptide compositiort of avian infectious bronchitis virus particles

Polypeptides of the surface projections and the ribonucleoprotein of avian infectious bronchitis virus

The polypeptides of human and mouse eoronavirusos

Acrylamide-gel electrophorograms by mechanical fractionation

Coronavirus ; A comparative review

The polypeptides of hemagglutinating encephalomyelitis virus and isolated subviral particles

Characteristics of coronavirus causing vomition and wasting in pigs

Some characteristics of corona-like virus isolated from infant mice with diarrhea and inflamentory submaxillary gland of rats

Molecular weight estimation of polypeptide chain by etectrophoresis in SDS-polyacrylamide gets

Isolation and characterization of two distinct types of I-IVJ (Sendai virus) spikes

Characterization of a coronavirus. I. Structural proteins: effects of preparative conditions on the migration of protein in polyacrylamide gels

Characterization of a coronavirus. 2. Glycoproteins of the viral envelope; Tryptie peptide analysis

Structural polypeptides of the marine eoronavirus JHM

Received February 4, 1980