key: cord-355512-tycuoslv
authors: Storz, J.; Zhang, X. M.; Rott, R.
title: Comparison of hemagglutinating, receptor-destroying, and acetylesterase activities of avirulent and virulent bovine coronavirus strains
date: 1992
journal: Arch Virol
DOI: 10.1007/bf01309637
sha: 
doc_id: 355512
cord_uid: tycuoslv

Hemagglutinating and acetylesterase functions as well as the 124 kDa glycoprotein were present in the highly cell-culture adapted, avirulent bovine coronavirus strain BCV-L9, in the Norden vaccine strain derived from it, and in 5 wild-type, virulent strains that multiplied in HRT-18 cells but were restricted in several types of cultured bovine cells. The BCV-L9 and the wild-type strain BCV-LY-138 agglutinated chicken and mouse erythrocytes. The acetylesterase facilitated break-down of the BCV-erythrocyte complex with chicken but only to a minimal extent with mouse erythrocytes in the receptor-destroying enzyme test. Purified preparations of the vaccine and the wild-type strains agglutinated chicken erythrocytes at low titers and mouse erythrocytes at 128 to 256 times higher titers whereas receptor destroying enzyme activity was detectable only with chicken erythrocytes. When wild-type strains were propagated in HRT cells at low passage levels, they produced 5×10(5) to 4.5×10(6) plaque forming units per 50 µl which agglutinated erythrocytes from mice but not from chickens. Diisopropylfluoro-phosphate moderately increased the hemagglutination titers, but completely inhibited the receptor destroying enzyme of purified virus of all strains. It had virtually no influence on the plaque-forming infectivity of the different BCV strains. The acetylesterase of strain BCV-L9 reacting in the receptor-destroying enzyme test was stable for 3h at 37 and 42°C. It was inactivated within 30 min at 56°C while the hemagglutinin function of this strain was stable for 3 h at 37, 42, and 56°C, but it was inactivated at 65°C within 1 h.

Bovine coronavirus (BCV) has a hemagglutinin (HA) for some erythrocytes with an approximate molecular mass of 62 kDa and 124 kDa in the reduced and the nonreduced forms. This structural protein composes the short spikes of the viral envelope [4, 9, 1 I] . Acetylesterase activity (AE) was found associated with this glycoprotein of a BCV strain isolated from calves in the Netherlands and of the strain BCV-L9 [15, 16, 28] . This glycoprotein is now referred to as the hemagglutinin-esterase (HE) of BCV [2] . The AE inactivates the receptors for BCV of susceptible cells by hydrolyzing an ester bond to liberate acetate from C-9 of sialic acid [15, 16, , an enzyme function first detected in the influenza C virus by Herrler and coworkers [5,1. The gene of the HE glycoprotein of BCV is located upstream of the S gene and predicts a protein with 424 amino acids [7, 12, 33,1. As an enteropathogen BCV causes severe diarrhea in neonatal calves, and it is also considered to be etiologically involved with winter dysentery of adult cattle [4, 11, 14] . BCV represents one of the better characterized coronaviruses with HA properties. Four major structural proteins are associated with infectious BCV. The HE as well as the spike or S glycoprotein (190kDa) and the integral membrane glycoprotein M (23) (24) (25) (26) are associated with the viral envelope while the phosphorylated N protein (50-54kDa) functions as a nucleocapsid [3, 8, 20, 23] . Proteolytic cleavage of the S glycoprotein precursor into S 1 and S 2 of 100 and 110 kDa is required for cell fusion activity [20, 22,1. The S 1/$2 glycoproteins facilitate virus attachment to susceptible cells, and also binding to erythrocytes, cell fusion, and induction of neutralizing antibodies [t7, 20, 22, 24, 25,1. The exact functions of HE and S 1/$2 and their interplay in the infectious process in vitro and in vivo are not fully defined [17, 24] .

The presence and function of the HE was analyzed in two cell-adapted BCV strains [15, 24, 28, 29] . Some of the BCV isolates from winter dysentery agglutinated rodent erythrocytes and others did not [1,1. The objective of our investigations was to identify the HE of wild-type BCV strains at low passage levels in H R T cell cultures and to compare it with the HE of prototype BCV-L9 and vaccine strains by relating its functions to the interaction with different erythrocytes, the effects of enzyme inhibitors, and to plaque-forming infectivity.

The cell culture-adapted prototype BCV-L9 was originally isolated in bovine fetal kidney (BFK) cells from diarrhea fluid of a calf [11, 18] . Our virus strain had been passaged 42 times in BFK cells, 16 times in bovine fetal brain cells, 15 times in bovine fetal spleen (BFS) cells, and 5 to 10 times in human rectal tumor (HRT-18) cells [20] [21] [22] . Five other wildtype BCV isolates, initially maintained by calf inoculation, were adapted to HRT-18 cells from diarrhea fluid or intestinal mucosal scrapings of calves with clinical diarrhea and electron microscopic evidence of coronavirus infection [4, 21] . These strains are: BCV-LY-138, BCV-C-50, BCV-Miller, BCV-Meeker, and BCV-Fisher. The vaccine strain of BCV was cultured in HRT-18 cells from the vaccine of Norden Laboratories, Omaha, Nebraska [11, 18] . Importantly, cultivation of the BCV wild-type strains remained impossible for us until it was demonstrated that the cytopathic expression of BCV-L9 in cultured bovine cells was enhanced by trypsin [22] and that HRT-18 cells were susceptible to BCV [10] . These wild-type strains were completely restricted in a variety of cultured bovine cells even in the presence of trypsin [21] .

The HRT-18 cells were cultured with Dulbecco's modified Earle's medium (DMEM) with 5% bovine fetal calf serum which was omitted from cultures used for virus propagation. A plaque assay in HRT-18 cells overlaid with 3ml of 1.2% (w/v) agarose in DMEM containing 4 gg/ml final trypsin concentration was used to measure infectivity [22, 27] . Samples of the 7 BCV strains were propagated in HRT cells and purified as described [6, 20, 30] .

The BCV-L9-infected cell sediment of the first step in the virus purification was suspended in 7.5ml phosphate buffered saline (pH of 7.4), treated with sound 3 times for 15s at a power setting of 4 (Branson Sonifyer), and centrifuged at 3000 x g for 30 min. The supernatant fluid represented the BCV-L9-infected cell lysate. Uninfected HRT-18 cells were treated the same way and were used in control tests.

Acetylesterase activity of purified BCV preparations was determined according to Vlasak et al. [28] . Purified preparations of the different BCV strains were added in 5 gl quantities to 1 ml of phosphate buffered safine containing I mM p-nitrophenylacetate (PNPA). The PNPA was initially dissolved in ethyl alcohol. Hydrolysis of the substrate was monitored at 405 nm in a Beckman spectrometer with multiple cuvette sets and a chart recorder. The reaction was measured at 1 min intervals for 5 min. The 5 rain value was used for comparisons [15, 16] .

The HA test was employing 0.5% suspensions of erythrocytes from adult chickens or mice [5] . The tests with 2-fold BCV dilutions in 50 gl quantities were initially incubated at 4 °C for 1 h to assess the HA titers. Thereafter, the microtiter plates were shifted to 37 °C for 1 h to monitor inactivation of receptors reflected by the breakdown of the BCV-erythrocyte complexes mediated by the AE in the RDE assay. The extent of disappearance of the BCVerythrocyte aggregates was recorded in a final evaluation made following an additional 4 to 6 h at room temperature when unequivocal interpretation of the results was possible [24] .

Diisopropylfluorophosphate (DFP) was used as an inhibitor of serine esterase activity to assess the effect of AE from different BCV strains on the HA and RDE functions as well as on BCV infectivity. A sample of 50 gl of each purified BCV strain was treated with 1 mM of DFP in 10 gl for 10 rain at room temperature. The virus samples were then diluted in 2 fold steps and tested for HA and RDE activities. Untreated samples were tested in parallel. Similarly, 10 gl aliquots of the purified BCV strains were treated with 1 mM of DFP in 10gl for 10min at room temperature. Samples were then mixed with 9801~1 of PBS, and further decimal dilutions were made. The infectivity of DFP-treated and untreated BCV preparations was assayed in the plaque test [22] . Bovine submaxillary muein (BSM) was tested with 2-fold dilutions of the purified virus preparations by adding 50 gl of the 1 mg/ml BSM. Alternatively, the indicated inhibitor concentrations was diluted 2-fold and 8 HA units of the strains BCV-L9 or BCV-LY-138 were added. Chicken erythrocytes as well as mouse erythrocytes were employed in these tests.

Aliquots of the BCV-L9-infected cell lysate having HA and RDE titers of 128 with chicken erythrocytes were incubated at 37 °C, 42 °C, 56 °C and 65 °C. Samples were withdrawn at intervals of 10 min for 1 h and every 30 min for the next 2 h to be tested for HA and RDE activities.

Sodium dodesyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) was performed in 12% slab gels under denaturing and under mildly denaturing conditions. Detailed procedures for the analysis of BCV proteins by SDS-PAGE were described [6, 20] .

The different viral functions were assessed on purified BCV samples containing infectivities of 2.5 x 10 6 t o 1.2 x 108 PFUs per 50 p.1 (Table 1) . AE activity was detected in the PNPA test and the gp 124HE structural protein was seen in PAGE of all BCV strains tested. The highest AE activity was detected in the BCV-LY-138 strain and this strain had also the highest infectivity even though it was only in the 5th passage in HRT-18 cells. Purified virus preparations of the strains BCV-L9 and BCV-LY-138 agglutinated chicken erythrocytes at titers of 1024 and 4096, respectively, but the HA titers were 8 and 4 fold higher with mouse erythrocytes. In contrast, the vaccine and three wild-type BCV strains had HA titers of 32 to 128 with chicken erythrocytes, and 128 to 256 fold higher titers were recorded with mouse erythrocytes. RDE activity was detected at titers of 16 to 64 with chicken erythrocytes whereby H A : RDE ratios of 1 : 1 were evident for the strains BCV-Meeker, BCV-Fisher and BCV-Calf-50. The RDE activity was minimal or not detectable in tests involving mouse erythrocytes. The PFUs per HA unit ranged from 2.6 × 10 4 to 1.5 x 10 6 for chicken erythrocytes and 3 x 10 2 to 7.3 x 103 for mouse erythrocytes (Table 1) .

The HA pattern of the highly cell-culture adapted and the wild-type BCV strains differed. The BCV-L9 strain was in the 78th cell culture passage while the wildtype strains had been passed only 3 to 7 times in HRT-18 cells. Strains BCV-L9 and BCV-LY-138 agglutinated chicken erythrocytes while the other BCV strains did not ( Table 2 ). All strains agglutinated mouse erythrocytes. The infectivity of the preparations ranged from 5 x 105 to 4.5 x 10 6 PFUs. RDE activity for receptors on chicken erythrocytes was detected in samples from BCV-L9 and BCV-LY-138. 

The BCV infectivity was virtually unaffected by D F P (Table 3 ). In contrast, the P D E activity for chicken erythrocytes was completely inhibited. The H A titers for chicken erythrocytes were 2 to 64 times higher after D F P treatment. The R D E activity for mouse erythrocytes was eliminated by D F P and an enhancing effect on H A was not noticed. BSM inhibited H A by BCV-L9 and the vaccine strain, while it reduced the H A titers of the other strains for chicken erythrocytes (Table4). Mouse erythrocytes became aggregated in the control tests by BSM at the concentration used.

The BCV-L9-infected cell lysate was used to assess thermal inactivation of the H A and R D E functions for chicken erythrocytes. As documented in Figs. 1 and 2 these activities remained unaffected at temperatures of 37 °C and 42 °C for 3 h. The R D E titer was reduced from 256 to < 2 within 30 min at 56 °C while H A remained constant. This activity was lost within 60 min at 65 °C.

All BCV strains tested had AE activities when purified preparations were assayed in the P N P A test [28] . This enzyme was inhibited by the serine esterase inhibitor D F P when assayed in the AE-mediated breakdown of BCV-chicken erythrocyte complexes yet the H A titers of purified preparations of wild-type BCV strains were increased 2 to 64 fold at 4 °C. The infectivity of identically DFP-treated Table 3 . (Table 1 ). These results indicate that the AE of the BCV is probably not directly involved in viral uptake during the infectious process. Receptor binding and viral attachment to susceptible cells in infections of non-HA and HA coronaviruses are mediated by the S glycoprotein [19] . The HE protein is not expressed in some murine coronavirus strains [31] . Our finding contrasts a previous report of a 100-fold reduction of infectivity by DFP of another BCV strain plaque-assayed in Madin-Darby bovine kidney cells [28] . Infectivity assayed by the plaque test does not address the potential significance of AE in viral spread within an infected cell culture or animal. The AE activity of the HE protein may play a role in facilitating virus release from infected cells and viral spread. BCV particles were seen adsorbed abundantly to the plasmalemma or microvilli. Adsorbed BCV covered the entire exposed surfaces of enterocytes in infections of calves [4] and HRT cells [I 3] . Further insight into the role of HE will be gained through investigations of the interactions of HE with cellular receptors in vitro as well as with the neuraminidate-containing glycokalyx protecting mucous membranes of the respiratory and intestinal tracts of hosts. The strongly inhibitory function of BSM on HA points in this direction (Table 4) . One difference between the BCV-L9 and BCV-LY-138 strains and the vaccine, Meeker, Fisher and Calf-50 strains involves their interaction with different erythrocytes. Chicken erythrocytes were agglutinated by BCV-L9 and BCV-LY-138 and the RDE functioned. These strains also agglutinated mouse erythrocytes, but elution through RDE activity was minimal. The approximately 100-fold concentrated purified preparations of the other BCV strains agglutinated chicken erythrocytes at low titers which were increased by DFP. The HA titers with mouse erythrocytes were 128 to 256 fold higher, but virtually no RDE function was detectable (Table 1) . This difference was further substantiated 

X v X "" X X in HA tests with fluids from HRT-18 cell cultures infected with the different BCV strains at tow passages. The vaccine, Meeker, Fisher, Calf 50 and Miller BCV strains agglutinated mouse but not chicken erythrocytes at infectivity yields of 5 x 105 to 4.5 x 106pFU s/50gl (Table2). The difference in the HA pattern of the avirulent and virulent strains may be determined by straindependent receptor binding properties, or possible differences between Neu 5,9 Acz-containing receptors on chicken and mouse erythrocytes, or the greater abundance of receptors on mouse erythrocytes [ 17] . Further differences between the avirulent strain BCV L9 and the virulent strains are host cell restriction [21, 22] , epitopes for S-specific monoclonal antibodies [6] , and amino acid substitutions among the S and HE proteins [32, 33] . Adaptation of wild-type BCV strains to cell cultures and the associated viral selection clearly involve significant genetic and functional changes. The difference in thermal sensitivity separated AE or R D E activity from the H A function. These studies do not define the structural proteins of BCV involved in HA. Monoclonal antibodies against H E as well as against S inhibited H A by BCV-Lg. The activity of anti HE monoclonal antibodies was predominately against R D E [24] . Purified H E of BCV-L9 had A E activity but agglutinated only mouse erythrocytes while purified S was a more powerful H A for chicken and mouse erythrocytes [17] . The functions of S and HE, and their interplay in receptor binding, HA, and the infectious process of BCV need to be defined.

Cell culture propagation of a coronavirus isolated from cows with winter dysentery

Recommendations of the coronavirus study group for the nomenclature of the structural proteins, mRNAs, and genes of coronaviruses

Structural proteins of bovine coronavirus and their intracellular processing

Morphology and morphogenesis of a coronavirus infecting intestinal epithelial cells of newborn calves

The receptordestroying enzyme of influenza C is virus neuraminidate-O-acetyl-esterase

Comparison of bovine coronavirus (BVC) antigens: monoclonal antibodies to glycoprotein GP 100 distinguish between vaccine and wild-type strains

Structure and orientation of expressed bovine coronavirus hemagglutinin-esterase protein

Bovine coronavirus structural proteins

Bovine coronavirus hemagglutinin protein

Neonatal calf diarrhea: propagation, attenuation, and characteristics of coronavirus-like agents

Hemagglutinating and acetylesterase activities of bovine coronavirus strains 203

Cloning and in vitro expression of the gene for the E 3 hemagglutinin glycoprotein of bovine coronavirus

Scanning electron microscopic characterization of bovine coronavirus plaques in HRT cells

Winter dysentery in adult dairy cattle: detection of coronaviruses in the feces

Hemagglutinating encephalomyelitis virus attaches to N-acetyt-9-0-acetylneuraminic acid-containing receptors on erythrocytes: comparison with bovine coronavirus and influenza C virus

Isolated HE protein from hemagglutinating encephalomyelitis virus and bovine coronavirus has receptor-destroying and receptor-binding activity

The S protein of bovine coronavirus is a hemagglutinin recognizing 9-0-acetylate sialic acid as a receptor determinant

Characterization of a calf diarrheal coronavirus

Coronaviruses: structure and genome expression

Structural proteins of bovine coronavirus strain L 9: effects of the host cell and trypsin treatment

Bovine coronavirus-induced cytopathic expression and plaque formation: host cell and virus strain determine trypsin dependence

Enhancement of plaque formation and cell fusion of an enteropathogenic coronavirus by trypsin treatment

On enteropathogenic bovine coronavirus

Monoclonal antibodies differentiate between the hemagglutinating and the receptordestroying activities of bovine coronavirus

The molecular biology of coronaviruses

Proteolytic cleavage of the E 2 glycoprotein of murine coronavirus: activation of cell-fusing activity of virions by trypsin and separation of two different 90 k cleavage fragments

Cultural and antigenic properties of newly established cell strains derived from adenocarcinomas of the human colon and rectum

The E 3 protein of bovine coronavirus is a receptor-destroying enzyme with acetylesterase activity

Human and bovine coronaviruses recognizes sialic acid-containing receptors similar to those of influenza C viruses

Structural polypeptides of the murine coronavirus JHM

Hemaggtutinating and acetylesterase activities in BCV strains ture, and biological activities of envelope protein gp 65 ofmurine coronavirus

Comparison of the nucleotide and deduced amino acid sequences of the S genes specified by virulent and avirulent strains of bovine coronaviruses

The hemagglutinin/esterase glycoproteins of bovine coronaviruses: sequence and functional comparisons between virulent and avirulent strains

We thank Dr. Georg Herrler for enhancing discussions and help with the acetylesterase tests. The dedication and excellent technical assistance of Eva Kroell is greatly appreciated. These investigations were made possible through a sabbatical leave in the Institut ffir Virologie, Justus Liebig Universitfit, Giessen, granted to J. S. by Louisiana State University, a reinvitation grant by the Alexander von Humboldt Foundation, through research grants 86-CSRS-2-2871 and 89-34116-4675 from the United States Department of Agriculture, and a grant from the Louisiana Education Quality Support Fund.

Received August 8, 1991