key: cord-0004992-85kn968m
authors: Hughes, G.; Babiuk, L. A.; van Drunen Littel-van den Hurk, S.
title: Functional and topographical analyses of epitopes on bovine herpesvirus type 1 glycoprotein IV
date: 1988
journal: Arch Virol
DOI: 10.1007/bf01319808
sha: e7c60108c8b5809a7c38d486078ec53e2bc39ba8
doc_id: 4992
cord_uid: 85kn968m

Bovine herpesvirus type 1 (BHV-1) glycoprotein gIV was purified by affinity chromatography. Purified preparations showed two distinct components of 71 K and 140 K following electrophoresis in sodium dodecyl sulphate polyacrylamide gels. The polypeptides were separated, excised from the gel and used to immunize rabbits; the resulting antisera showed a high degree of cross reactivity indicating that these polypeptides represent monomeric and dimeric forms of the same glycoprotein. Purified gIV was also used to develop a gIV-specific panel of monoclonal antibodies. Neutralizing monoclonal antibodies directed against gIV were conjugated to horseradish peroxidase and subjected to competition binding assays by ELISA. Three distinct neutralizing antigenic domains on gIV were identified. Domain 1 comprised two overlapping epitopes, whereas domain 2 was represented by a single monoclonal antibody. The third antigenic domain was made up of a complex of four identical or overlapping epitopes designated 3a, b, c, and d. Evidence is presented suggesting that domain 1 of gIV may be involved in penetration of the virus into the cell.

Bovine herpesvirus type 1 (BHV-1), a member of the alphaherpesvirinae, has been shown to express three major glycoproteins on the surface of infected cells and the viral envelope [-19-] . These glycoproteins, designated gI (130 K, cleaved to give 74K and 55K), gIII (91K with a 180K dimer), and gIV (71 K) are thought to be of major importance in host response to infection and in virus attachment and entry into the cell. Earlier studies utilizing affinity-purified glycoproteins as subunit vaccines have shown that all three glycoproteins can induce protection from a lethal challenge with BHV-1 followed by Pasteurella haemolytica, with gIV appearing to be of major importance in this regard [1] .

The range of glycoproteins identified in herpes simplex virus (HSV) is much wider numbering seven to date [13] . The three BHV-1 glycoproteins (gI, gIII, Published with the permission of the Director as Journal Series No. 67 and gIV) have been shown to be analogous to glycoproteins in HSV: gI to gB ofHSV, gIII to HSV gC and gIV to HSV gD [Zamb, manuscript in preparation]. HSV-1 gD induces high titres of virus neutralizing antibodies and is able to protect animals from experimental challenge [-6, 11, 12] . In HSV, gD is thought to be involved in virus penetration [3, 8] .

Thus far, we obtained one gIV-specific monoclonal antibody following immunization with purified BHV-1 [16] . In order to study the role of gIV in BHV-1 infection more fully, a panel of monoclonal antibodies was generated against affinity-purified gIV. The resulting neutralizing monoclonal antibodies were used to identify discrete epitopes of gIV on the surface of virions. Similar studies involving other glycoproteins of BHV-1 have revealed one neutralizing antigenic domain on gI and two on gIII using one panel of monoclonal antibodies [5] . Six neutralizing epitopes and one non-neutralizing epitope on gI as well as one neutralizing epitope on gIII were identified [18] with a different panel of monoclonal antibodies. Three neutralizing antigenic domains on gIII and one on gIV with a panel of monoclonal antibodies have been identified independently [ 15] .

In this study the neutralizing monoclonal antibodies obtained were used to try to elucidate the function of gIV of BHV-1. To date no specific function has been assigned to this glycoprotein in BHV-1 though recent studies with HSV [-3, 8] suggest that the equivalent glycoprotein (gD) is involved in virus penetration of the host cell.

The P8-2 strain of BHV-1 was propagated in Georgia Bovine Kidney cells (GBK cells) grown as monolayers in Eagle's Minimal Essential Medium (MEM) (Grand Island Biological Co., Grand Island, NY) supplemented with 10% fetal bovine serum (FBS) (Grand Island Biological Co.).

BHV-1 was purified from the culture medium of rapidly dividing GBK cells infected at 24 h post inoculation with a multiplicity of infection (moi) of 1. Following low speed and high speed centrifugation (5 minutes at 5 g and 90 minutes at 83,000 g, respectively) the virus pellets were centrifuged through potassium tartrate gradients as described previously [17] .

Glycoprotein gIV was purified on an immunoadsorbent column consisting of monoclonal antibody 3D9S [-1, 16], linked to Affigel-10. BALB/c mice (Jackson Laboratories, Bar Harbour, ME) were injected intraperitoneally with 0.1 ml (50 gg) of purified gIV mixed by sonication with an equal volume of Freund's complete adjuvant. One week later the mice were inoculated intraperitoneally with 50 gg of gIV in Freund's incomplete adjuvant. This was repeated after two weeks and three weeks. The final immunization on day 35 consisted of 50 gg of gIV in phosphate buffered saline inoculated into the tail vein. Fusion was Function and topography of epitopes on BHV-1 gIV 49 performed three days later by a modification of the method of Kennett et al. [91. The supernatants of growing hybridoma cell lines were screened for gIV-specificity using an ELISA with purified BHV-1 virions as the antigen. Clones showing the required specificity were subctoned in microtitre plates using the limiting dilution method.

Ascites fluids were obtained by intraperitoneal inoculation of 106 to 107 specific antibody-producing hybridoma cells into BALB/c mice that had been primed 2-3 weeks previously with pristane (2,6,10,t4-tetramethylpentadecane; Aldrich Chemicals, Montreal, Quebec).

Glycoprotein gIV was purified by immunoaffinity chromatography, followed by preparative sodium dodecyl sulphate polyacrylamide gel electrophoresis. The 71 K and 140 K glycoprotein bands were excised from the gel, emulsified in PBS and an equal volume of Freund's complete adjuvant. New Zealand white rabbits were given one initial immunization of each of the glycoproteins in Freund's complete adjuvant, followed by two subsequent immunizations with Freund's incomplete adjuvant at 4-week intervals. The rabbits were bled approximately 3 weeks after each immunization.

The ability of the hybridoma culture supernatants to mediate virus neutralization was tested as described previously [17] using 2-fold dilutions in the absence and presence of guinea pig serum (at 1/40 dilution) to determine complement-enhanced neutralization. Ascites fluids were tested for their neutralization titres which were judged to be the highest dilution resulting in a 50% reduction of plaques relative to a virus control.

The monoclonal antibodies were tested for their ability to neutralize virus after viral attachment by a modification of the standard neutralization assay. Virus was adsorbed to cell monolayers in 96-well plates for 2 h at 4 °C and incubated with dilutions of the monoclonal antibodies for a further 2 h at 4 °C. Finally the monoclonal antibody was removed and replaced with MEM containing 2% FBS and incubated at 37 °C for approximately 36 h. The monolayers were then stained and the levels of neutralization determined as a 50% reduction in plaques relative to a virus control.

Complement-dependent cell lysis was determined as described previously [2] .

The HRPO conjugated monoclonal antibodies were prepared as previously described [18] . Briefly, immunoglobulin was precipitated from ascites fluid by 50% (NH4)2SO 4 overnight, and the pellets were collected and desalted on PD 10 columns (Pharmacia, Montreal, Quebec) in 0.01 M NaHCO3/Na2CO3 pH 9.6. The IgG was then incubated for 3 h at 22 °C with activated HRPO at a ratio of 30mg IgG to 10mg HRPO. The reaction was halted by the addition of 10mg NaBH4 and the conjugates were dialyzed overnight at 0°C in phosphate buffered saline (PBS: 0.01 M NaH2PO4/NazHPO4, 0.15 M NaC1 pH 7.4).

Microtitre plates (Immulon 2, Dynatech Laboratories Inc., Alexandria, VA) were coated with 200 ~tl/well of BHV-t at 10 gg/ml in carbonate/bicarbonate buffer (0.05 M NaHCO3/ Na2CO3 pH 9.6) at 37 °C for 2 h or 4 °C overnight. Plates were washed six times in PBS with 50 G. Hughes et al.

0.05% TWEEN 20 (PBST). Monoclonal antibodies were screened at a 1/10 dilution in PBST (200 gl/well), and ascites fluids were tested to their dilution end points. After incubation for 2 h at 37 °C, the plates were again washed six times in PBST, they were then incubated for a further 2 h at 37 °C with 200 gl/well of affinity-purified, HRPO-conjugated goat antimouse IgG (Boehringer-Mannheim, Dorval, Quebec) diluted 1/1,000 in PBST. The plates were washed six times in PBST, and the substrate used was 5-aminosalicylic acid at 0.8 mg/ml with 0.006% H202. The cotour was allowed to develop for 30 rain. The absorbance at 492nm was measured in a Dynatech Microelisa reader (Model MR580) using wells without monoclonal antibody as controls.

The ELISA technique described above was modified for competitive binding assays by substituting the monoclonal and HRPO-conjugated second antibodies with HRPO-conjugated monoclonat antibodies. Color development and measurement of absorbance was as described e~bove. The percentage competition was calculated using the formula

where A is the optical density (OD) in the absence of competitor (i.e. unrelated monoclonal antibody conjugated with peroxidase), B is the OD in the presence of homologous antibody and n is the OD in the presence of competitive antibody [10] . Competition was regarded as any level greater than 80%. Intermediate competition was judged as between 60--80% competition, levels between 60% and -2 0 % were considered as non-competitive, less than -2 0 % was designated enhancement (negative competition).

Sodium dodecyl sulphate (SDS) PAGE was carried out according to the semi-denaturing method of Cohen et al. [4] . In this system the stacking gel buffer was 0.125 M Tris-sulphate pH 6.6 and the running buffer Tris-borate pH 8.3 containing 0.1% SDS. Samples were diluted in stacking gel buffer with 0.5% SDS but were not boiled before application to the gel. Gels were of 7.5% acrylamide (29 : 1, acrylamide : bisacrylamide) and 0.75 mm thickness. Electrophoresis was performed in a Hoeffer Mighty Small II apparatus (Hoeffer, San Francisco, CA) at 25 mAmps.

After termination of electrophoresis, gels were equilibrated in transfer buffer (0.192M glycine, 0.025 M Tris, 0.1% SDS, 20% v/v methanol) for 5 min before subjection to electrophoretic transfer to nitrocellulose. Transfer was carried out in a Hoeffer Mighty Small Transphor unit at 100 V for 30 rain in transfer buffer. Following transfer the nitrocellulose was washed in distilled water and then equilibrated in 0.02 M Tris-HC1, 0.5 M NaC1, pH 7.5 (TBS). A blocking reaction was performed in 3% gelatin in TBS for 1 h on a rotary shaker. The monoclonal antibody was applied at a dilution of 1/50 in 1% gelatin in TBS and the blot incubated overnight on a rotary shaker at room temperature. After washing in TBS the reaction was amplified by incubation for 2 h with rabbit anti-mouse IgG (heavy and light chains) (Cappel) at a dilution of 1/100 in TBS with 1% gelatin. Protein A-peroxidase conjugate (Bio-Rad) was applied after washing three times in TBS at a dilution of 1/2,000 in TBS with 1% gelatin for 2h. The colour development reagent 4-chloronaphthol was used according to the supplier's instructions (Bio-Rad).

When affinity-purified gIV was separated on a polyacrylamide gel, two bands were observed with apparent molecular weights of 140K and 71 K (Fig. 1) . Identical material after transfer to nitrocellulose and detection with a different monoclonal antibody against gIV than that used for purification is shown in the adjacent lane (Fig. 1 B) . This lane clearly shows two bands corresponding to the two stained bands in lane A (Fig. 1) .

Samples of protein from each of the two bands were excised from the acrylamide gel after electrophoresis and used to immunize rabbits. The rabbit sera were then titrated by ELISA against the homologous and heterologous proteins as shown in Table 1 , indicating extensive cross reactivity. Virus neutralization titres in the presence of complement are also shown. 

Monoclonal antibodies were screened first by their ability to recognize whole virions in ELISA; using this method fifty positive clones were obtained. These clones were then tested for their capacity to mediate virus neutralization. The resulting seven neutralizing clones were subcloned and ascites fluid was prepared from selected subclones which continued to demonstrate virus neutralization. Ascites fluids from these seven monoclones were further characterized with respect to their reactivity with gIV. a Neutralization titers were determined in the absence of complement and expressed as the reciprocal of the dilution which caused a 50% reduction of plaques relative to the virus control bAbC' was expressed as the percentage lysis at an antibody dilution of 1:10 c NT Not tested

All of the monoclonal antibodies were able to precipitate the 71 K and 140 K forms of gIV from [35S] methionine-labelled BHV-1 infected cells (not shown). Similarly, they all recognized gIV in its native state, as determined in an ELISA assay (Table 2) . However, the ability of the different clones to detect semidenatured gIV in a Western blot assay was much more variable; three clones (4C 1, 9 D 6, and 10 C2) showed a strongly positive reaction; two (3 C 1 and 136) were weakly positive, whereas the remaining three (2 C 8, 1 and 2, and 3 E 7) were negative. Only two monoclones, 9 D 6 and 10 C 2, were reactive with denatured gIV in an ELISA assay. Isotyping of these neutralizing monoclonal antibodies showed that they were all of the IgG subclass 1. The monoclonal antibodies were also analyzed with respect to their function (Table 3) . Whereas all seven monoclonal antibodies were able to neutralize BHV-1 at very high dilutions, only two of them, 9 D 6, and 136, were able to prevent penetration of the virus into the cell after attachment. In contrast, the only dilution of antibody resulting in any antibody-mediated, complement lysis was 1 : 10.

In order to analyze epitopes on glV, reciprocal competition binding assays (CBA's) were performed using all of the neutralizing monoclonal antibodies. Only one subclone of each monoclonal antibody was used since after characterization (Tables 2 and 3 ) the data indicated that the subclones included were not significantly different from each other. Figure 2 shows the CBA's for 3 E 7 as the conjugated antibody. In order to facilitate interpretation the graphs have been split. The only antibody showing any competition with the conjugated antibody was the homologous 3 E 7; 4 C 1, 10 C2, 2 C 8, and 3 C 1 all showed an absence of competition giving levels 136. The log 10 of the dilution is indicated on the X-axis between 10% and -1 0 % competition. Competing antibodies 9 D 6 and t36 showed "negative competition", i.e. enhancement. Figure 3 shows the competition binding assays for two conjugated monoclonal antibodies, 9 D 6 and 136. These two show similar patterns, they competed with each other for antigen and show similar patterns of non-competition/ enhancement. The graphs of 9 D 6 showed the classical patterns of 100% competition and absence of competition. When 136 was the conjugate some unusual features emerged. Firstly, 9 D 6 appeared to compete with 136 more effectively than the homologous antibody, at levels in excess of 100% competition. Secondly, the binding of some other antibodies enhanced the binding of conjugated 136 giving "competition" levels as low as -1 1 0 % , by 3 E7.

The competition patterns shown by four of the remaining monoclonal antibodies are shown in Fig. 4 . When 2 C 8 was used as the conjugated antibody, 3 E 7 showed enhancement of conjugate binding. The non-competitor in this 56 G. Hughes et al. graph was the 9 D 6 monoclonal antibody and 136 (not shown) showed a very similar pattern. All of the other monoclonal antibodies, 4 C 1, 10 C 2, and 3 C 1, competed successfully with 2 C 8 for antigen at levels between 80-110%. Generally similar results were found when the monoclonal antibodies were used in competition with the conjugated 4 C 1 antibody. However, 3 E 7 did not show enhancement with 4 C 1 (Fig. 4C ) as it did with 2 C 8 (Fig. 4A) . Monoclones 9 D 6 and 136 showed the same non-competitive pattern with 4 C 1 as with 2 C 8. The graph (Fig. 4B) shows that 10 C 2 competed most strongly with 4 C 1 with a maximum of 144% competition. Other antibodies (3 C 1, 2 C 8) also competed with 4 C 1 but at slightly lower levels, 2 C 8 at the same level as the homologous antibody, whereas 3 C 1 reached a maximum of 80%. The graphs showing competition with 10 C 2 ( Fig. 4 D and E) show a broadly similar pattern to those with the 2 C 8 and 4 C 1 conjugates. The same set of monoclonal antibodies competed at equivalent levels, that is, the homologous antibody 10 C 2 showed the maximum level of competition. The same was true for 10C2 in competition with 9 D 6 and 136, which did not compete at all. Monoclone 3 E 7 was also a non-competitor, although it showed slightly higher (though not significant) levels of competition than 9 D 6 and 136. The results of all of the competition binding assays are summarized in Table 4 .

Upon purification of glycoprotein glV (71 K) by affinity chromatography and analysis by SDS-PAGE and Western blotting, an additional polypeptide with an apparent molecular weight of 140 K was identified. The antibody responses of the rabbits that were immunized with the individual 71 K and 140 K polypeptides were evaluated by ELISA and neutralization assays, and showed ex-tensive cross-reactivity between these two species further indicating that they are related. Judging by their apparent molecular weight, the 140 K polypeptide is probably a dimer of gIV (71 K). The dimer was not reduced to a monomer by boiling in 2-mercaptoethanol, indicating that it is not linked by disulfide bonds. However, following purification of gIV, the dimer is more prevalent than after immunoprecipitation of [3SS]methionine-labelled BHV-t infected cells. Since during immunoaffinity purification of gIV the NP 40 concentration is decreased [17] , the 140 K species is probably a noncovalently linked dimer of 71 K-gIV. This is in agreement with the observations reported by Marshall et al. [14] for their 77 K/150 K polypeptide.

Extremely high neutralizing antibody titres were obtained by these monoclonal antibodies even in the absence of complement, expecially when compared with other reports [5, 15] . Only two of these antibodies, 9 D 6 and 10C2, recognized a continuous epitope, whereas the remaining five reacted with structural epitopes, which may explain at least some of the high neutralizing antibody titres. None of these monoclonal antibodies mediated immune lysis of virusinfected cells to any extent, whereas some monoclonal antibodies against BHV-1 glycoproteins gI and gIII showed lysis activity at dilutions greater than 1 : 10,000 [19] . This may be due to an inappropriate exposure of the critical gIV-specific epitopes on the cell membrane, making them inaccesible to either antibody or complement. Modifications or conformational changes in the glycoproteins during virus maturation may account for such a difference in topological orientation.

The competition binding assays suggested the presence of three different antigenic domains: the 136, 9 D 6 group was assigned epitopes la and lb respectively; the 3 E 7 group, antigenic domain 2 and the final group, antigenic domain 3 with four members 3a, 3b, 3c, and 3d (10C2, 4C 1, 2C8, and 3C 1 respectively).

Antigenic domain 2 comprised a single monoclonal antibody which showed the classical behaviour when in competition with other monoclones, i.e. 100% homologous and 0% heterologous competition. This suggests that this antigenic domain is a separate, independent epitope. Furthermore, it can be implied from the monoclonal antibody characterization data that this epitope is a discontinuous epitope containing amino acid residues from different parts of the glycoprotein. This was suggested by the inability of the antibody to recognize denatured antigen in ELISA or detect gIV on a Western blot. It is also important to note that 3 E 7, when acting as a competitor, showed some interaction with other antigenic domains. There appeared to be intermediate, non-reciprocal competition with 3 C 1, a member of antigenic domain 3. It also appeared to enhance the binding of 2C8 of domain 3 and 136 of domain 1, in a nonreciprocal manner. Interestingly, neither of these monoclonal antibodies recognised denatured antigen in ELISA and therefore probably recognise discontinuous epitopes. A possible mechanism for this process is that 3 E 7 binds to domain 2, causing distortion of gIV in the envelope resulting in alterations in other unrelated epitopes, seen as enhancement of binding. Coronavirus studies also indicate enhancement of up to 100% in the binding of specific monoclonal antibody pairs [7] , and this was also shown with BHV-1 gIII [15] .

Antigenic domain 1 was comprised of two monoclonal antibodies that compete with each other; they represent epitopes 1 a and 1 b. The differences in their levels of reactivity may be due to differences in avidity when competing for an identical site, or they may represent overlapping sites. The latter is probably the case when taken with the monoclonal antibody characterization data. The two clones reacted similarly in all tests with the exception of recognition of denatured antigen in ELISA; 9 D 6 was positive in this test and 136 was negative. Data from the post-adsorption neutralization assays strongly suggest that this domain is responsible for the penetration of the virus into the cell after viral attachment. Both monoclonal antibodies exhibited post-adsorption neutralizing ability although 136 was more efficient than 9 D 6. None of the other monoclonal antibodies showed any trace of post-adsorption neutralization. This confirms the observation that gIV is responsible for penetration of BHV-1 and indicates that antigenic domain 1 is the same as the antigenic domain II a on gIV reported earlier [ 15] . Evidence from gD of HSV-1 suggesting that gD is not involved in attachment of the virion but in its uptake into the cell [3, 8] also strengthens this hypothesis. Monoclone 136 showed enhanced binding following incubation with competing monoclonal antibodies 3 E 7 from domain 2 and 10C2 from domain 3, which suggests that binding to these epitopes caused distortion of the glycoprotein permitting easier access for monoclonal antibody 136 to its discontinuous epitope. However, no or very slight enhancement of binding of monoclone 9 D 6 was observed, which is consistent with the observation that the epitope recognized by this antibody is probably continuous, since it reacted in Western blotting and it showed an ability to recognize gIV after denaturation in an ELISA ( Table 2) .

The final antigenic domain, domain 3, is recognized by a group of monoclonal antibodies which recognized either exactly the same epitope or overlapping epitopes on gIV. The situation in domain 3 is probably similar to that of domain 1 in that the different monoclonal antibodies represent overlapping epitopes rather than avidity differences for the same epitope. This was indicated by the antibody characterization data shown in Table 2 . Each clone showed a different pattern of behaviour in Western blotting and ability to recognize denatured antigen. Clone 10 C 2 competed most strongly with the other members of this group. This may be due to high avidity or it may overlap with each of the other 3 epitopes. This clone also shows the maximum range of detection ability. Epitopes 3b and 3d probably overlap with each other and with 3a because they share similar characteristics, whereas epitope 3c probably only overlaps with 3a. The proposed interrelationship between these epitopes is shown in Fig. 5 . Domain 3 is probably discontinuous, at least in part, as has been indicated earlier in the discussion of domains 1 and 2. This domain was involved in non-reciprocal enhancement and competition by other domains. Furthermore, the antigen recognition data suggest that different parts of the glycoprotein brought together by secondary and/or tertiary structure contribute to the epitopes.

Taken together, the competition binding data and monoclonal antibody characteristics suggest three distinct neutralizing antigenic domains on gIV. Two of these domains, 2 and 3 are mostly discontinuous, whereas the third domain, 1, is partially continuous and responsible for virus penetration into the host cell.

Protection of cattle from bovine herpes virus type I (BHV-1) infection by immunization with individual viral glycoproteins

Defense mechanism against bovine herpesviruses: relationships of virus-host cell events to susceptibility to antibody-complement cell lysis

Entry of herpes simplex virus 1 in BJ cells that continously express viral glycoprotein of D is by entocytosis and results in degradation of virus

Localization of discontinuous epitopes of herpes simplex virus glycoprotein D: use of a nondenaturing

Neutralizing determinants defined by monoclonal antibodies on polypeptides specified by bovine herpesvirus 1

Vaccinia virus recombinant expressing herpes simplex virus type 1 glycoprotein D prevents latent herpes in mice

Antigenic structure of transmissible gastroenteritis virus II. Domains in the peplomer glycoprotein

Neutralizing monoclonal antibodies specific for herpes simplex virus glycoprotein D inhibit virus penetration

Hybrid plasmacytoma production: fusion with adult spleen cells, monoclonal spleen fragments, neonatal spleen cells and human spleen cells

Topographical analysis of antigenic determinants on envelope glycoprotein V 3 (E) of Japanese encephalitis virus, using monoclonal antibodies

Protection of mice from lethal herpes simplex virus infection by vaccination with a secreted form of cloned glycoprotein D

Glycoprotein D protects mice against lethal challenge with herpes simplex virus types 1 and 2

Identification of a herpes simplex virus 1 glycoprotein gene within a gene cluster dispensable for growth in cell culture

Characterization of envelope proteins of infectious bovine rhinotracheitis virus (Bovine herpesvirus-l) by biochemical and immunological methods

Mechanisms of neutralization by monoclonal antibodies to different antigenic sites on the bovine herpesvirus type 1 glycoproteins

Synthesis and processing of bovine herpesvirus-1, glycoproteins

Antigenic and immunogenic characteristics of bovine herpesvirus typel glycoproteins GVP 3/9 and GVP 6/11 a/16, purified by immunoadsorbant chromatography

Topographical analysis of bovine herpesvirus type-1 glycoproteins: use of monoclonal antibodies to identify and characterize functional epitopes

Interactions of monoclonal antibodies and bovine herpesvirus type-1 (BHV-1) glycoproteins: Characterization of their biochemical and immunological properties

We thank Elaine Gibbons and Jim Gilchrist for their technical assistance and Marilee Hagen for typing this manuscript. This work was funded by a grant from the Medical Research Council and the National Science and Engineering Research Council of Canada and from the Saskatchewan Agricultural Research Foundation.

Authors' address: Dr. S. van Drunen Littel-van den Hurk, Veterinary Infectious Disease Organization, 124 Veterinary Road, Saskatoon, Saskatchewan S 7 N 0W0, Canada.Received August 2, 1988