HYBRIDOMA Volume 10, Number S, 1991 Mary Ann Liebert, Inc., Publishers

Generation of Monoclonal Antibodies Against Surface Antigens of Moraxella bovis JOSÉ A.G. ALEIXO1-2, REJANE C. BARBOSA3'4,

and

CARLOS GIL-TURNES1'3 'Centro de Biotecnología, Universidade Federal de Pelotas (UFPel), Pelotas, RS, Brazil 2Faculdade de Nutriçâo, UFPel sFaculdade de Veterinaria, UFPel 4Present address: Departamento de Zootecnia, Universidade Estadual de Maringá, Maringá, PR, Brazil

ABSTRACT Six hybridoma lines producing monoclonal antibodies (MAbs) against noraxella bovis were established from fusions between the SP2/0 myeloma cells and BALB/c mice splenocytes. Three antibodies were of the IgGl isotype, two were IgG2a, and one was IgG2b. The specificity of the antibodies was determined by indirect enzyme-linked immunosorbent assay (ELISA) using whole cells of ft. bovis and of other Gram-negative bacteria, and lipopolysacharide (LPS) from ft. bovis JUR2 and E. coli as antigens. Ascitic fluid produced by the six hybridoma lines inhibited hemagglutination by ft. bovis GF9. One MAb (35F) reacted specifically with purified ft. bovis LPS in the ELISA test. The MAb panel detected heterogeneity among the isolates recovered from different geographical regions.

INTRODUCTION Noraxella bovis has been widely recognized as one of the more important etiological agents of infectious bovine keratoconjunctivitis (IBK), a highly contagious disease affecting cattle on a worldwide scale. The disease is of great importance since it affects animal health and causes economic loss(l,2). Pathogenesis of ft. bovis isolates is frequently associated with the existence of pili for adhesion (3,4,5) and with the ability to produce pit-like depressions in corneal epithelial cells (3,6). Pathogenic isolates also are hemolytic on blood agar and produce a variety of hydrolytic enzymes capable of damaging corneal cells (7,8). However, the importance of the hemolysin and the cornea damaging enzymes in IBK, and their relationship with the pitting factor(s) is still unclear. Several attempts have immunization with bacterins

made to protect calves against IBK through purified pili, but protection was higher against homologous challenge (7,10,11,12). Different serological types of pili have been identified in ft. bovis and recent studies indicated that a single strain of this organism is able to produce two or three types of pilins (4,13). Bacterial surface antigens play a critical role in the establishment of infections at mucosal surfaces since it is against these structures that the host immune response is mainly directed. The identification and isolation of surface antigens common to all ft. bovis strains would be of great value in the development of a reliable vaccine against IBK. The present study was undertaken to produce, purify and characterize monoclonal antibodies to H. bovis surface antigens. been

or

625

MATERIALS AND METHODS

Myeloma Cells and Antigens

Myeloma cell line SP2/0. Ag 14 was provided by N. Nardi, Departamento de Genética, UFRGS, Porto Alegre, RS, Brazil. H. bovis isolates Rl to R7 were clones obtained from the same animal; JUR, JURI, JUR2, JUR3, JUR5 and JUR7 were isolated from different animals of the same herd in successive years; GF9, FN, 17A and Al were isolated from different herds, all from Brazil. Isolates 2439-1, 2419-3 and 2358-1 were recovered in Uruguay and FLA64 and Epp63 in USA. Other bacterial strains are part of the culture collection maintained at our laboratory. Cultures used in experiments were grown on 105: sheep blood agar (Blood Agar Base, Biobrás, Montes Claros, Brazil). LPS was obtained by the method of Uestphal & Jann (1965) modified by Stead et al. (1975) and kindly provided by F. L. Araujo (14). Production of hybridomas The protocols used to establish the hybridomas and to produce the monoclonal antibodies (MAbs) were those described in the manual Hybridoma Techniques (EP1B0, SKMB Course 1980, Basel, Switzerland) with modifications. BALB/c mice were immunized by intraperitoneal injection of ca. 2xlOT cells of N. bovis GF9. Booster doses were given 2 and 4 weeks after the first immunization and spleens were harvested 3 to 4 days after the last dose. Spleen cells were mixed with SP2 cells at ratios 2-5:1 and fused in the presence of polyethylene glycol 1450 (J.T.Baker, Phillipsburg, NJ). Hybrid cells were plated on 96 well microtiter plates at a concentration of ca. 2x10* cells/well. The medium used to grow hybrid cells was Dulbecco modified Eagle medium (DMEM, Flow Labs., McLean, VA) containing 102 heat-inactivated fetal bovine serum (FBS) (Cultilab, Campinas, SP, Lsodium 2-mercaptoethanol, pyruvate, penicillin-streptomycin, Brazil), glutamine, hypoxanthine, aminopterin and thymidine (all from Sigma Chemical Co., St. Louis, NO). Supernatant fluids of wells showing growth were screened for reactivity by ELISA as described below. Cells from positive wells were cloned at least two times by the limiting dilution technique. The cloned cells were expanded to large culture flasks to obtain culture supernatants and to inject into mice to obtain ascites. Culture supernatants and ascitic fluids were stored at -20°C. Hybridoma cells were frozen at -70°C in FBS containing 10* DMS0 and stored in liquid nitrogen. Immunoglobulin class and sub-class of each monoclonal antibody were determined on 20-fold concentrated culture supernatants by double immunodifusion with specific antisera (Sigma). ELISA procedures Standard ELISA procedures were used to screen hybridoma culture supernatants for the presence of MAbs and to determine MAb cross-reactivity. Whole bacterial cells were suspended in carbonate-bicarbonate buffer (pH 9.6) at a concentration equivalent to McFarland No. 3 tube. Lipopolysaccharide (LPS) from H. bovis and from Escherichia coli (Sigma) were suspended in the same buffer at a concentration of 10 ug/ml. Bacterial and LPS suspensions (50 ul) were added to each well of 96-well PVC microtiter plates (Hemobag, Campinas, SP, Brazil) and incubated overnight at 4°C. Plates were washed three times with PBS containing 0.055: Tween 20 (PBS-T) and non-specific binding sites were blocked by incubation for lh at 37°C with a \>. solution of BSA in PBS. Fifteen microliters of culture supernatants or MAb partially purified from ascitic fluids with ammonium sulfate precipitation were added to each well and incubated for 3h at room temperature. Unbound antibodies were removed by washing three times with PBS-T and 50 ul of rabbit anti-mouse immunoglobulin conjugated to peroxidase (Dako, Denmark) was added to each well and incubated for 3h at room temperature. After washing the plates five times with PBS-T, 50 ul of a 2mPI ABTS solution (Sigma) was added to each well. Color development was detected visually and positive or negative results were determined by comparison with positive and negative control wells, made with sera obtained from immunized mice, as already described, and normal mouse serum, respectively.

626

Hemagglutination inhibition titers The capacity of MAbs purified from ascites to inhibit hemagglutination by isolates of ft. bovis was tested using the microplate technique. Protein concentration of each MAb preparation was determined by the Lowry procedure (15) and adjusted to 5mg/ml. Fifty microliters of doubled dilutions of each MAb in saline were mixed with one hemagglutinating unit of the isolate GF9. After incubation at 37°C for 30 min, an equal volume of a 1>. suspension of sheep red blood cells was added. The plates were placed at 4°C and readings performed when the showed clear hemosedimentation. One controls hemagglutination and hemagglutinating unit (HAu) was defined as the least quantity of bacteria contained in 50 ul of suspension that agglutinated an equal volume of l'< sheep red blood cells.

LPS treatments Heat treatment of ft. bovis LPS was done by subjecting a 0,1 mg/ml LPS solution in distilled water or in V. sodium dodecyl sulfate (SDS, Sigma) plus 5* 2-mercaptoethanol (Sigma) to 100°C for 10 min. Oxidation of the LPS solution was done according to the procedure described for the oxidation of peroxidase (15). The reactivity of the treated LPS was then assayed by ELISA.

RESULTS Six hybrid cell lines designated 47D, 410D, 311B, 52E, 35F and 36E were recovered from fusions between the SP2/0 myeloma cells and splenocytes from BALB/c mice immunized with whole ft. bovis cells. The MAbs secreted by the hybridoma lines were characterized by immunoglobulin's isotype and ability to inhibit hemagglutination by H. bovis GF9 (Table 1). MAbs 35F, 36E and 410D were of the IgGl subclass; MAbs 311B and 52E were of the IgG2a subclass and MAb 47D the IgG2b subclass. All the MAbs inhibited at different concentrations was of agglutination of sheep erythrocytes by H. bovis GF9. Each MAb was tested for immunoreactivity against various ft. bovis isolates, other Gram-negative bacteria, and LPS purified from ft. bovis JUR 2 and E. coli. These results are shown in Table 2. MAbs 47D, 311B and 35F reacted with the majority of the ft. bovis isolates tested. MAb 410D reacted only with the isolates from Brazil while MAb 36E reacted

TABLE 1

Immunoglobulin Isotype MAb

47D

*

**

and

Hemagglutination Immunoglobulin isotype •

Inhibition Titers of MAbs

HI titer

IgG2b

16

410D

IgGl

16

31 IB

IgG2a

8

52E

IgG2a

4

35F

IgGl

2

36E

IgGl

2

**

Determined by double immunodifusion with supernatants from hybridoma cultures Determined with MAbs from ascitic fluids. Values express the reciprocal of the maximal dilution that inhibited one HA unit of N. bovis GF9.

627

with some of the Brazilian isolates and with on« isolate from Uruguay. MAb 52E reacted with all the antigens tested. Besides reacting with whole N. bovis cells, MAb 35F reacted specifically with purified H. bovis LPS. N. bovis LPS dissolved in water or in SDS with mercaptoethanol and heated at 100°C for 10 min reacted with MAbs 35F and 52E in an ELISA, but reacted with none after meta-periodate oxidation.

DISCUSSION Six stable hybrid cell lines secreting MAbs of the immunoglobulin G class obtained from four fusion experiments. The MAbs were further characterized by hemagglutination inhibition (HI) assay and ELISA. All the MAbs inhibited hemagglutination by H. bovis GF9. MAbs 47D and 311B showed a strong inhibitory activity with isolate GF9 and reacted with 95* and 100*, respectively, of the isolates tested by ELISA. It is known that hemagglutination is related to virulence of H. bovis (17). Thus, these results suggest that MAbs 47D and 311B could be useful for the detection and isolation of common surface antigens. Two MAbs that reacted with antigenic determinants in LPS also showed HI activity. One (35F) recognized an epitope specific to N. bovis LPS, while the other (52E)

were

TABLE 2 Reactivities of MAbs anti-Noraxella bovis with Different

Monoclonal

Antigen

Antigens

in

an

antibody*

_

47D

410D

31 IB

52E

35F

36E

+

Noraxella bovis GF9 JUR JURI JUR3 JUR5 JUR7 Rl R2 R3 R4 R5 R7 17A FN Al 2439-1 2419-3 2358-1 FLA64

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

«

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+ +

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+ + + + -

Salmonella sp Neisseria sp Escherichia coli N. bovis LPS** E. coli LPS

-

=

-

+ -

-

+

-

-

__-+--

-

-

-

+

-

-

-

+

+

-

-

purified from ascites Lipopolysaccharide no reaction positive reaction;

MAbs

**

-

_--+_-

-

+

-

-

Epp63

*

-

-

-

=

-

628

-

-

ELISA

recognized an epitope common to N. bovis and E. coli LPS. The explanation of this fact was not elucidated, but it could be suggested that the MAb molecules produce a steric hindrance of adhesins thus preventing the recognition and adhesion to erythrocyte receptors. Steric hindrance of surface antigens by large molecules such as S-LPS of Brucella (17) and Coxiella (18) has been reported. In the ELISA, MAb 36E was the only one that did not react with all the Brazilian isolates tested. The other five reacted specifically with N.bovis suggesting that they share at least one common antigen. In contrast, strains from the USA and Uruguay had distinct reactivity patterns. These results agree with previous studies in which antigenic diversity of whole cells and pili antigens among strains of N. bovis from different geographical regions was detected with polyclonal antisera (20, 21). The epitope recognized by MAb 35F is in the H. bovis LPS as demonstrated by the maintenance of its activity after heat, SDS and 2-mercaptoethanol treatments and its disappearance after meta-periodate oxidation. In addition this MAb inhibited local Shwartzman reaction in white rabbits (data not shown). These results encourage its use for the study of the role of LPS in the pathogenicity and in the identification of H. bovis, such as the MAb specific for the putative lipid A region reported recently (22). This panel of MAbs detected differences among several strains of N. bovis that could not be detected with polyclonal sera (23). Further investigations, particularly the identification of antigens by immunoblotting and their isolation by affinity chromatography, are necessary to better characterize the antigens which might be potential candidates for more efficient vaccines against IBK. ACKNOWLEDGMENTS This work

was

(FINEP), Fundaçao

supported by grants from Financiadora de Estudos e Projetos Brasil, and Fundaçao de Amparo a Pesquisa do Estado do

Banco do

Rio Grande do Sul (FAPERGS). We thank Aleganí V. Monteiro for excellent technical assistance.

REFERENCES 1.

BAPTISTA, P.J.H.P. (1979). Br. Vet. J. 135, 225.

2.

PUNCH,

3.

CHANDLER, R.L., SMITH,K. and TURFREY.B.A. (1985). Exposure of bovine cornea to different strains of Noraxella bovis and to other bacterial species in vitro. J. Comp. Pathol. 95, 415.

4.

Infectious bovine keratoconjunctivitis. A review.

P.I. and SLATTER, D.H. (1984). keratoconjunctivitis. Vet. Bull. 54, 193.

A

review

of

infectious

bovine

MARRS, CF., SCH00LNIK, G., K00MEY, J.M., HARDY, T., ROTHBARD, T. and FALKOW, Cloning and sequencing of a Noraxella bovis pilin gene. J. 163, 132.

S. (1985). Bacteriol.

5.

6.

PEDERSEN, K.B., FR0H0LM, L.O. and BOVRE, K. (1972). Fimbriation and colony type of Noraxella bovis in relation to conjunctival colonization and development of keratoconjunctivitis in cattle. Acta Pathol. Microbiol. Scand. Sect. B. 80, 911. VOGELWEID, CM., MILLER, R.B., BERG, J.N. and KINDEN, D.A. (1986). Scanning microscopy of bovine corneas irradiated with sunlamps and challenge exposed with Noraxella bovis. Am. J. Vet. Res. 47, 378. electron

629

J.B. and L.C. Infectious bovine (1968). HENSON, GRUMBLES, keratoconjunctivitis. III. Demonstration of toxins in Noraxella (Heaophilus) bovis cultures. Cornell Vet. 51, 835.

PUGH,

G.W.

HUGHES,

D.E. (1968). Experimental bovine infectious caused by sunlamp irradiation and Noraxella bovis infection. Correlation of hemolytic ability and pathogenicity. Am. J. Vet. Res. 29, 835. and

keratoconjunctivitis

HUGHES,

D.E. and PUGH, G.W. (1976). Experimentally induced infectious bovine keratoconjunctivitis: Effectiveness of intramuscular vaccination with viable M. bovis culture. Am. J. Vet. Res.

32,

879.

LEHR. C, HUCHAPPA, G.J. and GOODNOW, R.A. (1985). Serologie and protective characterization of Noraxella bovis

PUGH,

pili. Cornell

Vet.

75, 484.

HUGHES, D.E. and BOOTH, G.D. (1977). Experimentally induced bovine keratoconjuntivitis: Effectiveness of a pilus vaccine exposure to homologous strains of H. bovis. Am. J. Vet. Res. 38,

G.W.,

infectious

against 1519.

WEBBER, J.J. and SELBY, L.A. (1981). Effects of Noraxella bovis vaccination schedules on experimentally induced infectious bovine keratoconjunctivitis. Am. J. Vet. Res. 42, 1181. LEPPER, A.W.D. and POWER, B.E. (1988). Infectivity and virulence of Australian strains of Noraxella bovis for the murine and bovine eye in relation to pilus serogroup sub-unit size and degree of piliation. Aust. Vet. J. 65, 305. ARAUJO,

F.L.,

ALVIANO,

Chemical composition Microbiol. 20, 165.

of

C.S., ANGLUSTER, J. 1ipopolysaccharide

and from

RICCIARDI, Noraxella

I.D. (1989). bovis. Vet.

LOWRY, D.H., ROSEBROUGH, N.J., FARR, A.L. and RANDALL. R.J. (1951). Protein phenol reagent. J. Biol. Chem. 193, 265.

measurement with the Folin

WILSON, M.B. and NAKANE, P.K. (1978). Recent developments in the periodate method of conjugating horseradish peroxidase (HRPO) to antibodies, in: Immunofluorescence and Related Techniques, W. Knapp, H. Holubar, and G. Wich (eds.), Elsevier/North-Holland, Amsterdam p. 215.

GIL-TURNES, C. (1983). Hemagglutination, autoagglutination and of Noraxella bovis strains. Can. J. Comp. Med. 47, 503.

pathogenicity

A., de WERGIFOSSE, P., DUBRAY, G. and LIMET, J.N. (1990). Identification of seven surface-exposed Bruceila outer membrane proteins by use of monoclonal antibodies: immunogold labeling for electron microscopy and enzyme linked immunosorbent assay. Infect. Immun. 58, 3980.

CLOCKAERT,

HACKSTADT,

T.

(1988).

Steric

hindrance

630

of

antibody

binding

of

surface

proteins

of Coxiella

burnetti

by phase I 1ipopolysaccharide. Infect. Immun.

56, 802.

GIL-TURNES, C. and ARAUJO, F.L. (1982). Serological characterization of strains of Noraxella bovis using double immunodiffusion. Can. J. Comp. Med. 46, 165.

LEPPER, A.W.D. and HERMANS, L.R. (1986). Characterization and quantitation of pilus antigens of Noraxella bovis by ELISA. Aust. Vet. J. 63, 401. WANNEMUEHLER, Y., JOHANSEN, K. and ROSENBUSCH, R. (1989). Identification of bovis by using a monoclonal antibody to a 1ipopolysaccharide epitope. J. Clin. Microbiol. 27, 2881. Noraxella

(1991). Quantification of Noraxella bovis adhesins with monoclonal antibodies. Lett. Appl. Microbiol.,

GIL-TURNES, C. and ALEIXO, J.A.G.

hemagglutinating in press.

Reprints shall be requested to: Dr. José Antonio G. Aleixo UFPel, Centro de Biotecnología 96100 Pelotas, RS, Brazil

631