! "eterinao' Microbiology, 29 ( 1991 ) 75-83 Elsevier Science Publishers B.V., Amsterdam

75

A unified serotyping scheme for Moraxeiia bovis L.J. Moorea and A.W.D. Lepper b ~Department of Animal Health. The University of Sydney. Camden. N.S. ~t:. 25 70. Australia ~CS, LR.O. Division of Animal Health. Animal Health Research Laborator.v. Parkville. l'ictoria, 3052, Australia (Accepted 14 February 1991 )

ABSTRACT Moore, L.J. and Lepper, A.W.D., 1991. A unified serotyping scherae for Moraxella botts. Vet. Microbiol., 29: 75-83. Fifty-three Australian, seven British, two American and ,.wo New Zealand isolates of Mora_x'ella boris were classified into seven serogroups on tt~e basis el their variable fimbrial (lfilus) antigens using whole cell slide agglutination (SA), enzyme-linke.:l immunosorbent assays (ELISA) and tandem-crossed immunoelectrophoresis (TCIE). Although results of serogroup classification by SA and ELISA were identical .~n 68.7% of isolates, it v,as found necessary to resolve the discrepancies between the two systems using TCIE. Results suggest that world-wide variation in the potentially host-protective fimbrial antigens of M. boris may be relalively limited. It is proposed that the previous numerical classifications of British and Australian serogroups are appropriately amalgamated as a result of this latest study and are designated as serogroups A to G inclusive. A protocol for the further serotyping of fresh, fimbriate isolates of M. boris is suggested.

INTRODUCTION

Moraxella boris is the bacterial cause of infectious bovine keratoconjunctivitis (IBK) which is the most common ocular disease of cattle throughout the world. The disease has important implications both for animal welfare and lost production (Wilcox, 1968; Bedford, 1976; Baptista, 1979; Punch and Slatter, 1984). The fimbriae ofM. bovis are essential virulence determinants, promoting bacterial adherence to the bovine ocular epithelial cells and thus establishing infection (Wilcox, 1970; Pugh and Hughes, 1971; Pedersen et al., 1972; Chandler et al., 1979, 1980, 1983, 1985; M~)ore and Rutter, 1989). The fimbrial protein is also an important immunogen (Pugh et al., 1977, 1984; Lehr et al., 1985 ~. However, these protective fimbrial immunogens differ in many isolates ofM. bovis (Sandhu et al., 1974; Pugh and Hughes, 1976; Pugh et al., 1984; Lehr et al., 1985) and more recently the extent of such variation has been investigated to assess the possibility of future control of IBK with a polyvalent fimbrial vaccine. Thirty-eight British isolates of M. boris were di0378- I 135/9 i/$03.50

© 1991 - - Elsevier Science Publishers B.V.

76

L.J. MOOREAND A.W,D, LEPPER

vided into seven fimbrial serogroups using tandem-crossed immunoelectrophoresis (TCIEL an enzyme-linked immunosorbent assay (ELISA) and a slide agglutination assay (SA) (Moore and Rutter, 1987). By contrast, nine serogroups were identified among 66 isolates by ELISA, of which 56 were from Australia, six from Britain, two from the United States of America and two from New Zealand (Lepper and Hermans, 1986; A.W.D. Lepper, unpublished results). Recent collaboration between the authors of this work has provided the opportunity to compare the two schemes, in order to define one rational typing system which could be used to define the antigens which might serve as the basis for a polyvalent fimbrial vaccine to prevent IBK. MATERIALS AND METHODS

Isolates of M. boris The 55 Australian, six British, two American and two New Zealand isolates of M. boris, as serogrouped by the ELISA and used in the current investigations, are shown in Table 1. A further two isolates, one Australian (WW489) and one British (EFR27), were available but classified as untypeable by this method and isolate Dal 1 of the original serotyping scheme was unavailable (Lepper and Hermans, 1986; A.W.D. Lepper, unpublished results). The two American isolates were kindly supplied by Dr. G.W. Pug,h of the National Animal Diseases Center, Ames, IA, USA, the two New Zealand isolates by Dr R.B. Marshall of Massey University, P~lmerston North, New Zealand and Australian isolate UQV 149 by Dr J. Wooicock of the University of Queensland, St. Lucia, Australia. Freeze-dried isolates were resuspended in trypticase soy broth and cultures were grown aerobically on 5% bovine blood agar (5% (v/v) bovine blood, 1% (w/v) Difco Bacto peptone, 0.5% (w/v) Oxoid Lab Lemco meat extract, 0.5% (w/v) sodium chloride, 1.5% (wlv) Difco Bacto agar) at 37°C for 18 h.

Fimbrial antigen preparations Fimbrial antigens, prepared as described by Lepper and Hermans (1986) and stored at - 20°C were available from most of the M. boris isolates for use in the present investigations.

Fimbrial antisera Hyperimmune antisera raised to fimbrial antigens of the nine putative El_.lSA-serogroup representative isolates of M. boris (see Table 1 ) in rabbits and goats (Lepper and Hermans, 1986) and stored at - 2 0 ° C were used for the present investigations. Rabbit antisera were used in the SA assay and in TCIE for all serogroups with the exception of serogroup VII; due to the low

5RC6 3WO6 3WO7 3WO8

$276R $260R 25R 4L

EPP63 (USA)I V240(NH)-" V26g(NH) Ran4 Di96a V240

Ill (EPP63)

Dal20 Dallc Fla64 (USA) JlO H388 H359 H389 CECBD CECBE CECBF K38 MGR6 MGL3 709 (NZ) ~ 719 (NZ) M¢n750L

IV (Dal2d)

~( USA ) Isolate from the United States of America. -"(N H ) Non-hacmol.x tic isolate. ~! UK) Isolate from Great Britain. ~ NZ ) Isolate from New Zealand.

!! (5RC6)

I ($276R)

Scrogroup ( Representative isolate }

R593L Gal 339L H417 H358/CS MG35 14b LangB ( U K ) 137H D93a 352L Toc6547 L! MF3946L MF3946R CAV2S (UK)

V (R593L)

Initial serotyping scheme of M. bol.is isolates based on the ELISA

TABLE ~

Tat849 Tat848 Tat842 DJSI DJS5 DJS7 V 188L(NH ) 18L 102CL 102LC(NH ) 352R(NH) UQV 149

VI (Tat849)

FL462 Q220 ( U K ) 3

VIIi (FL462)

WW489 EFR27 ( U K )

Isolates nol typeable by ELISA:

218R Maff( I ) C561/6 (UK)

VIi (218R)

MF39721. I (48)L ( U K ) EFL27 ( U K )

IX (MF3972L)

w

78

L.J. MOORE AND A.W.D. LEPPER

titre of the rabbit antiserum to the fimbrial protein of serogroup VII, goat antiserum was used instead. Serological assays Fimbriate whole cell suspensions were serotyped against the panel of nine fimbrial antisera using the double antibody sandwich (quantitative) ELISA (Lepper and Hermans, 1986) and the SA assay (Moore and Rutter, 1987 ). Purified fimbriai antigens were investigated against the same panel of antisera using the chequerboard single step and competitive inhibition (absorption) ELISAs (Lepper and Hermans, 1986) and TCIE (Moore and Rutter, 1987).

Experimental design Preliminary screening of Australian, New Zealand and British isolates of M. boris after 1986 was performed using quantitative ELISA; if necessary, more precise assignment of an isolate to one of the six existing serogroups (Lepper and Hermans, 1986) or to a new serogroup was then performed using the chequerboard and absorption ELISAs. The 67 isolates of M. boris serotyped by ELISA (Table 1 ) were then serotyped by SA; TCIE was used only when SA could not be performed, the SA and ELISA results disagreed or results from the absorption ELISA were inconclusive. TABLE 2 Isolates of.ll, t,oris exhibiting ambiguous serogrouping by ELISA and SA isola;.~ ~

5RC6 EPP63 (USA) V240{NH) V268( NH j Ran4 DIO6a V240 18L 352R(NH) MF3O721_ I;48JL (UK) EFL27 I l I K } WW48 t~ EFR27 ( LIK ) ~See Table I for legend.

Serogroup based on ELISA

SA

I! I11 III Iil ill I!1 111 Vl V[ IX IX IX

VI VII VII Vii V V V VII V V V V V V

untypeablc untypeable

79

UNIFIED SEROTYPING FOR MOR.4XELL.I BOI "IS

TABLE 3

Revised serolyping scheme ofM. bol.is isolates based on the ELISA, SA and TCIE

Serogroup ' ( Representative isolate ) a ill 8 [nl C [~V l ($276R) {3WO7) (Dal2d)

D IV l (R593L)

E [Wl (Tat849)

F [Vll] (218R)

G IVm] (FL462}

$276R $260R 25R 4L

R593L Gal 339L H417 H358/CS MG35 14b LangB ( U K } i 37H D93a 352L Toc6547L 1 MF3946L MF3946R CAV2S (UK) Ran4 D 196a -'V240 352R(NH ) MF3~72L I (48~L (UK) EFL-~7..( !JK ) -~WW489 3EFR27

Tat849 Tat848 Tat842 VI88L(NH) 102LC 102LC(NH ) UQV149 5RC6

2 ! 8R Maff( 1 ) C561/6 (UK} EPP63 (USA) V240(NH~ V268(NH) 18L

FL462 Q220 (UK)

3WO6 3WO7 3WO8

Dal2d Dallc Fla64 (USA) Jl0 H388 H359 H389 CEC8D CEC8E CECSF K38 MGR6 MGL3 709 (NZ) 719 (NZ) Men750L

(UK) 'See Table I for legend. -'Isolate typed only by SA. ~lsolate typed only by SA and TCIE.

RESULTS

Slide agglutination assay For 46 of the 67 isolates investigated (68.7%), there was a direct correlation between the serogroups based on the ELISA and the SA assay. Of the remaining 21, four isolates (102LC, DJSI, DJS5 and DJS7) were non-timbriate upon reconstitution from the freeze-dried culture and therefore could not be serotyped by the SA assay. In three isolates (MGR6, V 188L(NH ) and D93a) the autoagglutination was .of sufficient strength to prevent the preparation of a suspension suitable for use in this assay. Fourteen isolates (Table

80

L,J. MOORE AND A W.D. LEPPER

2) were separated into different serogroups using the SA compared to the ELISA.

Tandem-crossed immunoelectrophoresis The TCIE results were divided into four categories based on the interaction of immunoprecipitation peaks as described previously (Moore and Rutter, 1987) and were classified as either: immunologically identical, closely related, remotely related or unrelated antigens. Fimbrial antigens in the first two categories were considered to be of the same serotype and were therefore grouped together. Of the four isolates ( 102LC, DJS 1, DJS5 and DJS7) which were non-timbriate at the time of the present investigations, a fimbrial preparation was only available for isolate 102LC. The TCIE results confirmed the position of this isolate in serogroup VI, as predicted by ELISA. Fimbriai preparations were available for all three excessively autoagglutinating isolates (MGR6, V 188L(NH) and D93a) and again the TCIE results confirmed those of the ELISA. For 12 of the 14 isolates shown in Table 2, the TCIE results confirmed those of the SA assay. For the remainder, isolate 352R (NH) was shown to be closely related to ELISA serogroup VI but identical to serogroup V and isolate V240 was not tested by TCIE because a fimbrial preparation was not available. Serogroups based on the ELISA, SA and TCIE The new serotyping scheme based on the ELISA, SA and TCIE as a re3ult of the present investigation is shown in Table 3. DISCUSSION

In the majority of cases investigated there was a correlation between the ELISA and SA serotyping results, indicating that the two assays detect the same antigenic epitopes of the fimbrial protein. Where the SA assay could not be performed due to lack of fimbriation or excessive autoagglutination, the TC|E assay confirmed the ELISA serogrouping of these isolates. The three isolates (DJSI, DJS5 and DJS7 ) which were non-fimbriate at the time ofthe present investigations, and for which fimbrial preparations were no longer available, have been eliminated from the revised serotyping scheme because they can no longer be used in fimbrial vaccines and challenge experiments. In i 3 of the 14 cases where there was a discrepancy between the ELISA and SA results, TC!E confirmed the serogroups predicted by the latter, the exception being an isolate for which a fimbrial preparation was not available and which was therefore untested by TCIE. Explanations for these discrepancies were found through a study of laboratory records for the ELISA. The serotyping of isolate 5RC6 (the original ELISA serogroup II represent-

UNIFIED SEROTYPING FOR MORAXELLA BO~'IS

8!

ative) into group VI by SA and TCIE was associated with the preparation of fimbrial and antiserum stocks for use in the ELISA from an unrecorded isolate, mistakenly labelled 5RC6. A new serogroup II representative isolate, 3WO7, has therefore been chosen and fimbrial protein and antisera are currently being prepared for use in the ELISA serotyping system. Isolate 352R(NH) was originally published as ELISA serogroup VI; however existing records, that had been overlooked, described an equally strong ELISA reaction of this isolate with serogroup V. There was, therefore, no true discrepancy between results obtained from all three serotyping methods. It is therefore recommended that isolate 352R(NH) be moved to serogroup V since this is the stronger reaction; however, it should be noted that this isolate also demonstrates a high level of cross-reactivity with serogroup VI. The relocation of ELISA serogroup III isolates into groups V and VII and of isolate 18L into group VII by SA and TCIE was associated with the initial serotyping of these isolates (using crude antigen preparations) by the quantitative ELISA only, the more specific chequerboard and absorption ELISAs were not performed in these cases. Similarly, the relocation of ELISA serogroup IX isolates and two other isolates, untypeable by ELISA, into serogroup V by SA and TCIE was associated with insufficiently defined results from quantitative and absorption ELISAs, although laboratory records did indicate a weak ELISA reaction with this antiserum. Overall, therefore, a study of the ELISA laboratory records and the use of TCIE have shown that the ELISA and SA techniques do give the same serotyping results, the original discrepancies being due to laboratory errors, insufficiently detailed investigations, the inability to perform assays for technical reasons, or difficulty in the interpretation of the assay results. The original ELISA-based scheme has therefore been reduced from nine to seven serogroups using SA and TCIE and we therefore propose a new system of nomenclature of serogroups A to G in order to avoid confusion with previous literature (Table 3). It should be noted that in addition to the two American and two New Zeao land isolates of M. boris, this new system includes six British isolates representative of British serogroups I to V and VII (Moore and Rutter, 1987). British serogroup VI is also represented by Australian isolate Men750L (L.J. Moore, unpublished results). The inclusion of these isolates within an Australian based serotyping scheme suggests that the range of variation among potentially host-protective fimbrJal antigens of M. bovis may be relatively limited. However, the level of protection offered by such antigens, differing to a lesser degree within a particular serogroup, has yet to be established. It should also be noted that any future work must involve the use of the serogroup representative fimbrial antigen as the protective antigen; whereas the ~erogroup-specific antise~m recognises all antigens in that group, by design, the reverse is not always the case.

82

LJ. MOORE AND A.W.D. LEPPER

Based on the findings of the present investigations we propose some recommendations for the serotyping of new isolates of M. bovis. Initially the isolate should be tested against the panel of seven serogroup representative antisera (serogroups A to G) by the SA assay. In a small number of cases where this is made impossible by excessive autoagglutination a preliminary indication of the serogrouping should be obtained using a quantitative ELISA and if considered necessary purified fimbriae should then be prepared and used in a chequerboard or absorption ELISA. Care must be taken over the interpretation of the ELISA results, and if they are equivocal, as occurred in the present investigation and as reported previously (Moore and Rutter, 1987 ) TCIE should be performed. Finally, if the new fimbriae are antigenically unrelated to any of those already characterised, antisera to the purified protein should be raised in rabbits and goats and a new serogroup established. ACKNOWLEDGEMENTS

We are indebted to the Australian Meat and Livestock Research and Development Corporation for its support and to Mrs Linda Morgan and Mrs Somsrong Spiess for their excellent technical assistance.

REFERENCES Baptista. P.J.H.P., 1979. Infectious bovine keratoconjunctivitis, a review. Br. Vet. J., i 35: 225242. Bedford, P.G.C., 1976. Infectious bovine keratoconjunctivitis. Vet. Rec., 98:134-135. (~ - ' h n n H h ~ r . . . . .

u . ~ . ~

I~ !

I~.~nl;ct~

P I 14 P ~ n , t T , r t ' r , ~

R

. . . ~ . ~

~

,

~ . ,

1

.

.

~

,

~

. ~ . , , , ,

.

~ , , ~

,

~,~,~.,,~

10"70 ~ t . d ; , ~ c n n lh,~ n ~~ l, h, n~ n , ~~ n. . i., ~- '. ;, ~ ,~f . . . . 1.lilt. . . . . •

*



~.

~

,

~

~ , ,

,,,~

~..

~)

#

,xc//~ ~o~'i~ in relation to infectious bovine keratoeon~unctivitis. J. Comp. Pathol.. 89: 441448. Chandler. R.L., Turfrey, B., Smith, K. and Gourlay, R.N., 1980. Virulence of M#r,.~e//a b~'i~ in gnotobiotie calves. Ve;. Rec., 106: 364-365. Chandler, R.L., Bird, R.G.. Smith, ~.D., Anger, H.S. and Turfrey, B.A., 1983. Scanning electron microscope studies on preparations of bovine cornea exposed to Moraxella bo~'is. J. Comp. Pathol., 93: I-8. Chandler. R.L.. Smith, K. and Tuffrey, B.A., 1985. Exposure ofbovine cornea to different strains of.|lora.~'clla l~m'is and to other bacterial species in vilro. J. Comp. Pathol., 95:415-423. Lchr, C.H., Ja.vappa, H.G. and Goodnow. R.A., 1985. Serologic and protective characterization of .llora.~clla I,ot'is pill. Cornell Vet., 75: 484-492. Leppcr. A.W.D. and Hermans, L.R., 1986. Characterisalion and quantitation of pilus antigens of.ll,ra.vcila bori~ by ELISA. Aust. Vet. J., 63: 401-405. Moore, L.J. and Rutter, J.M.. 1987. Antigenic analysis of fimbrial proteins from Moraxella l,,~'i.~. J. Clin. Microbiol., 25: 2063-2070. Moore, L.J. and Rulter. J.M., 1989. Attachment of .ll¢,'a.w'lla bo~'is to calf corneal cells and inhibition by antiserum. Aust. Vet. J.. 66: 39-42. Pedersen, K.B., Froholm. L.O. and Bovre, K.. 1972. Fimbriation and colony type ofMora.xclla t,m'is in relation to conjunctival colonization and development of keratoconjunctivitis in cattle. Acla. Palhol. Microbiol. Scand., 80B: 91 i-918.

UNIFIED SEROTYPING FOR MOK.IXELL.,! BO! "IS

83

Pugh, G.W. and Hughes, D.E., 1971. Infectious bovine keratoconjunclivitis induced by different experimental methods. Cornell Vet., 61: 23-45. Pugh, G.W. and Hughes, D.E., 1976. Experimental production of infectious bovine keratoconjunctivitis: comparison of serological and immunological responses using pill fractions of Moraxella boris. Can. J. Comp. Meal., 40: 60-66. Pugh, G.W., Hughes, D.E. and Booth, G.D., 1977. Experimentally induced infectious bovine keratoconjunctivitis: effectivenessof a pilus vaccine against exposure to homologous strains of Moraxella boris. Am. J. Vet. Res., 38:1519-1522. Pugh, G.W., Kopecky, K.E. and McDonald, T.J., 1984. Infectious bovine keratoconjunctivitis: enhancement of Moraxella bo~is pili immunogenicity with diphtheria-tetanus toxoids and pertussis vaccine. Am. J. Vet. Res., 45:661-665. Punch, P.I. and Slatter, D.H., 1984. A review of infectious bovine keratoconjunctivitis. Vet. Bull., 54: 193-207. Sandhu, T.S., White, F.H. and Simpson, C.F., 1974. Association of pili with rough colony type of Moraxella boris. Am. J. Vet. Res.. 35: 437-439. Wilcox, G.E., 1968. Infectious bovine keratoconjunctivitis: a review. Vet. Bull., 38: 349-360. Wit~ox, G.E., 1970. The aetiology of infectious bovine keraLo~onjunctivitis in Queensland. l. Moraxella boris. Aust. Vet. J., 46:409-414.

A unified serotyping scheme for Moraxella bovis.

Fifty-three Australian, seven British, two American and two New Zealand isolates of Moraxella bovis were classified into seven serogroups on the basis...
470KB Sizes 0 Downloads 0 Views