FEMS Immunology and Medical Microbiology 10 (1994) 47-54 © 1994 Federation of European Microbiological Societies 0928-8244/94/$07.00 Published by Elsevier

47

FEMSIM 00447

Vaccination with Staphylococcus aureus fibrinogen binding proteins (FgBPs) reduces colonisation of S. aureus in a mouse mastitis model

W u b s h e t M a m o a,1 Maria Bod6n b and Jan - Ingmar Flock ,,b,2

Swedish University of Agricultural Sciences, Department of Veterinary Microbiology, Box 583, S-751 23 Uppsala, Sweden, and b Center for Biotechnology, Karolinska Institutet, NoL, um, S-14I 57 Huddinge, Sweden

(Received 12 April 1994; revision received 23 August 1994; accepted 1 September 1994)

Abstract: A mouse mastiffs model was used to study the effect of vaccination with fibrinogen binding proteins and collagen binding protein from Staphylococcus aureus against challenge infection with S. aureus. The mice vaccinated with fibrinogen binding proteins showed reduced rates of mastitis compared with controls. Gross examination of challenged mammary glands of mice showed that the glands of mice immunized with fibrinogen binding proteins developed mild intramammary infection or had no pathological changes compared with glands from control mice. Histopathological examination of tissue sections from challenged glands showed that most glands from mice vaccinated with fibrinogen binding protein developed disseminated necrosis or had no pathological changes. A significantly reduced number of bacteria could be recovered in the glands from mice immunized with fibrinogen binding proteins as compared with controls. In a similar study, immunization of mice with collagen binding protein did not induce protection against challenge infection with S. aureus.

Key words." Staphylococcus aureus; Mastitis; Vaccination; Fibrinogen binding protein; Collagen binding protein

Introduction

Many pathogenic bacteria bind to extracellular matrix proteins, directly to cells or to plasma proteins as a step in the process of infection.

* Corresponding author. Tel: + 46 (8) 746 5802, Fax: + 46 (8) 711 3918. Present address: Section of Immunology, Preclinical R&D, AII, Astra Arcus, S-151 85 S6dertiilje, Sweden.

2 Present address: Karolinska Institutet, Division of Oral Microbiology, Huddinge University Hospital, F88, S-141 86, Huddinge, Sweden.

Staphy lococcus aureus has been shown to bind to fibronectin [1], collagen [2], laminin [3], elastin [4], fibrinogen [5,6] and vitronectin [7]. The bind- ing is mediated by surface located proteins. Many of these proteins have been characterized, puri- fied, and the corresponding genes cloned [8,9]. The exact role of these binding functions in vivo and in relation to the initiation of infection has not been established. However, the importance of fibronectin binding proteins (FnBPs) and colla- gen binding protein (CnBP) in endocarditis [10] (Schennings and Flock, unpublished), and arthri- tis [11] has been shown. The role of fibrinogen binding in connection with S. aureus binding to

SSDI 0928-8244(94)00056-5

48

foreign body implants has also been demon- strated [12,13].

We have previously characterized three differ- ent fibrinogen binding proteins (FgBPs), with molecular masses 19, 60 and 87 kDa, from S. aureus [14,15]. The 60 and 87 kDa proteins have coagulase activity and the 60 kDa protein can also bind to prothrombin. These FgBPs are mainly found in the extracellular medium. However, pu- r if ication on f ibr inogen Sepharose f rom lysostaphin-digested cellular fractions showed that a small amount of these proteins is associated with the cells. The 19 and 87 kDa proteins have been shown to be exposed on the cell surface [14] and it is possible that these proteins are responsi- ble for the binding of S. aureus to intact fibrino- gen and fibrin. The abundance of fibrin in wounds suggests a potential role for FgBPs in bacterial adherence during colonization and infection. FgBPs may also be targets for anti-microbial ther- apy by eliminating the fibrinogen binding func- tion(s). Immunizations with fusion proteins en- compassing the fibronectin binding D-domains of a FnBP from S. aureus have been shown to induce the production of antibodies with adher- ence blocking activity [16,17]. It was recently re- ported that immunization with these fusion pro- teins induced protection against S. aureus rat endocarditis [18] and mouse mastitis [19].

Bovine mastitis is one of the most costly dis- eases in dairy cattle and is mainly caused by S. aureus [20]. S. aureus strains isolated from masti- tis have been shown to bind to soluble fibronectin and collagen [21]. A mouse model for S. aureus

mastitis has been established [22] and this model makes it possible to follow the clinical aspects of the infection and to assess the protective role of antibodies against S. aureus after immunization with different immunogens. In this study FgBPs and CnBP from S. aureus were used to immunize mice.

Materials and Methods

C h e m i c a l s

Incomplete and complete Freund's adjuvants were purchased from Difco (Difco Laboratories,

Detroit, MI). All other chemicals were of analytic grade.

A n t i g e n p r e p a r a t i o n

The FgBPs of 19 and 87 kDa were prepared as described before [14]. Briefly, bacteria were grown in Brain Heart Infusion medium for 3 -4 h and the filtered supernatant was loaded onto a fib- rinogen Sepharose column. Bound material was eluted with 0.7% acetic acid containing 0.05% Nonidet P-40. A preparative polyacrylamide gel was run with the purified material and stained with Coomassie brilliant blue. The 19 and 87 kDa bands were cut out from the gel and eluted in a Model 422 Electro-Eluter (Bio-Rad, Hercules, CA). The collagen binding domain of CnBP from S. aureus was obtained from M. H66k, (Texas A & M University, Houston, TX). This domain, the A-region, was expressed as a fusion protein with glutathione S-transferase as described in [23].

A n i m a l s

BSVS mice strain [24], maintained by uncon- trolled random mating at the National Veterinary Institute, Uppsala, Sweden, were used through- out this trial. Female mice (2-3 months old), weighing 25-30 g, were intramammarily inocu- lated following their first pregnancy at days 10-15 of lactation [22]. The offspring was removed 2 h before challenge.

V a c c i n a t i o n s chedu le

A total of 10 mice were vaccinated with FgBPs and 15 mice with CnBP, while 5 mice were con- trols for the FgBPs vaccinated group and 10 mice were used as controls for the group of mice vaccinated with CnBP. Mice were vaccinated sub- cutaneously in the neck region with 22 /xg of the antigens emulsified with Freund's complete adju- vant (primary vaccination) 2 days after mating. Booster doses were given after 14 days with the same amount of immunogen, but emulsified with Freund's incomplete adjuvant.

Bac te r ia

S. aureus strain SAl l3 [25] was used to chal- lenge mice. Bacteria were cultured at 37°C for 18 h in tryptic soy broth (Difco). The bacteria were

49

washed three times in sodium phosphate buffer (PBS) pH 7.4, and the number of bacteria in the suspension was standardized to 1 × 10 6 c f u / m l in peptone water. These aliquots were kept at -196°C until used.

Challenge experiment Mice were challenged 14 days after the booster

dose by intramammary inoculation (mouse masti- tis model) [22]. Briefly, mice were anaesthetized using ether inhalation, the teat tips aseptically removed and a 0.1 ml bacterial suspension (1 × 106 c fu /m l ) was inoculated into both left and right 4th (L-4 and R-4) teats of a mouse. The clinical condition of challenged mice was regu- larly observed and recorded. Mice that died be- fore 48 h post-challenge were immediately exam- ined, whereas surviving mice were killed at 48 h, and their mammary glands removed for gross and histopathological examination and for quantifica- tion of the number of bacteria.

Gross examination of challenged mammary glands Challenged mammary glands (L-4 and R-4)

were examined under the stereo microscope (40X magnification) and pathological changes (types of lesions) were evaluated and recorded according to their macroscopic appearance [26].

Histopathological examination of challenged mam- mary glands

The L-4 glands were sectioned, stained with haematoxylin-eosin (HE) and histopathological changes evaluated under the microscope (400X

magnification). Histopathological changes of tis- sue sections were evaluated and recorded as de- scribed by [26].

Recovery of bacteria from the challenged mam- mary glands

Dissected R-4 mammary glands were placed separately in glass tubes containing 5 ml of sterile saline and were homogenized. Ten-fold serial di- lutions of homogenates were prepared, and 25/~1 of each dilution was plated on surface-dried blood agar. Colonies were counted after 24 h incubation at 37°C and the number of bacteria (cfu) in each mammary gland was calculated.

Evaluation of protection In this study, protection was defined as the

failure to reveal mastitis among vaccinated and challenged groups of mice as compared with a control group. The number of typical and atypical lesions developed in the glands of vaccinated versus control groups of mice were compared using X 2 analysis. The means and standard devia- tions (means + S.D.) of recovered bacteria were used to compare the results from intramammary growth using a two-tailed Student's t-test.

Results

Vaccination of mice with fibrinogen binding pro- teins (FgBPs)

Mice were vaccinated with the 19 and 87 kDa

Table 1 Gross and histopathological examination of challenged a glands of mice vaccinated with fibrinogen binding proteins and controls

Mice Gross examination c vaccinated No. glands Type of lesions: no. (%) with examined

+ + + / + + +/0

Histopathological examination d

No. of Type of lesions: no

glands A / B C examined

FgBPs (n = 10) 22 4, (18%) 18, (82%) 5 1 2 2

Control b

(n = 5) 10 6, (60%) 4, (40%) 4 3 1 0

a Mice were challenged with Staphylococcus aureus strain SAll3. b Control mice were injected with adjuvant without immunogens. c Gross examination was scored as + + + / + + = gross /medium mastitis; + / 0 = low g r a d e / o r no pathological change. d Histopathological examination was recorded as: A = non-reactive total necrosis; B = advanced regressive and mild inflammatory

reaction, C = disseminated and focal necrosis and 0 = no pathological reaction.

50

A B

10

8

T O 6

EO

c" '° 4 m~

- - 0

< ~ 2

Vaccinated

1 0 -

8 -

6 -

4 -

2 -

i i i

Controls Vaccinated Controls

Fig. 1. Average number of bacteria (log cfu + S.D.) recovered from the mammary glands of mice vaccinated with fibrinogen binding proteins (n = 10) and controls (n = 5) (A) and recovered from the mammary glands of mice vaccinated with collagen binding protein

(n = 15) and controls (n = 10) (B). The horizontal line indicates the number of bacteria with which the mice were inoculated.

FgBPs and challenged with S. aureus SA 113. Forty-eight hours later, challenged mammary glands were taken out for gross and histopatho- logical examination and to quantify the number of bacteria recovered in the glands. As shown in Table 1, 48 h post intramammary inoculation the number of challenged glands displaying severe or medium mastitis was reduced from 60 to 18% as a result of vaccination (P < 0.05). Eighty-two per-

cent of the glands from vaccinated mice showed low grade mastitis or had no pathological changes. The histopathological examination showed a ten- dency (not statistically significant at a 95% confi- dence level) towards a milder inflammatory reac- tion in the vaccinated group. Figure 1A shows the logarithmic number of bacteria recovered from the glands of mice vaccinated with FgBPs and controls. The average logarithmic number (means

Table 2 Gross and histopathological examination of challenged a glands of mice vaccinated with collagen binding protein and controls

Mice Gross examination c vaccinated No. glands Type of lesions: no. (%) with examined + + + / + + +/0

Histopathological examination d

No. of Type of lesions: no

glands A / B C examined

CnBPs (n = 15) 44 31, (70%) 13, (30%) 5 5 0 0

Control b (n = 10) 30 26, (87%) 4, (13%) 5 5 0 0

a Mice were challenged with a Staphylococcus aureus strain SAll3. b Control mice were injected with adjuvant without immunogens. c Gross examination was scored as + + + / + + = gross /medium mastitis; + / 0 = low g r a d e / o r no pathological change. d Histopathological examination was recorded as: A = non-reactive total necrosis; B = advanced regressive and mild inflammatory

reaction, C = disseminated and focal necrosis and 0 = no pathological reaction.

_+ S.D.) of bacteria recovered from the glands of vaccinated mice was 6.0 _+ 1.2, while 9.2 _+ 0.6 was recovered from the glands of control mice (P < 0.01).

Vaccination o f mice with collagen binding protein (CnBP)

Gross examination of mammary glands of mice vaccinated with CnBP showed a slight reduction (not statistically significant at a 95% confidence level) of mastitis compared with controls (Table 2). Histopathological examination of glands showed that all glands of mice examined from both groups developed non-reactive total necrosis with regressive inflammatory reaction (Table 2). After 48 h post intramammary inoculation, an average logarithmic number of 9.0 _+ 0.4 bacteria could be recovered in the glands of mice vacci- nated with CnBP and 9.6 _+ 1.0 could be recov- ered from control glands (Fig. 1B).

Discussion

In this experimental infection study, a mouse mastitis model was used to investigate the effect of immunization with fibrinogen binding proteins (FgBPs) and collagen binding protein (CnBP) on challenge infection with S. aureus. The mouse mastitis model largely mimics a bovine mastitis, where bacteria penetrate the teat canal into teat cisterns and eventually become established in the mammary glands. However, the size of the mouse mammary gland is relatively small and during the procedure it is severely traumatized. This model may therefore mimic not only mastitis but also wound infection caused by S. aureus.

In earlier experiments, we have demonstrated that antibodies raised against the D-domains of a fibronectin binding protein (FnBP-A) can inter- fere with the binding of FnBP to fibronectin [16,17] and with the binding of S. aureus to immobilized fibronectin [18]. As a result of vacci- nation with FnBP-A a reduced number of bacte- ria were recovered from heart valves in a rat endocarditis infection model [18] and from mam- mary glands in a mouse mastitis model [19]. We believe that this effect is partly due to an adher-

51

ence blocking effect. It has also been demon- strated that antibodies produced in mice against the D-domains of FnBP-A are opsonizing, thereby stimulating phagocytosis [27].

The mechanism behind the protective effect of anti-FgBPs antibodies seen here can only be speculated upon as long as the exact role of the FgBPs is unknown. It seems most likely that the FgBPs attached to the cell wall contribute to bacterial adherence to fibrinogen and fibrin, for example in wounds with blood clots. The antibod- ies might thus prevent efficient bacterial binding to the wounded tissue. However, most of the FgBPs are found in the extracellular medium, and the C-terminals of both the 19 kDa (Bod6n and Flock, unpublished) and the 87 kDa proteins [28] lack typical cell wall anchoring regions [29]. It has been suggested that coagulase from S. aureus strain SA 113 may be a virulence factor in a mouse mastitis model [25], but more recent data have shown that the gene for coagulase from S. aureus strain 8325-4 is not important for viru- lence in subcutaneous and intramammary infec- tions of mice [28]. However, immunization with coagulase might contribute by reducing the rate of infection and coagulase could thus be an im- portant antigen in a vaccine preparation against S. aureus. The function of the 19 kDa protein during the cause of staphylococcal infection is not known.

According to the immunization results pre- sented here, mice vaccinated with the glutathione S-transferase-CnBP fusion protein were not pro- tected against challenge infection with S. aureus. Histopathologically examined gland tissues from vaccinated and control mice showed no differ- ence, nor did the number of bacteria recovered from the glands. Earlier reports have also indi- cated that vaccination with CnBP fusion protein did not induce protection in a rat endocarditis infection model [18], despite the finding that col- lagen binding is important in maintaining endo- carditis (Schennings and Flock, unpublished). The fusion protein used here contains only the A domain responsible for collagen binding; it is still possible that CnBP presented in another way could induce protection.

Antibodies against the FgBPs were shown to

52

reduce colonization of S. a u r e u s in mice. This implies that it may be possible to design a vaccine against S. a u r e u s infections based on FgBPs. It is possible that passive immunization, using these proteins as antigens, may induce protection against certain hospital-acquired S. a u r e u s infec- tions, for example in patients with high suscepti- bilities to infections undergoing surgical treat- ments. Moreover, it could be anticipated that immunization with FgBPs might induce protec- tive antibodies against bovine intramammary in- fection caused by S. aureus .

Acknowledgements

This work was supported by Alfa Laval Agri- international and the Swedish Medical Research Council (Grant No. B94-16X-09092-05A).

References

1 Kuusela, P. (1978) Fibronectin binds to Staphylococcus

aureus. Nature 276, 718-720. 2 Speziale, P., Raucci, G., Visai, L., Switalski, L.M., Timpl,

R. and H66k, M. (1986) Binding of collagen to Staphylo-

coccus aureus Cowan I. J. Bacteriol. 167, 77-81. 3 Vercelotti, G.M., McCarthy, P., Lindholm. P, Peterson,

P.K., Jacob, H.S. and Furcht, L.T. (1985) Extracellular matrix proteins (fibronectin, laminin, and type IV colla- gen) bind and aggregate bacteria. Am. J. Pathol. 120, 13-21.

4 Park, P.W., Robert, D.D., Grosso, L.E., Parks, W.C., Rosenbloom, J., Abrams, W.R. and Mecham, R.P. (1991) Binding of elastin to Staphylococcus aureus. J. Biol. Chem. 266, 23399-406.

5 Hawiger, J., Kloczewiak, M. and Timmons, S. (1983) Inter- action of fibrinogen with staphylococcal clumping factor and with platelets. Ann. NY Acad. Sci. 408, 521-535.

6 Kuusela, P., Vartio, T., Vuento, M. and Myhre, E.B. (1985) Attachment of staphylococci and streptococci on fibrinogen, fibronectin fragments and fibrinogen bound to a solid phase. Infect. Immun. 50, 77-81.

7 Paulsson, M. and Wadstr6m, T. (1990) Vitronectin and type I collagen binding by Staphylococcus aureus and coag- ulase negative staphylococci. FEMS Microbiol. Immunol. 64, 55-62.

8 Flock, J.-I., Fr6man, G., Jonsson, K., Guss, B., Signals, C., Nilsson, B., Raucci, G., H66k, M., Wadstr6m, T. and Lindberg, M. (1987) Cloning and expression of the gene for a fibronectin-binding protein from Staphylococcus au-

reus. EMBO J. 6, 2351-2357. 9 Patti, J.M., Jonsson, H., Guss, B., Switalski, L.M., Wiberg,

K., Lindberg, M. and H66k, M. (1992) Molecular charac- terization and expression of a gene encoding a Staphylo-

coccus aureus collagen adhesin. J. Biol. Chem. 267, 4766- 4772.

10 Kyupers, J.M. and Proctor, R.A. (1989) Reduced adher- ence to traumatized rat heart valves by a low-fibronectin- binding mutant of Staphylococcus aureus. Infect. Immun. 57, 2306-2312.

11 Patti, J., Bremell, T., Krajewska-Pietrasik, D., Abdelnour, A., Tarkowski, A., Ryd~n, C. and H66k, M. (1994) The Staphylococcus aureus collagen adhesin is a virulence de- terminant in experimental septic arthritis. Infect. Immun. 62, 152-161.

12 Cheung, A. and Fischetti, V. (1990) The role of fibrinogen in staphylococcal adherence to catheters in vitro. J. Infect. Dis. 161, 1177-1186.

13 Herrman, M., Vaudaux, P., Pittet, D., Auckenthaler, R., Lew, D.P., Schumacher-Perdreau, F., Peters, G. and Waldvogel, F.A. (1988) Fibronectin, fibrinogen and laminin act as mediators of adherence of clinical staphylococcal isolates to foreign materials. J. Infect. Dis. 158, 693-701.

14 Bod6n, M.K. and Flock, J.-I. (1989) Fibrinogen-binding protein/clumping factor from Staphylococcus aureus. In- fect. Immun. 57, 2358-2363.

15 Bod6n, M. and Flock, J.-I. (1992) Evidence for three different fibrinogen-binding proteins with unique proper- ties from Staphylococcus aureus strain Newman. Microb. Pathogen. 12, 289-298.

16 Ciborowski, P., Flock, J.I. and Wadstr6m, T. (1992) Im- munological response to a Staphylococcus aureus fi-

bronectin binding protein. J. Med. Microbiol. 37, 376-381. 17 Luk, J., Flock, J.-I. and Wadstr6m, T. (1989) Detection in

rabbit sera of blocking antibodies against staphylococcal fibronectin binding protein by enzyme-linked immunosor- bent assay. FEMS Microbiol. Immunol. 47, 505-510.

18 Schennings, T., Heimdahl, A., Coster, K. and Flock, J.- I.(1993) Immunization with fibronectin binding protein from Staphylococcus aureus protects against experimental endocarditis in rats. Microb. Pathogen. 15, 227-236.

19 Mamo, W., Jonsson, P., Flock, J.-l., Lindberg, M., Muller, H.-P., Wadstr6m, T. and Nelson, L. (1994) Vaccination against Staphylococcus aureus mastitis; Immunological re- sponse of mice vaccinated with fibronectin binding pro- teins (FnBPs) to challenge with S. aureus. Vaccine 12, 988-992.

20 Anderson, J.C. (1983) Veterinary aspects of staphylococci. In: Staphylococci and Staphylococcal Infections (Easmon and Adlam, Eds.), pp. 193-241. Academic Press. London, New York.

21 Mamo, W., Fr6man, G. and Wadstr6m, T. (1988) Interac- tion of subepithelial connective tissue components with Staphylococcus aureus and coagulase-negative staphylo- cocci from bovine mastitis. Vet. Microbiol. 18, 163-176.

22 Chandler, R.L. (1970) Experimental bacterial mastitis in the mouse. J. Med. Microbiol. 3, 273-282.

23 Patti, J.M., Boles, J.O. and H66k, M. (1993) Identification and biochemical characterization of the ligand binding

domain of the collagen adhesin from Staphylococcus au-

reus. Biochemistry 32, 11428-11435. 24 Webster, L.T. (1937) Inheritance of resistance of mice to

enterobacterial and neurotropic virus infections. J. Exp. Med. 65, 261.

25 Jonsson, P., Lindberg, M., Haraldsson, I. and Wadstr6m, T. (1985) Virulence of Staphylococcus aureus in a mouse mastitis model: Studies of alpha haemolysin, coagulase and protein A as a possible virulence determinants with protoplast fusion and gene cloning. Infect. Immun. 49, 765-769.

26 Haraldsson, I. and Jonsson, P. (1984) Histopathology and pathogenesis of mouse mastitis induced with Staphylococ-

cus aureus mutants. J. Comp. Path. 94, 183-196.

53

27 Rozalska, B. and Wadstr6m, T. (1993) Protective opsonic activity of antibodies against fibronectin-binding proteins (FnBPs) of Staphylococcus aureus. Scand. J. Immunol. 37, 575-58O.

28 Phonimadaeng, P., O'Reilly, M., Nowlan, P., A.J. Bramley and Foster, T.J. (1990) The coagulase of Staphylococcus

aureus 8325-4. Sequence analysis and virulence of site- specific coagulase deficient mutants. Molec. Microbiol. 4, 393-404.

29 Fischetti, V., Pancholi, V. and Schneewind, O. (1990) Conservation of a hexapeptide sequence in the anchor region of surface proteins from gram-positive cocci. Molec. Microbiol. 4, 1603-1605.