67
FEMS Microbiology Letters 100 (1992) 67-70 © 1992 Federation of European Microbiological Societies 0378-1097/92/$05.00 Published by Elsevier
FEMSLE 80016
Enzyme-linked immunosorbent assay for the detection of antibodies to exocellular proteins of Staphylococcus aureus in bone infection P e t e r A. L a m b e r t ~, S t e p h e n J. Krikler b, Ragini P a t e l
a
and S a u d a P a r v a t h a n ~
Pharmaceutical Sciences Institute, Department of Pharmaceutical Sciences, Aston Unit;ersity, Birmingham, UK, and h Royal Orthopaedic Hospital, Birmingham, UK Received 9 June 1992 Accepted 18 June 1992
Key words: Staphylococcus aureus; Exocellular protein; Enzyme-linked immunosorbent assay; Bone infection; Serodiagnosis
1. S U M M A R Y An enzyme-linked i m m u n o s o r b e n t assay (ELISA) was developed for measurement of the antibody response to exocellular protein antigens of Staphylococcus aureus. The wells of a microtitre plate were coated with the exocellular proteins present in the tryptic soya broth growth medium from a stationary phase culture of a bone infection strain of S. aureus. The wells were then reacted sequentially with patient sera, protein A-peroxidase conjugate and chromogenic substrate. Serum from patients with S. aureus bone infection gave a significantly higher IgG titre than sera from patients with Staphylococcus epidermidis or Streptococcus sanguis bone infection or healthy uninfected individuals. The assay
Correspondence to: P.A. Lambert, Pharmaceutical Sciences Institute, Aston University, Aston Triangle, Birmingham B4 7ET, UK.
therefore appears to have potential in the serodiagnosis of S. aureus bone infection.
2. I N T R O D U C T I O N Detection of microbial infection of bone is a significant clinical problem [1]. It is particularly important to distinguish infection from mechanical loosening of a prosthesis because treatment of the two conditions is different [2]. Staphylococcus aureus is the most frequent isolate from osteomyelitis in humans [1,2]. However, its major cell wall antigens (peptidoglycan, wall and membrane teichoic acids, proteins) are of little value as serodiagnostic markers of infection because of the high levels of antibody found in healthy individuals [3]. Immunoblotting studies have shown that significant levels of exocellular protein antigens can be detected in culture supernatants [4]. These have potential both for typing strains [4] and in the serodiagnosis of S. aureus bone infection [5]. Our objective in this study was to investi-
68 gate further the serodiagnostic potential of extracellular antigens in S. aureus bone infection [5] by development of an ELISA system which can be used to screen large numbers of serum samples.
3. M A T E R I A L S AND M E T H O D S 3.1. Strains and growth conditions Seven strains of S. aureus were used: NCTC 6571 (Oxford strain); NCTC 10788; and five clinical isolates (two from osteomyelitis, two from cystic fibrosis lung infection and one from endocarditis). Cultures were grown in 100 ml of tryptic soya broth (TSB, Difco) for 18 h at 37°C with vigorous shaking in a rotary incubator. Cells were harvested by centrifugation (10000 × g, 10 min) and the culture supernatants stored at - 2 0 ° C for study of exocellular antigens. 3.2. Sera Serum was collected by venipuncture from five healthy adults and from patients with bone infection due to S. aureus, Staphylococcus epidermidis and Streptococcus sanguis and stored at -20°C. 3.3. Immunoblotting Electrophoresis and immunoblotting were carried out as described [5] using Mini Protean and Mini Trans Blot apparatus (Bio Rad). Nitrocellulose blots were reacted with sera diluted 1:400 with Tris-buffered saline/0.1% Tween 20 (TBST); bound IgG was then detected by reaction with protein A-peroxidase conjugate (Sigma) and 4c h l o r o n a p h t h o l / H 202 [5]. 3.4. ELISA 50 ml of culture supernatant was diluted with an equal volume of sodium carbonate buffer (0.05 M, pH 9.6) and 0.1 ml added to each well of a microtitre plate (Immulon, Dynatech). After incubation at 5°C for 15 h the plate was washed three times with phosphate buffered saline/0.1% Tween 20 (PBST). Unbound sites on the wells were blocked by incubation in the same buffer (1 h, 5°C). 0.1 ml samples of sera diluted in PBST were added to the wells and incubated for 3 h at 5°C. After washing three times in PBST bound
IgG was detected with protein A-peroxidase conjugate (0.5 txg/ml in PBST, 3 h at 5°C). The chromogenic substrate consisted of 3,3',5,5'-tetramethylbenzidine (10 mg) dissolved in 1 ml of dimethyl sulphoxide and diluted into 100 ml of sodium acetate/citrate buffer (0.1 M, pH 6.0) containing 10 txl H 2 0 2. 0.1 ml of the substrate mixture was added to each well. After suitable colour had developed (5 rain at 20°C) the reaction was stopped by addition of 50/xl 1 M HzSO 4 to each well. The colour was measured at 450 nm with an Anthos 2001 plate reader (Anthos Labtec Instruments).
4. RESULTS AND DISCUSSION Figure 1A shows that the seven strains grown in tryptic soya broth each gave a characteristic pattern of antigens which reacted strongly with IgG in serum from a S. aureus bone infection patient. Some bands were found in all the strains (e.g. the 61 kDa band), others were common to some but not all strains (e.g. the 81, 45 and 29 kDa bands). By contrast, the pooled sera from five healthy individuals showed virtually no reaction with proteins from the strains (Fig. 1B). We repeated the immunoblots with culture supernatants from the seven strains grown in four other laboratory growth media (brain heart infusion, Mueller-Hinton broth, iso-sensitest broth and nutrient broth). There was considerable variation in the pattern of antigens revealed. Each medium gave its own characteristic antigenic profile for each strain, and, in all cases, the patient's serum reacted strongly with bands on the blots, whereas the pooled sera from healthy individuals gave only a weak reaction. We selected one of the osteomyelitis strains grown in tryptic soya broth for development of an ELISA test. This strain produced four prominent protein bands at 81, 61, 45 and 29 kDa detected by the patient's serum (Fig. 1A, lane 4) plus many fainter bands. Its serodiagnostic potential was originally demonstrated by immunoblotting on nitrocellulose strips [5]. The control sera reacted only weakly (Fig. 113, lane 4). Figure 2 shows the ELISA titration curves for serial dilutions of sera
69
from patients and uninfected controls. Sera from the three patients with S. aureus bone infection showed high levels of IgG directed towards the
3"
2-
k
A 1
2
3
4
5
6
7
1-
v.
o.
1:100
816145-
1:400
1:1600
1:6400 1:25600
SERUM DILUTION Fig. 2. ELISA titration curves of sera from bone infection patients. Microtitre plate wells were coated with culture supernatant from the S. a u r e u s osteomyelitis strain in lane 4 of Fig. 1, then reacted with sera at the dilutions shown, followed by protein A-peroxidase and chromogenic substrate (Section 3.4). Sera were: l , • and A S. a u r e u s bone infection patients; D, * ~ and zx patients with bone infection due to S. e p i d e r m i d i s (three patients) and S. s a n g u i s respectively; [] pooled sera from five healthy individuals.
29-
B 1
2
3
4
5
6
7
81614.5-
29-
Fig. 1. Immunoblotting of culture supernatants from S. a u r e u s strains grown in tryptic soya broth: lane 1, NCTC 6571; lane 2, NCTC 10788; lanes 3 and 4 strains from osteomyelitis; lanes 5 and 6 strains from cystic fibrosis lung infection; lane 7, strain from endocarditis. Culture supernatants from each strain were denatured with an equal volume of sample denaturing buffer (5 min, 100°C) and 20 /~1 loaded on the gel. Blots were reacted with serum (diluted 1:50 in TBS/Tween) from the S. a u r e u s osteomyelitis patient from whom the strain in lane 4 was isolated (A) or pooled sera from five healthy controls (B). Bound IgG was detected with protein A-peroxidase and 4chloronaphthol/H 202.
exocellular antigens. A positive response (taken as an absorbance at 450 nm > 0.5) was obtained at dilutions up to 1:6400, whereas in uninfected controls, a negative response (absorbance < 0.5) was obtained at dilutions of 1:200 and above. Serum samples from patients with bone infection due to S. epidermidis (three strains) and S. sanguis were also tested. The titres ranged between 1:400 and 1:800 and were easily distinguished from those of the S. aureus bone infection patients. In general, the response of infected serum to whole-cell protein antigens revealed by immunoblotting cannot be distinguished from that of uninfected controls, reflecting the high endogenous antibody levels in the healthy population [3,5,10,11]. This study shows that, by careful selection of an exocellular antigen preparation, a test can be devised which selectively measures response to antigens prominent in bone infection. Clearly a large scale trial is now needed to evaluate the sensitivity and specificity. The identity of the antigens is not yet known. Up to 25 different exoceilular proteins have been reported in S. aureus [5,6], each involved in an
70
aspect of microbial pathogenesis. Comparison of molecular masses tentatively identifies the 81 kDa band as hyaluronidase [6]; the 61 kDa as acid phosphatase [6] or a high molecular mass form of coagulase [7]; the 45 kDa as exocellular protein A [8] or lysin [9]; whilst several exocellular proteins of molecular mass around 30 kDa have been reported, including epidermolysins A and B, B lysin, leucocidin and penicillinase [5]. It is interesting to speculate on why these exocellular antigens should be especially prominent in bone infection. It is known that bacteria in bone infections persist within a glycocalyx-enclosed biofilm adhering to the bone or to prosthetic devices [12-14]. In a chronic condition such as osteomyelitis, the host's immune system would be continually exposed to exocellular proteins released from organisms within the biofilm, whilst cellular antigens would remain in the protective structure of the enveloping glycocalyx [15]. Therefore, elevated levels of antibody directed towards these exocellular antigens can be distinguished from the pre-existing background of antibodies against cellular staphylococcal antigens.
REFERENCES [1] Green, S.A. and Dlabal, T.A. (1983) Clin. Orthop. 180, 117-124.
[2] Lifeso, F.M. and Faisal, A.-S. (1984) J. Bone Jt. Surg. 66B, 573-579. [3] Wergeland, H.I, Haaheim, L.R, Natas, O.B., Wesenberg, F. and Oeding, P. (1989)J. Clin. Microbiol. 27, 1286-1291. [4] Krikler, S.J., Pennington, T.H. and Petrie, D. (1986) J. Med. Microbiol. 21, 169-171. [5] Krikler, S.J. and Lambert, P.A. (1992)J. Med. Microbiol. 37, 227-231. [6] Arvidson, S.O. (1983) In: Staphylococci and Staphylococcal Infections (Easmon, C.S.F. and Adlam, C., Eds.), pp. 745-808. Academic Press, London. [7] Jeljaszewicz, J., Switalski, L.M. and Adlam, C. (1983) In: Staphylococci and Staphylococcal Infections (Easmon, C.S.F. and Adlam, C., Eds.), pp. 525-527. Academic Press, London. [8] Bjork, I., Peterson, B.A. and Sjoquist, J. (1972) Eur. J. Biochem. 29, 579-584. [9] Mollby, R. (1983) In: Staphylococci and Staphylococcal Infections (Easmon, C.S.F. and Adlam, C., Eds.), pp. 619-669. Academic Press, London. [10] Espersen, F. and Schiotz, P.O. (1981) Acta Pathol. Microbiol. Scand. Sect. C 89, 93-98. [11] Bell, J.A., Pennington, T.H. and Petrie, D.T. (1987) J. Med. Microbiol. 23, 95-99. [12] Gristina, A.G., Oga, M., Webb, L.X. and Hobgood, C.D. (1985) Science 228, 990-993. [13] Marrie, T.J. and Costerton, J.W. (1985) J. Clin. Microbiol. 22, 924-933. [14] Power, M.E., Olson, M.E., Domingue, P.A.G. and Costerton, J.W. (1990) J. Med. Microbiol. 33, 189-198. [15] Costerton, J.W., Cheng, K-J. and Geesey, G.G. (1987) Annu. Rev. Microbiol. 41,435-464.
FEMS Microbiology Letters 100 (1992) 67-70 © 1992 Federation of European Microbiological Societies 0378-1097/92/$05.00 Published by Elsevier
FEMSLE 80016
Enzyme-linked immunosorbent assay for the detection of antibodies to exocellular proteins of Staphylococcus aureus in bone infection P e t e r A. L a m b e r t ~, S t e p h e n J. Krikler b, Ragini P a t e l
a
and S a u d a P a r v a t h a n ~
Pharmaceutical Sciences Institute, Department of Pharmaceutical Sciences, Aston Unit;ersity, Birmingham, UK, and h Royal Orthopaedic Hospital, Birmingham, UK Received 9 June 1992 Accepted 18 June 1992
Key words: Staphylococcus aureus; Exocellular protein; Enzyme-linked immunosorbent assay; Bone infection; Serodiagnosis
1. S U M M A R Y An enzyme-linked i m m u n o s o r b e n t assay (ELISA) was developed for measurement of the antibody response to exocellular protein antigens of Staphylococcus aureus. The wells of a microtitre plate were coated with the exocellular proteins present in the tryptic soya broth growth medium from a stationary phase culture of a bone infection strain of S. aureus. The wells were then reacted sequentially with patient sera, protein A-peroxidase conjugate and chromogenic substrate. Serum from patients with S. aureus bone infection gave a significantly higher IgG titre than sera from patients with Staphylococcus epidermidis or Streptococcus sanguis bone infection or healthy uninfected individuals. The assay
Correspondence to: P.A. Lambert, Pharmaceutical Sciences Institute, Aston University, Aston Triangle, Birmingham B4 7ET, UK.
therefore appears to have potential in the serodiagnosis of S. aureus bone infection.
2. I N T R O D U C T I O N Detection of microbial infection of bone is a significant clinical problem [1]. It is particularly important to distinguish infection from mechanical loosening of a prosthesis because treatment of the two conditions is different [2]. Staphylococcus aureus is the most frequent isolate from osteomyelitis in humans [1,2]. However, its major cell wall antigens (peptidoglycan, wall and membrane teichoic acids, proteins) are of little value as serodiagnostic markers of infection because of the high levels of antibody found in healthy individuals [3]. Immunoblotting studies have shown that significant levels of exocellular protein antigens can be detected in culture supernatants [4]. These have potential both for typing strains [4] and in the serodiagnosis of S. aureus bone infection [5]. Our objective in this study was to investi-
68 gate further the serodiagnostic potential of extracellular antigens in S. aureus bone infection [5] by development of an ELISA system which can be used to screen large numbers of serum samples.
3. M A T E R I A L S AND M E T H O D S 3.1. Strains and growth conditions Seven strains of S. aureus were used: NCTC 6571 (Oxford strain); NCTC 10788; and five clinical isolates (two from osteomyelitis, two from cystic fibrosis lung infection and one from endocarditis). Cultures were grown in 100 ml of tryptic soya broth (TSB, Difco) for 18 h at 37°C with vigorous shaking in a rotary incubator. Cells were harvested by centrifugation (10000 × g, 10 min) and the culture supernatants stored at - 2 0 ° C for study of exocellular antigens. 3.2. Sera Serum was collected by venipuncture from five healthy adults and from patients with bone infection due to S. aureus, Staphylococcus epidermidis and Streptococcus sanguis and stored at -20°C. 3.3. Immunoblotting Electrophoresis and immunoblotting were carried out as described [5] using Mini Protean and Mini Trans Blot apparatus (Bio Rad). Nitrocellulose blots were reacted with sera diluted 1:400 with Tris-buffered saline/0.1% Tween 20 (TBST); bound IgG was then detected by reaction with protein A-peroxidase conjugate (Sigma) and 4c h l o r o n a p h t h o l / H 202 [5]. 3.4. ELISA 50 ml of culture supernatant was diluted with an equal volume of sodium carbonate buffer (0.05 M, pH 9.6) and 0.1 ml added to each well of a microtitre plate (Immulon, Dynatech). After incubation at 5°C for 15 h the plate was washed three times with phosphate buffered saline/0.1% Tween 20 (PBST). Unbound sites on the wells were blocked by incubation in the same buffer (1 h, 5°C). 0.1 ml samples of sera diluted in PBST were added to the wells and incubated for 3 h at 5°C. After washing three times in PBST bound
IgG was detected with protein A-peroxidase conjugate (0.5 txg/ml in PBST, 3 h at 5°C). The chromogenic substrate consisted of 3,3',5,5'-tetramethylbenzidine (10 mg) dissolved in 1 ml of dimethyl sulphoxide and diluted into 100 ml of sodium acetate/citrate buffer (0.1 M, pH 6.0) containing 10 txl H 2 0 2. 0.1 ml of the substrate mixture was added to each well. After suitable colour had developed (5 rain at 20°C) the reaction was stopped by addition of 50/xl 1 M HzSO 4 to each well. The colour was measured at 450 nm with an Anthos 2001 plate reader (Anthos Labtec Instruments).
4. RESULTS AND DISCUSSION Figure 1A shows that the seven strains grown in tryptic soya broth each gave a characteristic pattern of antigens which reacted strongly with IgG in serum from a S. aureus bone infection patient. Some bands were found in all the strains (e.g. the 61 kDa band), others were common to some but not all strains (e.g. the 81, 45 and 29 kDa bands). By contrast, the pooled sera from five healthy individuals showed virtually no reaction with proteins from the strains (Fig. 1B). We repeated the immunoblots with culture supernatants from the seven strains grown in four other laboratory growth media (brain heart infusion, Mueller-Hinton broth, iso-sensitest broth and nutrient broth). There was considerable variation in the pattern of antigens revealed. Each medium gave its own characteristic antigenic profile for each strain, and, in all cases, the patient's serum reacted strongly with bands on the blots, whereas the pooled sera from healthy individuals gave only a weak reaction. We selected one of the osteomyelitis strains grown in tryptic soya broth for development of an ELISA test. This strain produced four prominent protein bands at 81, 61, 45 and 29 kDa detected by the patient's serum (Fig. 1A, lane 4) plus many fainter bands. Its serodiagnostic potential was originally demonstrated by immunoblotting on nitrocellulose strips [5]. The control sera reacted only weakly (Fig. 113, lane 4). Figure 2 shows the ELISA titration curves for serial dilutions of sera
69
from patients and uninfected controls. Sera from the three patients with S. aureus bone infection showed high levels of IgG directed towards the
3"
2-
k
A 1
2
3
4
5
6
7
1-
v.
o.
1:100
816145-
1:400
1:1600
1:6400 1:25600
SERUM DILUTION Fig. 2. ELISA titration curves of sera from bone infection patients. Microtitre plate wells were coated with culture supernatant from the S. a u r e u s osteomyelitis strain in lane 4 of Fig. 1, then reacted with sera at the dilutions shown, followed by protein A-peroxidase and chromogenic substrate (Section 3.4). Sera were: l , • and A S. a u r e u s bone infection patients; D, * ~ and zx patients with bone infection due to S. e p i d e r m i d i s (three patients) and S. s a n g u i s respectively; [] pooled sera from five healthy individuals.
29-
B 1
2
3
4
5
6
7
81614.5-
29-
Fig. 1. Immunoblotting of culture supernatants from S. a u r e u s strains grown in tryptic soya broth: lane 1, NCTC 6571; lane 2, NCTC 10788; lanes 3 and 4 strains from osteomyelitis; lanes 5 and 6 strains from cystic fibrosis lung infection; lane 7, strain from endocarditis. Culture supernatants from each strain were denatured with an equal volume of sample denaturing buffer (5 min, 100°C) and 20 /~1 loaded on the gel. Blots were reacted with serum (diluted 1:50 in TBS/Tween) from the S. a u r e u s osteomyelitis patient from whom the strain in lane 4 was isolated (A) or pooled sera from five healthy controls (B). Bound IgG was detected with protein A-peroxidase and 4chloronaphthol/H 202.
exocellular antigens. A positive response (taken as an absorbance at 450 nm > 0.5) was obtained at dilutions up to 1:6400, whereas in uninfected controls, a negative response (absorbance < 0.5) was obtained at dilutions of 1:200 and above. Serum samples from patients with bone infection due to S. epidermidis (three strains) and S. sanguis were also tested. The titres ranged between 1:400 and 1:800 and were easily distinguished from those of the S. aureus bone infection patients. In general, the response of infected serum to whole-cell protein antigens revealed by immunoblotting cannot be distinguished from that of uninfected controls, reflecting the high endogenous antibody levels in the healthy population [3,5,10,11]. This study shows that, by careful selection of an exocellular antigen preparation, a test can be devised which selectively measures response to antigens prominent in bone infection. Clearly a large scale trial is now needed to evaluate the sensitivity and specificity. The identity of the antigens is not yet known. Up to 25 different exoceilular proteins have been reported in S. aureus [5,6], each involved in an
70
aspect of microbial pathogenesis. Comparison of molecular masses tentatively identifies the 81 kDa band as hyaluronidase [6]; the 61 kDa as acid phosphatase [6] or a high molecular mass form of coagulase [7]; the 45 kDa as exocellular protein A [8] or lysin [9]; whilst several exocellular proteins of molecular mass around 30 kDa have been reported, including epidermolysins A and B, B lysin, leucocidin and penicillinase [5]. It is interesting to speculate on why these exocellular antigens should be especially prominent in bone infection. It is known that bacteria in bone infections persist within a glycocalyx-enclosed biofilm adhering to the bone or to prosthetic devices [12-14]. In a chronic condition such as osteomyelitis, the host's immune system would be continually exposed to exocellular proteins released from organisms within the biofilm, whilst cellular antigens would remain in the protective structure of the enveloping glycocalyx [15]. Therefore, elevated levels of antibody directed towards these exocellular antigens can be distinguished from the pre-existing background of antibodies against cellular staphylococcal antigens.
REFERENCES [1] Green, S.A. and Dlabal, T.A. (1983) Clin. Orthop. 180, 117-124.
[2] Lifeso, F.M. and Faisal, A.-S. (1984) J. Bone Jt. Surg. 66B, 573-579. [3] Wergeland, H.I, Haaheim, L.R, Natas, O.B., Wesenberg, F. and Oeding, P. (1989)J. Clin. Microbiol. 27, 1286-1291. [4] Krikler, S.J., Pennington, T.H. and Petrie, D. (1986) J. Med. Microbiol. 21, 169-171. [5] Krikler, S.J. and Lambert, P.A. (1992)J. Med. Microbiol. 37, 227-231. [6] Arvidson, S.O. (1983) In: Staphylococci and Staphylococcal Infections (Easmon, C.S.F. and Adlam, C., Eds.), pp. 745-808. Academic Press, London. [7] Jeljaszewicz, J., Switalski, L.M. and Adlam, C. (1983) In: Staphylococci and Staphylococcal Infections (Easmon, C.S.F. and Adlam, C., Eds.), pp. 525-527. Academic Press, London. [8] Bjork, I., Peterson, B.A. and Sjoquist, J. (1972) Eur. J. Biochem. 29, 579-584. [9] Mollby, R. (1983) In: Staphylococci and Staphylococcal Infections (Easmon, C.S.F. and Adlam, C., Eds.), pp. 619-669. Academic Press, London. [10] Espersen, F. and Schiotz, P.O. (1981) Acta Pathol. Microbiol. Scand. Sect. C 89, 93-98. [11] Bell, J.A., Pennington, T.H. and Petrie, D.T. (1987) J. Med. Microbiol. 23, 95-99. [12] Gristina, A.G., Oga, M., Webb, L.X. and Hobgood, C.D. (1985) Science 228, 990-993. [13] Marrie, T.J. and Costerton, J.W. (1985) J. Clin. Microbiol. 22, 924-933. [14] Power, M.E., Olson, M.E., Domingue, P.A.G. and Costerton, J.W. (1990) J. Med. Microbiol. 33, 189-198. [15] Costerton, J.W., Cheng, K-J. and Geesey, G.G. (1987) Annu. Rev. Microbiol. 41,435-464.
