Research in Veterinary Science 1992, 52, 90-96

Surface hydrophobicity of Staphylococcus intermedius and

Staphylococcus hyicus R. T. GREENE, C. L,~MMLER, M. SCHMITT, Institut ffir Bakteriologie und Immunologic der Justus-Liebig-Universitgit Giessen, Frankfurter Strasse 107, D6300 Giessen, Germany

Surface hydrophobicity of 90 Staphylococcus intermedius and 55 S hyicus isolates was evaluated using the hexadecane adherence assay and the ammonium sulphate salt aggregation test. A strongly positive hydrocarbon adherence in the hexadecane adherence assay was demonstrated in 11 per a n t of the S intermedius isolates and 7 per cent of the S hyicus isolates. Bacterial aggregation in 1"6 M, or less, ammonium sulphate was observed in 28 per cent of the S intermedius isolates and 37 per cent of the S hyicus isolates. There was no statistical correlation between the two assays. The adherence of both bacterial species to hexadecane was eliminated when the cells were frst treated with pronase and trypsin, while it was mildly enhanced by prior heat treatment (60°C and 95°C for up to three hours). In contrast, aggregation of Sintermedius in ammonium sulphate was not influenced by trypsin pretreatment, and aggregation of both bacterial species was diminished, or eliminated, with pronase or prior 95°Cheat treatment. Surface hydrophobicity, as measured in both assays, appeared to have no relationship with growth patterns in serum soft agar or production of slime. Similarly, the presence or absence of substantial surface receptor activity to fibrinogen, fibronectin or IgG did not appear to be related to surface hydrophobicity.

THE surface hydrophobicity of microorganisms may play an integral role in disease pathogenesis (Ofek and Beachey 1980, Kabir and Ali 1983, Seltmann et al 1986). The hydrophobic nature of bacterial cells has been shown to influence phagocytic interactions (van Oss 1978), binding to host tissues (Edebo et al 1980), and attachment to catheters or prosthetic devices (Christensen et al 1982, Hogt et al 1983, Boujaafar et al 1990). Surface hydrophobicity can be determined by

measurement of the attachment of cells to various polymers, salt aggregation, or affinity of bacteria to hydrocarbon solvents in a two-phase sygtem. Investigators have found that in certain staphylococcal species pathogenic to humans, cell surface hydrophobicity may be influenced by the presence of capsules, protein A or fibronectin binding proteins on the cell surface (Hogt et al 1983, Jonsson and Wadstr6m 1984). In addition, it has been suggested that slime production influences the surface hydrophobicity in some coagulase-negative staphylococci (Hogt et al 1983). Data on the hydrophobicity of S intermedius-and S hyieus, two important staphylococcal species pathogenic to animals are minimal. These species are both associated with chronic seborrhoeic illnesses and serious septicaemic infections in animals. This study was, therefore, performed to evaluate the surface hydrophobicity of these two bacterial species and the relationship this bacterial property may have to other surface characteristics. Materials and methods

Bacterial strains Ninety S intermedius isolates were obtained from 90 dogs from North Carolina State University (Raleigh). Half of these isolates were from the nasal mucosa or external ear canal and were considered to have been part of the normal flora. The other 45 isolates were from dogs with skin (n = 22), urinary tract (n -- 19) or miscellaneous (n = 4) diseases. All isolates were coagulase positive, Voges-Proskauer (Baird-Parker 1966) negative. Most produced acid slowly on purple agar supplemented with 1 per cent maltose (Kloos and Schleifer 1975), and all were further identified to the species level using a commercial 90

Surface hydrophobicity of Staphylococcus species identification system (API Staph-Ident, Analytab Products). Forty-four S hyicus isolates were collected from pigs and 11 from cattle. These isolates have been characterised previously (L~tmmler 1991). Unless otherwise stated, isolates were grown in brain heart infusion broth (BH0 (Merck). Hexadecane adherence assay Adherence to n-hexadecane (Sigma) was assayed using minor modifications of the method described by Rosenberg (Rosenberg et al 1980, Rosenberg 1984). Washed cells (late exponential phase growth) were suspended in phosphate buffered saline (PBs, pH 7-5) to an absorbance (A540) of 0.5 at 540 nm (Spectronic 20, Bausch and Lomb). To 3 ml of this suspension, if not otherwise stated, 0.8 ml hexadecane was added. The tubes were mixed (30 seconds), and after phase separation (20 minutes at room temperature), A540 of the aqueous phase was measured. Bacterial suspensions without hexadecane were used to measure the initial A540. Samples were run in duplicate so the final adherence value was calculated using the mean. The percentage of bacteria that adhered to hexadecane was expressed by the equation: A540 (initial)

A540 (with n-hexadecane)

x 100 = % Adherence

A540 (initial)

Adherence was considered positive (++) if the adherence was over 75 per cent, intermediate (+) if 25 to 75 per cent, and negative (-) if under 25 per cent. Enzymatic and heat treatment The effects of pronase E or trypsin (Merck) treatment on the staphylococcal adherence to hexadecane were evaluated on selected positive (++), intermediate (+) and negative (-) isolates. Before the addition ofhexadecane, enzymes (final concentrations of 50 gg m1-1, unless otherwise stated) were incubated with 3 ml suspension of bacterial cells (A540 = 0.5) for one hour at 37°C. After this incubation, hexadecane was added, and the subsequent steps were as described above. PBs (50 gl) without enyzmes was used for the corresponding controls. The effect of heat treatment (60~C and 95°C

91

for one, two and three hours) on the staphylococcal adherence was evaluated by heating an adjusted bacterial suspension (A540 = 0.5) before the addition of hexadecane. Salt aggregation test Salt aggregation was performed as previously described (Rozgonyi et al 1985). Briefly, overnight cultures were washed twice in 0.002 M sodium phosphate buffer, pH 6-8, and resuspended to a 10 per cent transmission using 620 nm wavelength. Of this suspension, 50 gl were mixed with equal volumes of ammonium sulphate (3.2, 2.6, 2.0, 1.4, 0.8 and 0.2 M), to give final volumes of 1.6 M, 1.3 M, 0-7 M, 0.4 M and 0.1 M, respectively. A visible cell aggregation within five minutes was regarded as a positive reaction. A bacterial suspension in plain buffer served as a control. The lowest final ammonium sulphate molarity which gave visual bacterial cell clumping was scored as the value for bacterial surface hydrophobicity. Enzymatic and heat treatment The effects of pronase or trypsin on salt aggregation were evaluated by adding the enzymes (final concentration of 50 gg ml-1) to 3 ml of the photometrically adjusted suspension. Heat treatment was evaluated by heating 3 ml of the suspension at 60°C and 95°C for one, two or three hours. The treated suspensions were subsequently evaluated for their salt aggregation. Soft agar and serum soft agar growth The serum soft agar technique has previously been described (Yoshida and Takeuchi 1970). The optimum bacterial concewtration was obtained by diluting 100 pl from a gently agitated overnight bacterial culture into 10 ml of 0.15 M sodium chloride. This same 1:100 dilution was then repeated five more times before 100 pl was inoculated into 10 ml of medium. The soft agar medium contained BHI and 0.15 per cent agar. For serum soft agar, 100 gl fresh rabbit serum was added to the medium. Growth was interpreted as either compact or diffuse, using S aureus (Cowan I) and S epidermidis (ATCC 14990) as controls for compact and diffuse growth, respectively.

92

R. T. Greene, C. Lammler, M. Schmitt

Slide agglutination tests for detection of substantial fibrinogen, fibronectin or IgG receptors

Results

Slide agglutination reactions were used to evaluate the reactivity of the staphylococci to fibrinogen (Deutsche KabiVitrum), fibronectin (Behringwerke) or IgG (Winblad and Ericson 1973, L/immler and Blobel 1984). A commercial system with IgG bound to human erythrocytes (Sangocell C; Behringwerke) was used to detect binding of IgG (Schaeg et al 1979). After overnight growth in 2 ml of BHI, bacteria were washed twice and resuspended in 0.5 ml 0.15 sodium chloride. One drop of the staphylococcal isolate and an equal volume of a 2 per cent fibronectin, 2 per cent fibrinogen or the commercial erythrocyte-IgG system were then mixed. Visible clumping, within three minutes, indicated a positive clumping reaction. S aureus (K807) served as the positive control for fibrinogen assays, while S aureus (Cowan I) was used as the positive control for fibronectin and IgG tests. S epidermidis (ATCC 14990) was used as the negative control for all protein clumping studies.

The percentage bacterial adherence to hexadecane increased as the volume of hexadecane added to the bacterial suspension was increased. Even hydrophilic isolates were found to adhere to some degree if large volumes ofhexadecane were added (Figs 1 and 2). Experiments using increasing volumes of hexadecane established that 0.8 ml provided excellent distinction between hydrophilic and hydrophobic strains. For this reason, this volume was used throughout this study. The majority of the isolates from both bacterial species displayed intermediate (25 to 75 per cent) adher-

Hexadecane adherence assay

100-

80-

&

A [] []

60-

0

0 []

40.

[]

20-

Slime detection A slight modification of a previously described quantitative slime detection assay was used (Pfaller et al 1988). In each well of a microtitre plate, 10 ~1 of an overnight culture was added to 200 ~ of fresh medium. After 18 hours incubation at 37°C, the wells were washed three times with PBS, stained with methylene blue (250 ILl) for 15 minutes, and then washed another three times. Final colour development was analysed five minutes after the addition of 0.25 M EDTA (250 ~). An ELISA reader with a photometric filter of 590 nm was used. Positive and negative controls of S epidermidis RP62A and RP62NA were used, respectively (generously provided by Dr Gerald B. Pier, Harvard Medical School, Boston).

0 0

0 0.0

Q

0.2

0

9

0.4

,n

9

0.6 0.8 110 Volumehexadecane(mr/

1.

114

116

FIG 1 : Effects of increasing volumes of hexadecane on adherence using ~ (++) positive, [] (+)intermediate, and © (-) negative Sintermedius isolates 100[]

80-

,,,

[]

60A

0

40-

0 0

20-

0

[]

0

[]

Statistical analysis

0

,°2

0 0.0

Q

o.2

0

9

0.4

0

0

o:6

0:8

1:o

112

~14

116

Volume hexadecane(m~

Statistical analysis of the differences between the two hydrophobicity assays was performed using the Spearman rank correlation coefficient.

FIG 2: Effects of increasing volumes of hexadecane on adherence using • (++) positive, ~ intermediate (+), and © (-) negative S hyicus isolates

Surface hydrophobicity of Staphylococcus species TABLE 1 : Surface hydrophobicity of 90 Staphylococcus intermedius and 55 S hyicus isolates as determined by the hexadecane adherence assay and the salt aggregation test

Final ammonium sulphate molarity (M)

Salt aggregation* (Final ammonium sulphate molarity) (%) Hexadecane No adherence:~ aggregation 1.6 M 1.3 M 1.0 M 0.05). Also, no relationship was found between either of the hydrophobicity assays and the clinical source of the S intermedius isolates. Fibrinogen, fibronectin and IgG slide agglutination tests Fibrinogen- and fibronectin-clumping was detected in only a minority of the isolates (Table 4). Similar findings were obtained for IgG-clumping using S intermedius, but the data for IgG clumping were much different when S hyieus was used (Table 4). Those isolates that were positive in these tests did not necessarily display either hydrophobic or hydrophilic characteristics in either of the two hydrophobicity assays (data not shown).

Slime assay and serum soft agar growth Salt aggregation test Bacterial aggregation in 1.6 M or less ammonium sulphate was observed in 28 per cent of the S intermedius isolates and 37 per cent of the

Slime production was not detectable by any isolate. Diffuse growth in serum soft agar was detected only in three (3 per cent) S intermedius and two (4 per cent) S hyicus isolates

R. T. Greene, C. Lammler, M. Schmitt

94

TABLE 4: Summary of positive surface receptor activity for fibrinogen, fibronectin or IgGand other surface characteristics of 90 S intermedius and 55 S hyicus isolates

S intermedius Fibrinogen binding Fibronectin binding IgG binding Slime production Diffuse growth in serum soft agar

5 (6%) 1 (1%) 6 (7%) 0 3(3%)

S hyicus 1 (2'/o)

0

36 (65'/o)*

o 2 (4%)

* All isolates collected from cattle were negative, while only five of 41 (12 per cent) of isolates obtained from pigs were negative

(Table 4). The surface hydrophobicity test results for these isolates failed to suggest an association with diffuse growth. Discussion

Since the first step in bacterial infections involves contact of the bacterial surface with host cells, surface hydrophobicity has been postulated to be an important factor influencing the initial bacterial-host interaction. Various methods exist for studying bacterial surface hydrophobicity, but the exact properties being measured by these techniques are still unknown (Ofek et al 1983, Jonsson and Wadstr6m 1984, Reifsteck et al 1987). It is apparent from the present results with S intermedius and S hyicus that the hexadecane adherence assay and the salt aggregation test measured different surface characteristics contributing to hydrophobicity. This was demonstrated not only by the lack of correlation between the results from these assays, but also by the different effects that enzymatic and heat treatment had on the results. Capsules, protein A, fibronectin receptors, or slime production have been inconsistently associated with surface hydrophobicity in other staphylococci (Hogt et al 1983, Jonsson and Wadstr6m 1984, Reifsteck et al 1987). Nevertheless, the present authors were not able to demonstrate such associations. Evidence incriminating teichoic acid or lipoteichoic acid in the adherence of various Gram-positive cocci to host cells or hydrocarbons has also been described (Aly et al 1980, Simpson et al 1980, Mi6rner et al 1983). In these investigations involving lipoteichoic acid, hydrocarbon adherence was eliminated after enzymatic treatment, but yet was minimally changed after heat treatment. This

thermostability of hydrocarbon adherence was also demonstrated in- the present study, thus, possibly suggesting some lipoteichoic acid influence. However, surface proteins probably played an important role in salt aggregation. This is inferred from the fact that heating of the cells at 95°C caused a decline in aggregation, while heating at 60°C did not affect this property. In addition, pronase consistently eliminated reactivity. The failure of trypsin treatment to eliminate salt aggregation in S intermedius isolates reflects the specificity of this enzyme for limited proteins. Therefore, surface hydrophobicity, as measured in both assays, is most likely influenced by a complex mosaic of molecules, one of which may be lipoteichoic acid. These molecules interact to determine biospecific, electrostatic and hydrophobic interactions (Jonsson and Wadstr6m 1984). The growth characteristics in serum soft agar were used as a functional indicator of the presence of a capsule (Yoshida and Ekstedt 1968, Yoshida et al 1977). With this technique, encapsulated strains show diffuse growth, whereas compact growth indicates the absence of a capsule (Norcross and Opdebeek 1983, Anderson 1984, Opdebeek et al 1987). This methodology is more sensitive than the standard india ink method of capsule detection (Duguid 1951), but a more precise method of capsular antigen detection may eventually involve the use of specific anti-capsular monoclonal or polyclonal antibodies (Nelles et al 1985, Sutra et al 1990). These antibodies specifically react with the capsules of S aureus. Future investigations may eventually determine whether these antibodies react with capsules from S intermedius or S hyicus. Nevertheless, in this study, one known encapsulated S hyicus strain (Yoshida et al 1988) did demonstrate diffuse growth in serum soft agar. An association between surface hydrophobicity and pathogenicity has periodically been demonstrated for other bacterial infections (Pascual et al 1986, Boujaafar et al 1990). HoweVer, in this study, such a relationship was not observed for S intermedius (data not presented). Isolates considered to be collected from the normal flora were just as likely to be hydrophobic or hydrophilic as those isolated from dogs with clinical disease. Also, isolates from pyodermas gave similar results to those from urinary tract infections. It is apparent that surface characteristics, such

Surface hydrophobicity o f as the presence of substantial protein receptor activity, slime production or capsular material are not factors that dramatically influence surface hydrophobicity in these two pathogenic staphylococcal species. Other, as yet unknown, surface proteins or surface properties probably play an important role. Discovery of these surface factors may enhance our understanding of bacterial-host interactions. Certainly, the detection in these two bacterial species of sporadic isolates with extreme hydrophobicity and hydrophilicity provides evidence that further applications of surface hydrophobicity assays may be useful. For instance, these assays may possibly be used to separate certain hydrophobic or hydrophilic phenotypes from mixed cultures, or possibly as epidemiological typing tools. Acknowledgements The authors would like to give recognition to the fine technical assistance of Heike Runge. In addition, gratitude is extended to Pat Rottman and Dr W. E. Kloos for their initial research assistance. Finally, sincere thanks to Mr K. Failing and Mr H. Heiter for their statistical assistance. This research was supported by the Alexander yon Humboldt Stiftung. References ALY, R., SHINEFIELD, H. R., LITZ, C. & MAIBACH, H. I. (1980) Role of teichoic acid in the binding of Staphylococcus aureus to nasal epithelial ceils. Journal of Infectious Diseases 141,463-465 ANDERSON, J. C. (1984) Absence of encapsulation in strains of Staphylococcus aitreus isolated from bovine mastitis. Research in Veterinary Science 37, 359-361 BAIRD-PARKER, A. C. (1966) Identification Methods for Microbiologists, Part A. Ed B. M. Gibbs and F. A. Skinner. New York, Academic Press. pp201-210 BOUJAAFAR, N., FRENEY, J., BOUVET, P. J. M. & JEDDI, M. (1990) Cell surface hydrophobicity of 88 clinical strains of Acinetobacter baumannii. Research in Microbiology 141, 477-482 CHRISTENSEN, G. D., SIMPSON, W. A., BISNO, A. L. & BEACHEY, E. H. (1982) Adherence of slime-producing strains of Staphylococcus epidermidis to smooth surfaces. Infection and Immunity 37, 318-326 DUGUID, J. P. (1951) The demonstration of bacterial cell capsules and slime. Journal of Pathologic Bacteriology 63, 673-685 EDEBO, L., KIHLSTROM, E., MAGNUSSON, K-E. & STENDAHL, O. (1980) Cell Adhesion and Motility. Ed A. G. Curtis and H. Pitts. Cambridge University Press. pp 124-132 HOGT, A. H., DANKERT, J. & FEIJEN, J. (1983) Encapsulation, slime production and surface hydrophobicity of eoagulase-negative staphylococci. FEMS Microbiology Letters 18, 211-215 JONSSON, P. & WADSTROM, T. (1984) Cell surface hydrophobicity of Staphylococcus aureus measured by the salt aggregation test (SAT). Current Microbiology 10, 203-210 KABIR, S. & ALI, S. (1983) Charaterization of surface properties of Vibrio cholerae. Infection and Immunity 39, 1048-1058

S t a p h y l o c o c c u s species

95

KLOOS, W. E. & SCHLEIFER, K. H. (1975) Simplified scheme for routine identification of human staphylococcus species. Journal of Clinical Microbiology 1, 82-88 LAMMLER, C. (1991) Identification of Staphylococcus hyicus using the ATB 32 H Staph system and conventional tests. Journal of Clinical Microbiology 29, 1221-1224 LA.MMLER, C. & BLOBEL, H. (1984) Colony-formation of staphylococci in fibronectin-soft-agar. Zentralblatt ffir Bakteriologie Mikrobiologie und Hygiene Series [A] 257, 1-5 MIORNER, H., JOHANSSON, G., & KRONVALL, G. (1983) Lipoteiehoic acid is the major cell wall component responsible for surface hydrophobicity of group A streptococci. Infection and Immunity 39, 336-343 NELLES, M. J., NISWANDER, C. A., KARAKAWA, W. W., VANN, W. F. & ARBEIT, R. D. (1985) Reactivity of type-specific monoclonal antibodies with Staphylococcus aureus clinical isolates and purified capsular polysaccharide. Infection and Immunity 49, 14-18 NORCROSS, N. L., & OPDEBEEK, J. P. (1983) Encapsulation of Staphylococcusmureus isolated from bovine milk. Veterinary Microbiology 8, 397-404 OFEK, I. & BEACHEY, E. H. (1980) Bacterial Adherence Receptors and Recognition, Series B, vol 6. Ed E. H. Beachey. London and New York, Chapman and Hall. ppl-31 OFEK, I., WHITNACK, E. & BEACHEY, E. (1983) Hydrophobic interactions of group A streptococci with hexadeeane droplets. Journal of Bacteriology 154, 139-145 OPDEBEEK, J. P., FROST, A. J., O'BOYLE, D. & NORCROSS, N. L. (1987) The expression of capsule in serum-soft agar by Staphylococcus aureus isolated from bovine mastitis. Veterinary Microbiology 13, 225-278 PASCUAL, A., FLEER, A., WESDERDAAL, N. A. C. & VERHOEF, J. (1986) Modulation of adherence of coagulase-negative staphylococci to Teflon catheters in vitro. European Journal of Clinical Microbiology 5, 518-522 PFALLER, M., DAVENPORT, D., BALE, M., BARRETT, M., KOONTZ, F. & MASSANARI, R. (1988) Development of the quantitive micro-test for slime production by coagulase-negative staphylococci. European Journal of Clinical Microbiology and Infectious Diseases 7, 30-33 REIFSTECK, F., WEE, S. & WILKINSON, B. J. (1987) Hydrophobicity-hydrophilicity of staphylococci.JournalofMedical Microbiology 24, 65-73 ROSENBERG, M. (1984)Bacterial adherence to hydrocarbons: a useful technique for studying cell-surface hydrophobicity. FEMS Microbiology Letters 22, 289-295 ROSENBERG, M. D., GUTNICK, D. & ROSENBERG, E. (1980) Adherence of bacteria to hydrocarbons: A simple method for measuring cell-surface hydrophobicity. FEMS Microbiology Letters 9, 29-33 ROZGONYI, F., KATALIN, R., SZITHA, K. R., HJERTEN, S. & WADSTROM, T. (1985) Standardization of salt aggregation test for reproducible determination of cell surface hydrophobicity with special reference to Staphylococcus species. Journal of Applied Bacteriology 59, 451-457 SCHAEG, W., BRI3CKLER, J. & BLOBEL, H. (1979) Improved method for the demonstration of protein A of Staphylococcus aureus. Zentralblatt ffir Bakteriologie, Mikrobiologie und Hygiene Series [A] 245, 442-449 SELTMANN, G., PAL, T. & TSCHAPE, H. (1986) Surface hydrophobicity of plasmid-carrying virulent Shigella flexneri and their avirulent variants. Journal of Basic Microbiology 26, 283-287 SIMPSON, W. A., OFEK, I., SARASOHN, C., MORRISON, J. C. & BEACHEY, E. H. (1980) Characteristics of the binding of streptococcal lipoteichoic acid to human oral epithelial cells. Journal of Infectious Diseases 141, 457-462 SUTRA, L., MENDOLIA, C., RAINARD, P. & POUTREL, B. (1990) Encapsulation of Staphylococcus aureus isolates from mastitic milk: relationship between capsular polysaccharide types 5 and 8 and colony morphology in serum-soft agar, clumpingfactor, teichoic acid, and protein A. Journal of Clinical Microbiology 28, 447-451

96

R. T. Greene, C. Lammler, M. Schmitt

VAN OSS, C. J. (1978) Phagocytosis as a surface phenomenon. ~lnnual Reviews in Microbiology 32, 19-39 WINBLAD, S. & ERICSON, S. (1973) Sensitized sheep red cells as a reactant for Staphylococcus aureus protein A. Acta Pathologica et Microbiologica, Scandinavica Section B, Microbiology 81, 150156 YOSHIDA, K. & EKSTEDT, R. D. (1968) Relation of mucoid growth of Staphylococcus aureus to clumping factor reaction, morphology in serum soft agar, and virulence. Journal of Bacteriology 96, 902-908 YOSHIDA, K., ICHIMAN, Y., UMEDA, A. & TAKEUCHI, S. (1988) Demonstration of capsule in a strain of Staphylococcus

hyicus by an electron microscope. Journal of Applied Bacteriology 65, 483-486 YOSHIDA, K., OHTOMO, T. & MINEGISHI, Y. (1977) Mechanism of compact-colony formation by strains of Staphylococcus aureus in serum soft agar. Journal of General Microbiology 98, 67-75 YOSHIDA, K. & TAKEUCHI, Y. (1970) Comparison of compact and diffuse variants of strains of Staphylococcus aureus. Infection and Immunity 2, 523-527 Received May 30, 1991 Accepted September 9, 1991