Physicochemical Surface Properties of Nonencapsulated and ...

APPLIED AND ENVIRONMENTAL MICROBIOLOGY, Nov. 1989, P. 2806-2814

0099-2240/89/112806-09$02.00/0 Copyright © 1989, American Society for Microbiology

Vol. 55, No. 11

Physicochemical Surface Properties of Nonencapsulated and Encapsulated Coagulase-Negative Staphylococci HENNY C.

VAN DER

MEI,1* PETER BROKKE,2 JACOB DANKERT,2 JAN FEIJEN,3 PAUL G. ROUXHET,4

HENK J. BUSSCHER' Laboratory for Materia Technicha, University of Groningen, Antonius Deusinglaan 1, 9713 AV Groningen,' Department of Hospital Epidemiology, University Hospital, Oostersingel 59, 9713 EZ Groningen, and Department of Chemical Technology, Twente University of Technology, 7500 AE Enschede,3 The Netherlands, and Faculte des Sciences Agronomiques, Universite' Catholique de Louvain, B-1348 Louvain-la-Neuve, Belgium4 AND

Received 22 May 1989/Accepted 28 August 1989

Cell surfaces of three nonencapsulated and three encapsulated coagulase-negative staphylococci were characterized by their surface free energies, zeta potentials, and elemental and molecular compositions. Surface free energies were calculated from contact angle measurements with various liquids. All six strains showed a high surface free energy (103 to 126 mJ * m-2), estimated from the concept of polar and dispersion components. However, the hydrogen-donating surface free energy parameter was zero for all nonencapsulated strains. The zeta potential profile measured as a function of pH in phosphate-buffered saline for the nonencapsulated strains was completely different from that of the encapsulated strains. X-ray photoelectron spectroscopy was used to determine the elements (0, C, N, P, and K) in the outer 2 to 5 nm of the freeze-dried cell surface and showed that the hydrophilic character of the staphylococci was related to oxygen (O/C ratio, approximately 0.52)- and phosphorus (P/C ratio, approximately 0.03)-containing groups. Both the elemental and molecular characterizations (done by infrared spectroscopy) pointed to the presence of polysaccharides and polypeptides on the cell surface of the nonencapsulated and encapsulated strains.

Infections associated with prosthetic implants or medical devices are often due to coagulase-negative staphylococci (5, 11-13). The pathogenicity of these staphylococci is related to their capacity to adhere to solid surfaces (9, 10, 25) and therefore also with the physicochemical properties of their cell surface (17). Some coagulase-negative staphylococci have a capsule. The presence of a capsule in conjunction with the various physicochemical surface characteristics involved in adherence, e.g., surface free energy (8, 19, 26), zeta potential (24, 37), hydrophobicity (15, 28), and elemental and molecular compositions (33-35), may affect the adherence of coagulase-negative staphylococci (16). Attempts have been made to relate the absence or presence of a capsule with the hydrophobicity of coagulasenegative staphylococci. Although the majority of encapsulated strains are hydrophobic in nature (16), some nonencapsulated strains are hydrophobic and others are hydrophilic. Extensive determinations of surface free energies, zeta potentials, and chemical surface compositions of a series of oral streptococci (33, 34) and brewery yeast strains (1, 2) have yielded a comprehensive characterization of the surfaces of these microorganisms. The aim of this study was to characterize the surface of three nonencapsulated and three encapsulated coagulasenegative staphylococcal strains by their surface free energy, zeta potential, and chemical composition to obtain further knowledge of the staphylococcal cell surface properties involved in promoting their adherence.

gether with their known surface properties. The isolation, identification, and growth conditions of the strains have all been described in detail previously (15-17). However, for completeness, the growth conditions will be repeated here. Strains from a sheep blood (5% [vol/vol]) agar (Oxoid Ltd., London, England) plate were inoculated in Trypticase soy broth (BBL Microbiology Systems, Cockeysville, Md.), cultured for 20 h at 37°C, and harvested by centrifugation (3,000 rpm, 4°C, 10 min; model 3E-1 centrifuge; Sigma, Osterode am Harz, Federal Republic of Germany). Clusters of bacteria were removed by ejecting the bacterial suspensions through a 3-jim-pore-size polycarbonate filter (Nuclepore Corp., Pleasanton, Calif.). For contact angle measurements and electrophoresis, cells were subsequently washed three times with phosphate-buffered saline (PBS; 8.1 mM Na2HPO4, 1.5 mM KH2PO4, 140 mM NaCl, and 3 mM KCl [pH 7.2]) and suspended in PBS. Only for X-ray photoelectron spectroscopy (XPS) and infrared spectroscopy (IR) were cells washed with and suspended in distilled water. For XPS and IR experiments, the bacterial pellets were frozen immediately after centrifugation in a glass flask in liquid nitrogen and lyophilized at -5°C in a Lyovac GT4 (Leybold Heraeus) (1). All experiments described were carried out in duplicate on two separate bacterial cultures. Contact angle measurements and estimation of cell surface free energies. Bacteria were deposited on membrane filters to produce a lawn of 50 to 100 stacked cells, suitable for contact angle measurements (8, 39). After a standard drying time of 30 min, plateau contact angles (8) were determined at 25°C with sessile droplets of water, formamide, diiodomethane, a-bromonaphthalene, and a series of water-n-propanol mixtures. Surface free energies were estimated by two approaches. First, the water, water-n-propanol, and ot-bromonaphthalene contact angles were least-square fitted (6, 7) to:

MATERIALS AND METHODS

Bacterial strains and growth conditions. Table 1 lists the six coagulase-negative staphylococci involved in this study to*

Corresponding author. 2806

PHYSICOCHEMICAL SURFACE PROPERTIES OF STAPHYLOCOCCI

VOL. 55, 1989

TABLE 1. Staphylococcal species included in this study and their known surface properties (16) % Adherence to: Strain

Encapsulationa

Anionic exchange resinb

Xylene'

1 S. epidermidis SL58d 2 S. epidermidis NCTC 100835 3 S. epidermidis NCTC 100894 4 S. hominis SL33d S S. saprophyticus SAP1 6 5. epidermidis NCTC 100892

_ -

27 70 47

75 3 74 93 6e 72

+

+ +

79 99

a +, Encapsulated; -, nonencapsulated. b Percentage of bacteria adhering to an anionic exchange resin in column chromatography. c Percentage of bacteria adhering to xylene in a xylene-water two-phase partitioning test. d These strains are slime producing. e S. saprophyticus SAPM constitutes an exception in the general tendency observed for encapsulated strains to be predominantly hydrophobic in nature

(16).

2(yb )1)2 + 2(gY {)1/2

(1) in which d and p denote the dispersion and polar surface free energy components of the bacteria and the liquid used, respectively, and le is the equilibrium spreading pressure (6). All liquid parameters are known through calibration experiments (7). This approach has proven to be valid for low-energy surfaces but also for high-energy surfaces on which spreading pressures may become significant (6) and to predict correctly a number of widely varying biological adhesion phenomena (8, 26, 29). Recently, Van Oss et al. (39, 40) extended equation 1 in an attempt to obtain more molecularly relevant data from contact angles. In this approach, which neglects spreading pressures, the pure liquid contact angles are inserted in the

(cos 0 + 1) -yl

=

-

following equation: (cos 0 + 1) 7, = y1'2 + 2(yb y+1/2 (2) to yield a Lifshitz-van der Waals _yLW surface free energy component of the bacteria and hydrogen-donating -y and hydrogen-accepting -y- surface free energy parameters, which together determine the acid-base surface free energy component of the bacteria according to: (3))1/2 -y = 2 (-yb . where _yL'w, -y+, and -yF are the respective surface free energy components of the liquids used, known from separate experiments (38, 40). Note that in this approach, which neglects spreading pressures, the subscript bv is used for the bacterial cell surface free energy, whereas otherwise only the subscript b is used. Zeta potential measurements. For zeta potential measurements, bacteria were suspended in PBS at a concentration of 107 cells per ml. The pH of the suspension was adjusted to vary over the range from 2 to 7 by the addition of HCI. Zeta potentials were measured with a precision of +2 mV with a model 501 Lazer Zee meter (PenKem, Bedford Hills, N.Y.), which uses scattering of incident laser light to enable detection of the bacteria at a relatively low magnification. The method is based on the application of the Smoluchowski equation (14). XPS. After lyophilization, the bacterial powder was placed in a stainless steel trough and pressed. Six troughs could be 2(^

LW

L

XW)1/2 + 2(wy

2807

inserted simultaneously in the chamber of the spectrometer, a Vacuum Generators ESCA 3 Mk II instrument equipped with a Tracor Northern TN 1710 signal averager for signalto-noise ratio enhancement. A magnesium anode was used for X-ray production (14 kV, 20 mA). After a scan of the overall spectrum, peaks were recorded in the following order: C1s (5 min), O1, (5 min), N1l (15 min), K2S (25 min), P2p (25 min), and C1s (5 min) again. The area under each peak after linear background subtraction was used for calculation of the peak intensities, yielding the elemental surface concentration ratios O/C, N/C, K/C, and P/C with the sensitivity factors determined by Wagner et al. (41). The carbon peak was decomposed by a least-square fitting program into three gaussian components at 285.0 eV (representative for C-C and C-H bonds), 286.6 eV (C-0 and C-N bonds), and 288.4 eV ([C=O]-NH bonds) (27) by imposing a constant full width at half-maximum (2.15 eV) (1) for the components. The component set at 286.6 eV was used to determine the exact binding energies of all the peaks. Similarly, the oxygen peak was decomposed into two components representative for oxygen involved in C-OH bonds (533.2 eV) and in other functional groups (531.3 eV). For the decomposition of the oxygen peak a full width at half-maximum of 2.6 eV was imposed (20). IR. After lyophilization, the bacterial powder was combined with KBr (1:50 by weight), ground for 1 min, and pressed to a pellet. Transmission infrared absorption spectra were recorded on a Nicolet Instruments Fourier transform MX-S infrared spectrometer. The effective spectral resolution and wave number accuracy were 4 cm-1 and 0.01 cm-1, respectively. Five hundred scans were measured and averaged for each sample with a KBr pellet as a reference. The areas of the most important absorption bands were determined by integration after linear background subtraction and normalized with respect to the area of the CH stretch absorption region around 2,930 cm-1. All experimental methods used are summarized in Table 2. All these techniques, except for the IR technique, are known in physicochemistry as being surface sensitive. However, they can never probe the properties of the mathematically defined dividing plane between two phases but always yield information about a certain volume in the surface region. This so-called "depth of information" varies for the four techniques used in this study (Table 2). It has been shown that all four techniques give valuable macroscopic information on bacterial cell surfaces (33-35). Moreover, XPS and IR data can be used to explain cell surface free energies and zeta potentials, both relevant for adherence, on the basis of the average elemental and molecular compositions of the cell surface. RESULTS Contact angles measured on lawns of staphylococci (Table 3) demonstrated higher formamide contact angles for nonencapsulated strains than for encapsulated strains. These differences were much higher than the reproducibility of the measurements between two separate bacterial cultures, typically being +5 degrees. By consequence of the higher formamide contact angles for the nonencapsulated strains, the acid-base surface free energy component yAb for these strains was zero because of a hydrogen-donating surface free energy parameter of zero. The surface free energy approach based on dispersion and polar components did not reveal any systematic differences between the strains and showed that all strains involved possessed a high cell surface free energy.

2808

VAN DER

APPL. ENVIRON. MICROBIOL.

MEI ET AL.

TABLE 2. Techniques used in this study for the physicochemical characterization of coagulase-negative staphylococcal cell surfaces Technique

Parameter(s) derived

Contact angle measurements Electrophoresis XPS

Bacterial cell surface free energya Zeta potentials and isoelectric points Elemental surface concentration ratios O/C, N/C, K/C, and P/C; details on the molecular binding of the elements Molecular composition

IR

information

Reference(s)

0.4-0.5 nm b 2-5 nm

4, 8, 39 14, 18 1, 3

Micron rangec

32

a Two approaches were used to estimate surface free energies from measured contact angles: one based on the concept of polar and dispersion components (6, 7) and the other based on a new approach, separating surface free energies into acid-base and Lifshitz-van der Waals components (38, 40). b _, Zeta potentials are measured at the so-called plane of shear. The distance between the surface and the plane of shear is approximately 1.0 nm for PBS but can be greatly increased by the presence of surface appendages. c Although in general IR is considered a bulk technique, we have demonstrated that it reflects differences between the surfaces of various freeze-dried microorganisms (35).

Figure 1 shows the pH dependences of the zeta potentials of the strains in PBS. Zeta potentials of duplicate cultures coincided within 3 mV, the precision of an individual measurement. The zeta potential profiles of the encapsulated staphylococci were completely different from those of the nonencapsulated staphylococci. XPS spectra showed that the outer 2 to 5 nm of the cell wall predominantly consisted of carbon, oxygen, nitrogen, potassium, and phosphorus, detected at binding energies of 286.6, 532.8, 400.4, 378.1, and 134.0 eV, respectively. It should be noted, however, that all three encapsulated strains showed a slightly higher (0.3 eV) binding energy of the N1s peak than did the nonencapsulated strains. Neither the elemental surface concentration ratios nor the decompositions of the C1s and Ols peaks (Table 4) revealed systematic differences between the strains. Figure 2 demonstrates the principle of peak decomposition of a C1l peak and an 01s peak for a nonencapsulated strain and an encapsulated strain. Whereas the positions as well as the full widths at half-maximum of the peaks were identical for both cultures, a 10% variation in the compositional data was evident for two separate cultures of the same strain. The IR spectra of the strains were surprisingly simple (Fig. 3); the most important absorption bands were located at 2,930 cm-1 (CH2-CH3 band), 1,654 cm-1 (amide I band: C=O stretch in proteins), 1,542 cm-' (amide II band: N-H bending in proteins), and 1,237 cm-' and 1,070 cm-' (phosphates and sugars). In addition, most of the strains showed another peak at 1,742 cm-' (carboxyl stretch) which was, however, very small for SL58 and SAP1. The absorption band ratios with respect to the CH absorption region around 2,930 cm-1 are compiled in Table 5. The magnitudes of the

absorption band ratios

were

reproducible within 10% when

two cultures of the same bacteria were compared.

DISCUSSION In this study we determined a variety of physicochemical surface properties of nonencapsulated and encapsulated coagulase-negative staphylococci: the cell surface free energy, zeta potential, elemental surface composition, and presence of specific molecular groups. The various properties determined will now be successively discussed. Surface free energies. All investigated nonencapsulated and encapsulated staphylococcal strains had a hydrophilic character, according to the concept of dispersion and polar surface free energy components, taking into account spreading pressures. Hogt et al. (16), however, using a xylene test for determining bacterial hydrophobicities, observed a large variation among the hydrophobicities of the six strains used in this study, with only Staphylococcus epidermidis NCTC 100835 and S. saprophyticus SAP1 being extremely hydrophilic (Table 1). This result supports a previous conclusion of both Mozes and Rouxhet (21) and Van der Mei et al. (36) that there is in general no correlation between bacterial hydrophobicities derived from contact angles and those derived from two-phase partitioning tests. Apparently each type of test determines the hydrophobicity of the cell surface at a different level. The hydrogen-donating surface free energy parameter and, consequently, the acid-base surface free energy component yB in the approach of Van Oss et al. (38, 40) were zero for all three nonencapsulated strains. Thus, contact angle measurements and subsequent surface free energy

TABLE 3. Selection of measured contact angles 0 on lawns of staphylococci and surface free energies estimated by two approachesa Contact angle 0 (degrees) with:

strain

Nonencapsulated 1 2 3

Surface free energy (mJ m-2)

Water

Formamide

Diiodomethane

oa-Bromonaphthalene

-yb

-Y

'Yb

-L'r'

22 27 27

45 54 43

49 50 46

35 37 13

37 36 42

79 77 79

115 113 121

36 35 40

37 20 19

28 19 10

57 49 73

32 12 50

38 44 30

65 83 88

103 126 118

34 39 26

-Y

Ybv

Yb_v

Ybv

0 0 0

36 35 40

73 80 66

0 0 0

16 14 33

50 53 59

40 53 49

2 1 6

Encapsulated 4 5 6

y6

a One approach was based on the concept of dispersion -y and polar surface free energy components and accounted for spreading pressures (6, 7). The other approach was based on the concept of Lifshitz-van der Waals yr" and acid-base -yb" surface free energy components. 4B~was subsequently separated into hydrogen-accepting -vy and hydrogen-donating -y', parts (38, 40).

VOL. 55, 1989


t(mV)

2809

g (mV) +6 I

0

-6

-10

-14

-18

1 -22J -22J FIG. 1. Zeta potentials of nonencapsulated (1 to 3) and encapsulated (4 to 6) coagulase-negative staphylococci in PBS as a function of pH. The bar denotes the experimental precision. For nomenclature, see Table 1.

calculations done by the latter method might present an alternative for the unreliable India ink staining of cells and more tedious immunological methods (31) used for identifying the presence of a capsule, although research involving more strains is needed to verify this suggestion. Zeta potentials. The zeta potential is determined by the nature and number of ionogenic groups on the surface and by the properties of the suspending electrolyte, e.g., pH, ionic strength, or presence of surfactants. When the pH dependence of zeta potentials is used it is possible to identify the ionogenic groups at the bacterial cell surface in more detail. The nonencapsulated strains demonstrated a completely different profile than the encapsulated strains did, with the profiles of the encapsulated strains being typical for most

microorganisms (18, 30). The observed less negative zeta potentials of encapsulated strains at a low pH were presumably caused by a higher concentration of positively charged amino groups on the surface with respect to partly protonated phosphate groups and fully protonated carboxyl groups. At pH values of 6 to 7 the zeta potentials of the encapsulated strains became less negative again. The zeta potentials of nonencapsulated S. epidermidis SL58 scattered around 0 mV over the range of pH values tested. Although these low zeta potentials were in accordance with the low adherence of SL58 to an anionic exchange resin observed previously (Table 1), no correlation was established between the zeta potentials and the adherence to an anionic exchange resin of the other strains.

TABLE 4. XPS analysis of staphylococcal strains showing elemental surface concentration ratios and decomposition of the Cl, and 015 peaks Decomposition of:

Elemental surface concn ratio

Strain

O/C

N/C

K/C

01, peakb

Cl, peaka

P/C 285.0 eV

286.6 eV

288.4 eV

531.3 eV

533.2 eV


0.495 0.522 0.577

0.197 0.153 0.149

0.021 0.016 0.010

0.031 0.039 0.041

0.340 0.367 0.325

0.467 0.456 0.523

0.195 0.178 0.153

0.332 0.357 0.380

0.668 0.643 0.621

0.452 0.544 0.550

0.175 0.150 0.162

0.013 0.022 0.013

0.017 0.035 0.041

0.264 0.329 0.301

0.477

0.260

0.478 0.490

0.193 0.210

0.373 0.341

0.627 0.659 0.633

Encapsulated 4 5 6

0.368

a The carbon peak was decomposed into three components at 285.0, 286.6, and 288.4 eV, which were attributed to carbon involved in C-C- and -C-H, --C-0-and -C-N-, and -(C=O)-NH- bonds, respectively (1). b The oxygen peak was decomposed in two components at 531.3 and 533.2 eV, latter component being attributed to oxygen involved in -C-0 bonds (20).

2810


VAN DER MEI ET AL.

1:SL58 Relative intensity

Relative intensity

285

530

290 bindingenergy (ev)

535 bindingenergy (ev)

4:SL33 Relative intensity

Relative

intensity

285

FIG. 2. Carbon Cl, and oxygen experimentally measured Cl, and determined by decomposition.

01is O1,

535 530 290 bindingenergy (ev) bindingenergy (ev) peaks of a nonencapsulated strain (1) and an encapsulated strain (4). The solid lines represent the peaks; the dashed lines represent carbon and oxygen involved in specific molecular bonds, as

XPS analyses. The elements detected on the bacterial cell surface of the nonencapsulated and encapsulated staphylococci were essentially the same (Table 4). The binding energy for the N1s peak of the encapsulated strains was slightly higher than that for the N1s peak of the nonencapsulated strains, probably indicating a higher number of NH4+-containing components at the cell surface (1), although it should be admitted that the differences in binding energies mentioned were hardly significant. The elemental surface concentration ratios of the nonencapsulated and encapsulated strains were rather similar and can be used to model the chemical composition of coagulase-negative staphylococcal surfaces. The O/C surface concentration ratio was clearly too low for carbohydrates (theoretical O/C concentration ratio, approximately 1.0) but was, on the other hand, too high to represent a fully protein-covered surface (theoretical O/C concentration ratio, approximately 0.34), a theory which was confirmed by the relatively low N/C concentration ratio (theoretically approximately 0.27 for

proteins). In a model depicting the cell surface as consisting predominantly of proteins and carbohydrates, this would indicate that the surface layer is, on an average, approximately 61% composed of proteins and 32% composed of carbohydrates in both the encapsulated and nonencapsulated strains. The detection of sizeable amounts of phosphorus and potassium also points to the presence of (lipo) teichoic acid and other lipidlike components. IR spectra. Nichols et al. (23) discriminated two different groups of microorganisms with IR spectroscopy. One group, including Pseudomonas fluorescens, Desulfovibrio gigas, Staphylococcus aureus, Clostridium perfringens, Escherichia coli, Methylobacterium organophilum, and Methylosinus trichosporium, did not show a carboxyl stretch band around 1,740 cm-'. This band was found in Bacillus subtilis, Methylobacterium organophilum, and Nitrobacter winogradskyi. Our strains can be classified as belonging to the second group of microorganisms distinguished by Nichols et al.

(23).


VOL. 55, 1989

2811

u) U)

Absorbance 1.35

N

A

-

E

U)

N

0

N

E9

a

0

CO) 0

CY

CN

0.35

Ui -

sugar region

-0.25-J i3600

2

1

2850

2100

1350

600

Wavenumber (cm-i)

U) U) co 7

Absorbance 1.35

B

-

N

U)

C0

0.35

N

-

sugar

region

-0.25 -

I

I

3600

2850

1.

2100

1350

600

Wavenumber (cm-1)

FIG. 3. IR spectra of a nonencapsulated strain (1) (A) and an encapsulated strain (4) (B).

The cell wall of the staphylococcal strains studied possesses a thick peptidoglycan layer characteristic of grampositive bacteria (11) and representing up to 60% of the cell wall weight (22). This explains the similarity within our spectra, although small but significant differences existed in the absorption band ratios normalized with respect to the TABLE 5. IR absorption band ratios with respect to the CH absorption region around 2,930 cm-' for staphylococcal strains Ratioa Strain AmI/CH

AmII/CH

PICH

PII/CH

8.3 7.2 7.9

2.5 2.6 2.9

3.2 4.4 4.1

7.6 8.3 7.6

7.9 8.7 7.6

3.6 2.8 2.6

2.5 3.2 4.0

6.0 7.6 7.8


Encapsulated 4 5 6

a CH, CH2-CH3 band; AmI, amide I band (C=O stretch in proteins); AmII, amide II band (N-H bending in proteins); PI, phosphate band; Pll, phosphate-sugar band.

CH stretch region around 2,930 cm-1 (Table 5). These minor differences reflect the variability in the amounts of proteins (amide I band/CH2-CH3 band and amide II band/ CH2-CH3 band) and phosphates and carbohydrates (phosphate band/CH2-CH3 band and phosphate-sugar band/ CH2-CH3 band) at the cell surface. Relationships between various physicochemical parameters. To explain the physical properties of the strains on the basis of their chemical compositions, we sought possible relationships between the parameters measured. An increase in the surface free energy Yb of the coagulasenegative staphylococci was accompanied by an increase in the O/C concentration ratio and a decrease in the N/C concentration ratio (Fig. 4). Although the shift in binding energies towards higher values for the encapsulated staphylococci was extremely small, it is interesting to note that this shift was concurrent with the non-zero hydrogen-donating surface free energy parameter of these strains. No clear correlations were observed between the elemental surface concentration ratios of the nonencapsulated and encapsulated strains versus the zeta potentials. Principally, correlations between XPS data and IR data cannot be expected because of the much larger depth of

2812


VAN DER MEI ET AL.

O/C

O/C 0.60

-

0.6 -

0.55

0)

-

0D 0.4

0D 0

0.50

0

00

0.2 -

0.45

A1 0.00 0 O

0 120

110

100

0

130

40

80

120

130

Y b(mJ. m -2)

F b(mJ.m-2)

N/C

N/C

0D

0.20 -

0D

0.20-

0.15

69

0.18

0)

0.10

0

0.16 -

0

0.05 0.14

0.00

I

0

100

1

1

1

110

120

130

}' b (mJ.m- 2)

0

40

80

120

130

Yb (m .m-2)

FIG. 4. Elemental surface concentration ratios as a function of surface free energy for six strains of coagulase-negative staphylococci. For nomenclature, see Table 1. Symbols without numerals represent previously published data for oral streptococci (33).

information of IR. However, a definitive relationship was found between the P/C surface concentration ratio and the phosphate/CH2-CH3 absorption band ratio (Fig. 5), confirming not only that this band at 1,237 cm-' is due to phosphate but also that IR measurements reflect the phosphate concentration at the surface of the staphylococcal strains. Similarly, although less definitively, a relationship was obvious between the fraction of carbon atoms involved in peptide linkages and the amide I/CH2-CH3 absorption band ratio (C=0 stretch in proteins). However, the expected relationships between N/C and the amide II/CH2-CH3 absorption band ratio (N-H bending in proteins) and between the fraction of carbon atoms involved in polysaccharides (C-0 bonds) and the phosphate-sugar/CH2--CH3 absorption band ratio did not exist within the staphylococci investigated. Comparison with oral streptococci. Recently, we used the same variety of techniques to obtain a physicochemical characterization of oral streptococcal cell surfaces (33-35). Within our collection of oral streptococcal strains we observed the following relationships between the various parameters. (i) High surface free energies were always accom-

panied by high O/C and- low N/C surface concentration ratios and vice versa (Fig. 4). (ii) Isoelectric points increased linearly with N/C and showed an inverse relationship with O/C. (iii) A good relationship existed between XPS data and the IR absorption band ratios (Fig. 5). No information could be extracted from the P/C surface concentration ratio and the phosphate/CH2-CH3 IR absorption band ratios, and the oral streptococci did not show a carboxyl stretch band around 1,740 cm-'. The above-mentioned relationships were clear, despite the fact that all parameters were measured in a different state of (de)hydration of the cell surface, namely, zeta potentials were measured on hydrated cells, surface free energies were determined on partly dehydrated surfaces, and the elemental and molecular compositions were determined on freeze-dried cells. Compared with the staphylococci studied in this paper, our collection of streptococcal strains involved a much wider range in surface free energies (37 to 117 mJ m-2) and N/C (0.074 to 0.129) and O/C (0.312 to 0.495) surface concentration ratios, whereas P/C surface concentration ratios were relatively low (0.006 to 0.010) and potassium was only occasionally detectable and never quantified. The zeta po-

VOL. 55, 1989


N/C 0.20

P/C

0

-

0.10

0

8

-

0D 0D

0.03 -

0 0.02 -

0E

0 C)

0.05

0.00 v0.0 1.5

0.01 -

r

I

I

I

2.0

2.5

3.0

0.00 0

3.5

1

2

3

Am I/CH

N-H -

-

0.20

-

0.15

-

4

P l/CH

-C-0-

C=O

0.25

023

0.04 -

-

0.15

2813

0.6

-

0) 0.5 -

0)0

( 0.4 -

0

00

0.3

0 0.10

ooO

0.0 0.0 5.55

0

-

0

00

0

0.0< OA

I

I

1

6.5

7.5

8.5

9.0

Am I/CH

0.0

4.0

5.0

6.0

7.0

8.0

P ll/CH

FIG. 5. N/C and P/C surface concentration ratios and the fractions of carbon atoms involved in (C=O)-N-H bonds and C-0 bonds as a function of the amide II/CH2-CH3 (Am II/CH), phosphate/CH2-CH3 (P I/CH), amide I/CH2-CH3 (Am I/CH), and phosphatesugar/CH2-CH3 (P II/CH) IR absorption band ratios, respectively. For nomenclature, see Table 1. Symbols without numerals represent previously published data for oral streptococci (33).

tential profiles of the encapsulated staphylococcal strains were completely identical to those of the oral streptococci, with an isoelectric point of approximately 2.0. Such a low isoelectric point is, on basis of our experiments with streptococci, in accordance with the high O/C and low N/C surface concentration ratios, although admittedly the staphylococci contain much more nitrogen than do the streptococci.

In summary, a clear distinction can be made between the staphylococci and streptococci on the basis of the absolute values of their physicochemical surface properties. In addition,- the relationships between the various parameters are similar within both species and exist even when the results for staphylococci and streptococci are combined. ACKNOWLEDGMENTS We are greatly indebted to Marjon Schakenraad-Dolfing for the preparation of this manuscript. The financial support of Service de Programmation de la Politique Scientifique-Belgium (Concerted Action, Physical Chemistry of Interfaces and Biotechnology) is gratefully acknowledged. LITERATURE CITED 1. Amory, D. E., M. J. Genet, and P. G. Rouxhet. 1988. Application of XPS to surface analysis of yeast cells. Surf. Interface

Anal. 11:478-486. 2. Amory, D. E., N. Mozes, M. P. Hermesse, A. J. Leonard, and P. G. Rouxhet. 1988. Chemical analysis of the surface of microorganisms by X-ray photoelectron spectroscopy. FEMS Microbiol. Lett. 49:107-110. 3. Andrade, J. D. 1985. X-ray photoelectron spectroscopy (XPS), p. 105-195. In J. D. Andrade (ed.), Surface and interfacial aspects of biomedical polymers, vol. 1. Surface chemistry and physics. Plenum Publishing Corp., New York. 4. Bain, C. D., and G. M. Whitesides. 1988. Molecular-level control over surface order in self-assembled monolayer films of thiols on gold. Science 240:62-63. 5. Bayston, R. 1984. A model of catheter colonization in vitro and its relation to clinical catheter infections. J. Infect. 9:271-276. 6. Busscher, H. J., G. A. M. Kip, A. van Silfhout, and J. Arends. 1987. Spreading pressures of water and n-propanol on polymer surfaces. J. Colloid Interface Sci. 114:307-313. 7. Busscher, H. J., A. W. J. van Pelt, H. P. de Jong, and J. Arends. 1983. Effect of spreading pressure on surface free energy determination by means of contact angle measurements. J. Colloid Interface Sci. 95:23-28. 8. Busscher, H. J., A. H. Weerkamp, H. C. van der Mei, A. W. J. van Pelt, H. P. de Jong, and J. Arends. 1984. Measurement of the surface free energy of bacterial cell surfaces and its relevance for adhesion. Appl. Environ. Microbiol. 48:980-983. 9. Christensen, G. D., W. A. Simpson, A. L. Bisno, and E. H. Beachey. 1982. Adherence of slime-producing strains of Staph-

2814

10.

11. 12. 13.

14. 15. 16. 17.

18. 19.

20.

21. 22. 23.

24.

25.

VAN DER MEI ET AL.

ylococcus epidermidis to smooth surfaces. Infect. Immun. 37: 318-326. Christensen, G. D., W. A. Simpson, A. L. Bisno, and E. H. Beachey. 1983. Experimental foreign body infections in mice challenged with slime-producing Staphylococcus epidermidis. Infect. Immun. 40:407-410. Dankert, J., A. H. Hogt, and J. Feien. 1986. Biomedical polymers: bacterial adhesion, colonization, and infection. Crit. Rev. Biocompat. 2:219-301. Fulginity, V. A. 1984. Staphylococcus epidermidis septicemia in children: an emerging and difficult problem. J. Am. Med. Assoc. 252:1054. Gristina, A. H., and J. W. Costerton. 1984. Bacterial adherence to biomaterials: the clinical significance of its role in sepsis. Biomaterials Trans. 7:175. Hiemenz, P. C. 1977. Electrophoresis and other electrokinetic phenomena, p. 452-487. In J. J. Lagowski (ed.), Principles of colloid and surface chemistry. Marcel Dekker, Inc., New York. Hogt, A. H., J. Dankert, J. A. de Vries, and J. FeUen. 1983. Adhesion of coagulase-negative staphylococci to biomaterials. J. Gen. Microbiol. 129:2959-2968. Hogt, A. H., J. Dankert, and J. FeUen. 1985. Adhesion of Staphylococcus epidermidis and Staphylococcus saprophyticus to a hydrophobic biomaterial. J. Gen. Microbiol. 131:2485-2491. Hogt, A. H., J. Dankert, C. E. Hulstaert, and J. Feijen. 1986. Cell surface characteristics of coagulase-negative staphylococci and their adherence to fluorinated polyethylenepropylene. Infect. Immun. 51:294-301. James, A. M. 1979. Electrophoresis of particles in suspension. Surf. Colloid Sci. 11:174-177. Minagi, S., Y. Miyake, K. Inagaki, H. Tsuru, and H. Suginaka. 1985. Hydrophobic interaction in Candida albicans and Candida tropicalis adherence to various denture base resin materials. Infect. Immun. 47:11-14. Mozes, N., A. J. Leonard, and P. G. Rouxhet. 1988. On the relations between the elemental surface composition of yeasts and bacteria and their charge and hydrophobicity. Biochim. Biophys. Acta 945:324-334. Mozes, N., and P. G. Rouxhet. 1987. Methods for measuring hydrophobicity of microorganisms. J. Microbiol. Methods 6: 99-112. Naumann, D., G. Barnickel, H. Bradaczek, H. Labischinski, and P. Giesbrecht. 1982. Infrared spectroscopy, a tool for probing bacterial peptidoglycan. Eur. J. Biochem. 125:505-515. Nichols, P. D., J. M. Henson, J. B. Guckert, D. E. Nivens, and D. C. White. 1985. Fourier transform-infrared spectroscopic method for microbial ecology: analysis of bacteria, bacteriapolymer mixtures and biofilms. J. Microbiol. Methods 4:79-94. Olsson, J., P. 0. Glantz, and B. Krasse. 1976. Surface potential and adherence of oral streptococci to solid surfaces. Scand. J. Dent. Res. 84:240-242. Peters, G., R. Locci, and G. Pulverer. 1981. Microbial colonization of prosthetic devices. II. Scanning electron microscopy of naturally infected intravenous catheters. Zentralbl. Bakteriol. Abt. 1 Orig. B Hyg. Krankenhaushyg. Betriebshyg. Praev. Med. 173:293-299.


26. Pratt-Terpstra, I. H., A. H. Weerkamp, and H. J. Busscher. 1988. On a relation between interfacial free energy dependent and non-interfacial free energy dependent adherence of oral streptococci to solid substrata. Curr. Microbiol. 16:311-313. 27. Ratner, B. D., and B. J. McElroy. 1985. Electron spectroscopy for chemical analysis: applications in the biomedical sciences, p. 108-140. In R. M. Gendrau (ed.), Spectroscopy in the biomedical sciences. CRC Press, Inc., Boca Raton, Fla. 28. Rosenberg, M., and S. Kjelleberg. 1986. Hydrophobic interactions: role in bacterial adhesion. Microb. Ecol. 9:353-393. 29. Schakenraad, J. M., H. J. Busscher, C. R. H. Wildevuur, and J. Arends. 1988. Thermodynamic aspects of cell spreading on solid substrata. Cell Biophys. 13:75-91. 30. Small, D. A., and N. F. Moore. 1987. Measurement of surface charge of baculovirus polyhedra. Appl. Environ. Microbiol. 53:598-602. 31. Tojo, M., N. Yamashita, D. A. Goldmann, and G. B. Pier. 1988. Isolation and characterization of a capsular polysaccharide adhesin from Staphylococcus epidermidis. J. Infect. Dis. 157: 713-722. 32. Van der Maas, J. H. 1972. Basic infrared spectroscopy, 2nd ed. Pitman Press, Bath, United Kingdom. 33. Van der Mei, H. C., A. J. Ieonard, A. H. Weerkamp, P. G. Rouxhet, and H. J. Busscher. 1988. Properties of oral streptococci relevant for adherence: zeta potential, surface free energy and elemental composition. Colloids Surf. 32:297-305. 34. Van der Mei, H. C., A. J. Leonard, A. H. Weerkamp, P. G. Rouxhet, and H. J. Busscher. 1988. Surface properties of Streptococcus salivarius HB and nonfibrillar mutants: measurement of zeta potential and elemental composition with X-ray photoelectron spectroscopy. J. Bacteriol. 170:2462-2466. 35. Van der Mei, H. C., J. Noordmans, and H. J. Busscher. 1989. Molecular surface characterization of oral streptococci by Fourier transform infrared spectroscopy. Biochim. Biophys. Acta 991:395-398. 36. Van der Mei, H. C., A. H. Weerkamp, and H. J. Busscher. 1987. A comparison of various methods to determine hydrophobic properties of streptococcal cell surfaces. J. Microbiol. Methods 6:277-287. 37. Van Loosdrecht, M. C. M., J. Lyklema, W. Norde, G. Schraa, and A. J. B. Zehnder. 1987. Electrophoretic mobility and hydrophobicity as a measure to predict the initial steps of bacterial adhesion. Appl. Environ. Microbiol. 53:1898-1901. 38. Van Oss, C. J., M. K. Chaudhury, and R. J. Good. 1988. Interfacial Lifshitz-van der Waals and polar interactions in macroscopic systems. Chem. Rev. 88:927-941. 39. Van Oss, C. J., and C. F. Gillman. 1972. Phagocytosis as a surface phenomenon. I. Contact angles and phagocytosis of non-opsonized bacteria. J. Reticuloendothel. Soc. 12:283-292. 40. Van Oss, C. J., R. J. Good, and M. K. Chaudhury. 1986. The role of van der Waals forces and hydrogen bonds in "hydrophobic interactions" between biopolymers and low energy surfaces. J. Colloid Interface Sci. 111:378-390. 41. Wagner, C. D., L. E. Davis, M. V. Zeller, J. A. Taylor, R. H. Raymond, and L. H. Gale. 1981. Empirical atomic sensitivity factors for quantitative analysis by electron spectroscopy for chemical analysis. Surf. Interface Anal. 3:211-225.