FEMS MicrobiologyLetters 72 (1990)275-280 Publishedby Elsevier

275

FEMSLE04210

Purification and characterization from human parotid secretion of a peptide which inhibits hemagglutination of Bacteroides gingivalis 381 Yukitaka M u r a k a m i t, Atsuo A m a n o i, Masaru Takagaki 1 Satoshi Shizukuishi 1 Akira T s u n e m i t s u i and Saburoo A i m o t o 2 i Department of Preventive Dentist@, Facul(v of Dentisto'. and : Institute for Protein Research. Osaka University, Ovaka, Japan

Received30 May 1990 Revisionreceived25 June 1990 Accepted 10 July 1990 Key words: Salivary histidine-rich peptide; Hemagglutination; Bacteroides gingivalis

1. SUMMARY A peptide from human parotid secretion which inhibited hemagglutination of Bacteroides gingivalis 381 was purified by uhrafiltration followed by DEAE-Scphadex A-25 column chromatography and by gel filtration on Sephadex G-25, and then by reversed-phase HPLC. The complete amino acid sequence of the peptide, determined by automated Edman degradation was as follows: LysPhe-His-Glu-Lys-His-His-Sez-His-Arg-Gly-Tyr. The peptide contained 12 residues and the charged amino acids predominated with 4 histidine, 2 lysine, 1 arginine and 1 glutamic acid residues, thus being a histidine-rich peptide. The peptide was an active inhibitor of the hemagglutinating activity of B, gingivalis. Specific binding of tritium-labeled peptide to B. gingwalis cells was demonstrated. These results suggest that the histidine-rich peptide may function as a binding

Correspondence to: A. Tsunemitsu,Department of Preventive

Dentistry, Facultyof Dentistry,Osaka University,1-8 Yamadaoka, Suita, Osaka 565, Japan.

domain for the hcmagglutinins of B. gingivalis during agglutination.

2. INTRODUCTION High proportions of Bacteroides gingivalis have been found in periodontal pockets. It has been recognized that this organism can agglutinate erythrocytes, bind to sulcular epithelial cells and saliva-coated hydroxyapatite, and aggregate with a number of oral Gram-positive organisms [1-4]. Thus, it may play an important role in the etiology and pathogenesis of periodontal disease. However, the molecular interactions which permit this organism to adhere to the respective surfaces are poorly understood. Hemagglutinating activity of B. gingivalis has been investigated as one of the parameters that affect the adherence of this bacterium to cells of oral tissues. Data of some investigators have suggested that the ability of B. gingivalis to agglutinate erythrocytes is probably mediated by fimbriae [1,2], but there is conflicting evidence that a nonfimbrial hemagglutinin may be

0378-1097/90/$03.50© 1990 Federationof European MicrobiologicalSocieties

276 present on the cell surface of the bacterium [4-6]. It is known that the hemagglutinating activity of B. gingivalis is inhibited by whole saliva, especially hlgh-molecular-mass components of saliva [1,2]. However, little attention has been paid to the inhibitory substances such as low-molecular-mass peptides on hemagglutinating activity. This study deals with the purification, primary structure and some properties of a hemagglutination of B. gingivalis-:,nhibiting peptide (HIP) from human parotid secretion. 3. MATERIALS AND METHODS

3.1. Collection of human parotid saliva Parotid saliva was collected from healthy human adults, 20 to 30 years old, according to the method of Keene [7]. 3.2. Bacterial strains and growth condition B. gingivalis 381 was a gift from stock strains at the Research Laboratory of Oral Biology, Sunstar Inc., Osaka, Japan. The bacterial growth conditions were as in our previous procedure [8]. 3.3. Hemagglutination assay The hemagglutinating activity was determined using human O-type erythrocytes with a microtiter plate as described in our previous method [8]. Titer was expressed as the reciprocal of the highest dilution showing positive agglutination. The bacterial suspension was diluted in PBS to contain 4-times the minimal hemagglutinating titer. 3.4. Hemagglutination inhibition test For screening hemagglutination inhibition, 25 ~tl samples of fractions o(-,~ained by anion-exchange chromatography and gel filtration were mixed with 25/tl of bacterial suspension (containing 4-times the titer of hemagglutinating activity) and 50 ~tl of 2~ erythrocyte suspension, Hemagglutination was recorded after 90 rain. A reaction mixture that did not contain any sample was used as a control, Fractions having inhibitory activity were pooled. 3,5. Amino acid analysis Amino acid analysis was performed by a L-8500 amino acid analyzer (Hitachi Ltd,, Tokyo, Japan)

after hydrolysis for 24 h with 6 N HCI at l l 0 ° C in an evacuated sealed tube.

3.6. Peptide sequencing The sequence analysis of purified HIP was carried out on a 477A protein sequencer (Applied Biosystems Inc., Foster, CA, U.S.A.) attached to a 120A analyzer. 3. 7. Mass measurement The mass number of purified HIP was determined by FAB mass spectrometry using JMSHX100 (Jeol Ltd., Tokyo, Japan) attached to a JMA-3100 mass data system. 3.8. Peptide synthesis A HIP synthesis was carried out by a solidphase method using a peptide synthesizer 430A (Applied Biosystems Inc.) employing the Bocamino acid anhydride method. 3.9. Preparation of radioactive HIP Tritium labeling of the HIP was carried out using N-[ethyl-2-H]ethylmaleimide (NEM) according to the method described by Leslie et al. [9], Cys-HIP (Cys-Lys-Phe-His-Ghi-Lys-His-HisSer-His-ArgoGly-Tyr) was synthesized by a solidphase method. This peptide also exhibited inhibition of hemagglutination at 6.3 pM concentration (data not shown). The Cys-peptide and the HIP were equivalent in inhibitory activity. The Cyspeptide (0.18 pmol) was dissolved in 500 lal of 50 mM phosphate buffer (pH 7.0), and 0.023 pmol (1 mCi) of labeled NEM (NEN Research Products, MA, U.S.A.) was added. An excess of unlabeled NEM was added after 2 h standing at room temperature and the mixture was kept for another 2 h. After incubation, 1 mg of Cys. HCI was added. NEM and Cys. HCI did not affect the inhibitory activity, 3.10. Binding ability of radioactive HIP to cells Five hundred lal of cell suspensions (Asso of 1.6) were incubated with 50/~1 of labeled NEMCys-HlP for 15 rain at 37°C, The cells were recovered by centrifugation (2000 × g) for 20 rain. The resulting sediment was solubilized in 0,5 ml of tissue solubilizer (Luma solve, Lumac/3M by,

061

Schaesberg, Netherlands). The radioactivity was determined by liquid scintillation spectrometry (Liquid Scintillation Counter, Model LSC-950, Aloka, Tokyo, Japan).

o

4. RESULTS

4.1. Purification of HIP

o

~

,'o Retent,on t~me (re,n)

4.1.1. Step 1: ultrafiltration Five hundred ml of human parotid saliva was clarified by centrifugation at 10000 x g for 20 min. The supernatant was centrifuged at 200000 x g for 3 h. The supernatant obtained by ultracentrifugation was ultrafiltered (cut-off molecular mass: 50000). The ehiate was pooled and lyophilized.

Fig. 2. Elution profile for gel filtration on Sephadex G-25. The fractions having hemagglutination inhibition activity are shown by the thick bar.

sent in the unbound fraction as indicated by the thick bar (Fig. 1).

4.1.3. Step 3: Sephadex G-25 gel-chromatography 4.I.Z Step 2: DEAE-Sephadex A-25 column chromatography The sample was suspended in 20 ml of 50 mM CH3COONH 4 buffer (pH 8.0) and applied to a DEAE-Sephadex A-25 column (2.2 x 30 cm) equilibrated in the same buffer. The column was developed with a gradient from 0 to 1.5 M N H a H C O 3. The flow rate was 20 m l / h , and the eluate was collected in 6 ml fractions. Each fraction was assayed for the absorbance at 280 nm (A2so) and hemagglutination inhibition activity. The fractions showing inhibitory activity were pre-

[

lm

, , , / , , , ,,,

15.

The active fractions from DEAE Sephadex were lyophilized and applied to a Sephadex G-25 Superfine column (1.6 x 70 cm) equilibrated in 50 mM CH3COONH 4 buffer (pH 8.0). The flow rate was 2 ml/min, and the ehmte was collected into 1.5 ml fractions. In each fraction, A23o was monitored. The active fractions containing inhibitory activity were detected in the middle peak (Fig. 2).

4.1.4. Step 4: reversed-phase HPLC The peak of step 3 (sample A) was lyophilized and subjected to reversed-phase HPLC. The elution profile from a YMC-GEL ODS $5 column (10 x 250 m e , YMC, Kyoto, Japan) developed with an acetonitrilc gradient containing 0.01% tri-

o 0 ~. i

)5

¢o

,60

,~0

250

Frocuon number

Fig, i. Bulion profile for ion exchange chromatography on DEAE.Scphadex A-25. The fractions showing hemagglutination inhibitionactivityare indicated by the thick bar.

I

o~ o

i

i

i

,

Retenuon ume (rain)

Fig. 3. Elution profile for reversed-phase HPLC on a YMC-

GEL ODS $5. The highest peak indicated by an arrow was pooled and lyophilized.

fluoroacetic acid is shown in Fig. 3. The flow rate was 2.5 m l / m i n and ,4220 was monitored. The highest peak indicated by an arrow was pooled and lyophilized. This sample was further purified on a Cosmosil 5C18 column ( 4 . 6 × 2 5 0 mm, Nacalai Tesque, Inc., Kyoto, Japan) using an acetonitrile gradient containing 0.01 M ammonium acetate (pH 5.7) as an eluent. One minor component was separated from one major one just before a main peak (data not shown). The main peak was collected and lyophilized (sample B).

] a.

4.2. Amino acid analysis

l-HIP ooeeO (~O$(Jpm)

The amino acid compositions of the samples from Sephadex G-25 (sample A) and Cosmosil 5C18 (sample B) are shown in Table 1. The cationic amino acids histidine, arginine and lysine were present in large amount in sample A. Sample B contained 12 residues and cationic amino acids predominated with 4 histidine, 2 lysine and 1 arginine, and was hence characterized as a histidine-rich peptide.

Table 1 Amino acid compositionof HIP Amino acid Asp Thr Ser Glu Pro Gly Ala Val Met lie Leu Tyr Pbe Lys His Arg [M + HI +

Sample A 1.68 0.07 2.30 3,02 3,65 3.04 1 0.13 0.03 0.12 0.21 0,"/0

Sample B 0.07 0,00 0.91 (1) * 1.34 (1) 0,00 1 (1) 0,02 0,03 0,02 0,00 0,00 0,96 (!)

0,95

0.99 (I)

3.88 3,93 3.11

3,80 (4) 1,89 (2) 1,06(1)

Dig. 4. Binding ability of radioactive HIP to B. gingipalis 381. Native cells (0) or heated cells (100°C, l0 rain) (4) were incubated with increasingamounts of HIP. Each value is the averageof duplicate assays. 4.3. Amino acid sequence The amino acid sequence of sample B was determined to be Lys-Pbe°His-Glu-Lys-His-His Ser-His-Arg-Gly-Tyr by the Edman degradation method. This sequence coincided with the amino acid analysis data and the mass value. The expeered mass value of the pseudo~molccular ion ([M + H] +) of sample B was ]562.9. According to this sequence, a peptide was synthesized by solidphase method. The synthetic peptide was eluted at the same retention time as that of the native one on a reversed-phase HPLC under the two different elution conditions and had the inhibitory activity as potent as the native one.

4, 4. Binding of radioactive HIP to bacteria The tritium-labeled HIP specifically bound to

B, gingivalis cells as compared to that of cells heated at 100°C for 10 rain, although the values of nonspccific binding to reaction tubes were too high (Fig, 4),

5. D I S C U S S I O N

1562,9 re,u,

* The values in parentheses are the assumed integral numbers of residues,

Since the strongest hemagglutinating activity was found in R gingivalis 381 among black-pig-

mented Bacteroides species in our preliminary experiment, B. gingivalis 381 was used in this work. As shown in Fig. 3, there are several fragments except in sample B. M o r e detailed studies are required to fully characterize the a m i n o acid sequences a n d hemagglutination inhibitory activities o f the other fragments. In saliva, numerous polypeptides and proteins have been isolated and recognized for their ability to bind strongly to hydroxyapatite molecules and microorganisms [1,10]. The most studied is a histidine-rich protein ( H R P ) [11,11] and these peptides have antimicrobial effects including growth inhibition o f Candida aibicans and Streptococcus mutans [13,14]. Recently, it has been reported that there are twelve H R P s n a m e d as histatins by O p p e n h e i m et al. in h u m a n parotid secretion and their structural interrelationships have been clarified [15,16]. It is of interest that the sequence o f H I P having twelve a m i n o acid residues was unexpectedly the same as histatin 8 and also exists in the histatins 3, 4, 5, 6, 7, 9 and 10. Furthermore, the same a m i n o acid sequence o f Lys-Phe-His-Ghi-Lys-His-His-Ser-His-Arg which is present in the HIP, has bccn found in histatins 1 to 10, as an essential structure. Fibrinogen is known to bind to the cell surface o f B. gingivalis, and its binding is rapid, reversible, saturable and specific [17]. A t t e m p t s to derive similar data for the binding of H I P to B. gingivalis cells failed due to the ease of its removal by washing (unpublished observation). The data have however demonstrated that tritium-labeled H I P bind specifically to the hemagglutinins o f B. gingivalis, Based on the observed inhibition of hemagglutination, it is quite possible that the H I P may occupy the hemagglutinating adhesins which

are present on the cell surfaces a n d / o r fimbriae of B. gingivalis and play an important role in the colonization o f B. gingivalis in the oral cavity.

REFERENCES [1] Slots, J. and Gibbons, R.J. 0978) Infect. Immun. 19, 254-264. [2] Okuda, K., Slots, J. and Ganco, R.J. 0981) Curr. Microbiol. 6, 7-12. [3] Gibbons, R.J. and Etherdern, 1. (1983) Infect. Immun. 41, 1190-1196. [41 Boyd, J. and McBride, B.C. (1984) Infect. lmmun. 45, 403-409. [5] Okuda, K., Yamamotoo A., Naito, Y, Takazo¢, I.. Slots, J. and Genco, R.J. (1986) Infect. lmmun. 54, 659-665. [6] Yoshimura, F., Takahashi. K., Nodasaka, Y. and Suzuki, T. {1984) J. Bactenol. 160, 949-957. [7] Keene, H.J. {1963) J. Dent. Res. 42, 1041. [81 Inoshita, E., Amano. A., Hanioka, T., Tamagawa, H., Shizukuishi, S. and Tsunemitsu, A. 0986) Infect. Immun. 52, 421-427. [9] Leslie, J., Williams, D.L. and Gorin. G. 0962) Anal. Biochem. 3. 257-263. [lO] Cimasoni, G., Song, M. and McBride, B.C. 0987) Infect. Immun. 55,1484-1489. [lll Hay, D.I. 0975) Arch. Oral Biol. 20, 553-558. [12] Baum, B.J., Bird, J.L., Millar, D.B. and Longton. R.W. {1976) Arch. Biochem. Biophys. 177. 427-436. [13] MacKay, B.J., Dencpitiya, L., lacono, V.J., Krost, S.B. and Pollock, J.J. (1984) Infect. lmmun. 44. 695-701. [14] Pollock, J.J., Denepitiya, L., MacKay, B.J. and lacono, V.J. (1984) Infect. Immun. 44, 702-707. [151 Oppenheim, F.G., Xu, T., McMillian, F.M., Levitz, S.M., Diamond, R.D., Offner, G.D. and Troxler. R.F. 0988) J. Biol. Chem. 263, 7472-7477. [161 Troxler, R.F., Offner, G.D,, Xu, T., Vanderspek, J.C. and Oppenheim, F,G. 0990) J. Dent. Res. 69, 2-6. [17] Lantz, M.S., Rowland, R.W., Switalski, L.M. and Hook, M. 0986) Infect, immun. 54, 654-658.