Effect of a chemical or proteolytic modification on the biological activity of guinea-pig cationic peptide

ELSEVIER

et BioDhvsica Acta Biochimica et Biophysics Acta 1243 (1995) 295-299

Effect of a chemical or proteolytic modification on the biological activity of guinea-pig cationic peptide Tatsuhisa Yamashita

*, Shin

Yomogida, Isao Nagaoka, Kimiko Saito

Department OfBiochemistry, Juntendo University, School of Medicine, 2-l-l Hongo, Bunkyo-ku, Tokyo 113, Japan

Received 6 July 1994; accepted 15 August 1994

Abstract Guinea-pig neutrophil cationic peptides (GNCPS) are single polypeptides containing 31 amino acid residues and three intramolecular disulfide bonds, which show both antibacterial and histamine-releasing activities. Reduction and alkylation of the disulfide bonds of GNCP did not reduce both biological activities. When pyridylethylated GNCP was digested with trypsin, the biological activities were almost lost, whereas the chymotryptic digest retained the biological activities. Chymotrypsin digested fragments were purified by RP-HPLC, and three active peptide fragments containing two Arg residues at the N-terminal sequence were isolated. When the biological activities were examined using synthetic peptides containing various numbers of Arg residue at the N-terminus, the omission of the Arg residues was found to reduce remarkably the antibacterial and histamine-releasing activities. Together these observations indicate that the primary structures containing Arg residues at the N-terminus but not the intramolecular disulfide cross-linking are important for the expression of the biological activities of GNCP. Keywords:

Neutrophil; Cationic peptide; Antibacterial activity; Histamine-releasing activity; Modification; (Guinea-pig)

1. Introduction Neutrophils play a central role in host defense by ingesting and killing microorganisms. The antimicrobial systems of neutrophils can be divided into two categories [l-3]. One is an oxygen-dependent mechanism in which H,O,, 0, and HOC1 are involved. The other is an oxygen-independent mechanism in which granular antimicrobial proteins and peptides are involved. The most abundant of the granular antimicrobial components is a lowmolecular-weight cationic peptide which shows the potent microbial activity against bacteria, fungi and viruses [2-81. In the previous study [9], we purified two structurally homogeneous cationic peptides, GNCP-1 and GNCP-2, from the granular fraction of guinea-pig neutrophils, and showed that both peptides not only release histamine from rat mast cells but also are equally active against Grampositive and -negative bacteria. The composition and sequence analyses have revealed that both peptides are single polypeptides containing 31 amino acid residues and three intramolecular disulfide bonds [9-111, and that they differ

* Corresponding author. Fax: +81 3 38149300 0304-4165/95/$09.50 0 1995 Elsevier Science B.V. All rights reserved SSDI 0304-4165(94)00133-l

only by the substitution of an isoleucine (GNCP-1) for a leucine (GNCP-2) at position 21 [9]. It is reported that toxins are stable on exposure to acid and high temperature [12,13] but are inactivated by reduction and alkylation of their disulfides, suggesting that the disulfide bridges are required for the biological activity. It is of great interest, therefore, to examine the relationship between the structure and the biological activity of GNCP. In this report, we attempted to identify the possible structure required for the biological activity by a chemical or proteolytic modification, using GNCP-1, one of the guinea-pig neutrophil cationic peptides.

2. Materials

and methods

2.1. Materials Glycogen (Type II), trypsin (type XIII: TPCK-treated) and chymotrypsin (Type VII: TLCK-treated) were purchased from Sigma Chemical Co., St. Louis, MO; N-ethylmorpholine, sequanal grade, from Pierce Chemical Co., Rockford, IL; dithioerythritol and 4-vinylpyridine from Nacalai Tesque, Kyoto, Japan; high performance liquid

296

T. Yamashita et al. /Biochitnica et Biophysics Acta 1243 (1995) 295-299

chromatography grade acetonitrile from Kanto Chemical Co., Tokyo, Japan; and amino acid analysis grade trifluoroacetic acid from Wako Pure Chemical Industries, Osaka, Japan. 2.2. Purification of cationic peptides GNCPs were purified from guinea-pig peritoneal exudate neutrophils as previously described [9]. Neutrophils (purity > 95%) were suspended in ice-cold 0.34 M sucrose at a concentration of 2 . lo8 cells/ml, and disrupted by sonication in ice with four 10 s bursts at 25 W (Tomy ultrasonic disrupter, model UD-201, Tominaga Works Tokyo, Japan). The sonicate was centrifuged at 420 X g for 12 min at 4°C and the resulting supernatant was further centrifuged at 8200 X g for 15 min at 4°C. The resultant pellet (granule fraction) was suspended in 0.34 M sucrose, and sonicated for 1 min in ice at 168 W for solubilization. Cationic peptides (GNCP-1 and -2) were isolated by acid polyacrylamide gel electrophoresis [9]; the solubilized granule samples were electrophoresed on 15% polyacrylamide slab gel in @alanine buffer (pH 4.5) at 4°C for 4 h at 150 V. Thereafter, polypeptide bands were visualized by immersing the slab gel in a solution of 0.25% eosin Y in 0.1 N NaOH for 30 s. The band containing GNCP-1 and -2 was excised, ground, and subjected to electrophoretic elution in 0.16% acetic acid. GNCP-1 was further purified from GNCP-2 by reversed-phase high-performance liquid chromatography (RP-HPLC) on a Wakosil 5C8 column (7.5 X 300 mm; pore size 12.0 nm; Wako Pure Chemical). Water-acetnitrile gradients containing 0.1% trifluoroacetic acid were employed for elution. 2.3. S-alkylation of cationic peptide S-Pyridylethylation of cationic peptide was performed as described previously [9]. GNCP-1 (1 mg) was dissolved in 0.5 ml of 0.13 M Tris-HCl (pH 7.6) containing 6 M guanidine-HCl and 0.01% EDTA, and reduced for 3 h at room temperature by addition of 24.5 ~1 of 94.5 mg/ml dithioerythritol (about 20 molar excess over total disulfides). The free sulfhydryl groups were then exposed to 4.78 ~1 of 4-vinylpyridine (3:l molar ratio with respect to reducing agents used), and the solution was stirred for 90 min at room temperature. The S-pyridylethylated peptide solution was adjusted to pH 3 with 90 ~1 of glacial acetic acid, dialyzed at 4°C for 48 h against 0.16% acetic acid using a Spectrum Pore 6 of molecular mass cutoff of 1000 (Spectrum Med ical Industries, Los Angeles, CA), and lyophilized. 2.4. Enzyme digestion of pyridylethylated

GNCP-1

Pyridylethylated GNCP-1 (100 lug) was dissolved in 100 ~1 of 0.1 M N-ethylmorpholine acetate buffer (pH

8.2) and digested by trypsin or chymotrypsin at the final concentration of 50 pg/ml for 2 h at 37°C (2OO:l weight ratio with respect to enzyme used). The reaction was terminated by the addition of 15 ~1 of glacial acetic acid, and the digested peptides were lyophilized. The digested peptides were separated by RP-HPLC using a Wakosil 5C8 column (4.6 X 250 mm; pore size, 12.0 nm). The fractions were pooled, lyophilized, and subjected to amino acid analysis and bioassay. 2.5. Amino acid analysis Isolated peptides were hydrolyzed by 6 N HCl in an evacuated and sealed tube at 110°C for 24 h. Then, the acid was removed on rotary evaporator at approximately 40°C under reduced pressure, and the residue was dissolved in 0.01 N HCl. These samples were analyzed on a Hitachi 835 High-Speed Amino Acid Analyzer (Hitachi, Tokyo, Japan). 2.4. Preparation

of synthetic peptides

Peptides were synthesized with acetoamidomethylation of cysteine residues on a Model 430A Peptide Synthesizer (Applied Biosystems, Foster, CA), and their sequences were confirmed by a Model 477A Pulsed Liquid Sequencer (Applied Biosystems). 2.7. Antibacterial

assay

The bactericidal activities of peptides were tested against Staphylococcus aureus (NIHJ JC-1) as described previously [9]. After cells were grown in nutrient broth for 16 h at 37°C with shaking, the cells were washed twice with 10 mM phosphate buffer (pH 7.4) by centrifngation (6000 X g for 10 min), and diluted in the same buffer to give approximately 2 . lo6 colony-forming units/ml. Various concentrations of the peptides were added to 0.1 ml volumes of bacterial suspension in a total volume of 0.2 ml of 10 mM phosphate buffer, and were incubated for 20 min at 37°C. In control experiments, the cells were mixed with the solvent of the peptides (0.01% acetic acid). After a lOOO-fold dilution of the mixtures with 10 mM phosphate buffer, 0.1 ml aliquots were spread on nutrient agar plates and incubated for 18 to 20 h to allow full colony development. The resulting colonies were counted, and antibacterial activity was expressed as killing percentage, using the formula: Killing(%) =

I(

The number of colonies in the presence of the peptide The

numberof colonies in the absence of the peptide )

Xl00

2.8. Histamine release from rat mast cells Mast cells Sprague-Dawley

> 85%) (purity rats, and histamine

were obtained from release was assayed as

T. Yamashita et al. /Biochimica

et Biophysics Acta 1243 (1995) 295-299

described previously [9]. Histamine release was initiated by the addition of 0.05 ml of peptides in 0.01% acetic acid to mast cells (2. 104/ml) in a total volume of 1.5 ml Tris-ACM buffer (119 mM NaCl, 5 mM KCl, 0.6 mM CaCl,, 0.03% bovine serum albumin, and 31 mM Tris-HCl, pH 7.4), and mast cells were incubated at 37°C for 30 min. Histamine release was terminated by placing test tubes in an ice-water bath, followed by centrifugation at 220 X g for 10 min at 4°C. Histamine concentration was determined by the o-phthaldialdehyde spectrophotofluorometric procedure, as described previously [9]. Total cellular histamine was determined using the cell samples which had been incubated at 37°C with peptides for 30 min and then sonicated for 30 s. Histamine release is corrected for spontaneous release (3.0 f 0.4; mean f SD.; n = 91, and expressed as the percent of total histamine.

291

0

loo

B

1

3. Results and discussion 3.1. Effect of reduction and alkylation on the antibacterial and histamine-releasing activity of GNCP-1 In the preliminary experiments, the peptides were quite stable to heat (lOO”C, 10 min) and low pH, like toxins (data not shown) [12,13]. Then, the involvement of disulfide bond bridges in the expression of the biological activities were examined. After the reduction with dithioerythritol followed by the alkylation with 4-vinylpyridine, the dose dependence of biological activities of GNCP-1 were examined. As shown in Fig. lA, the ability of GNCP-1 to kill bacteria was not lost by the reduction and alkylation but rather potentiated; ED,, values of GNCP-1 and pyridylethylated GNCP-1 were approximately 2.7 pg/ml and 1.3 pg/ml, respectively. Fig. 1B shows the effect of a chemical modification on the ability of GNCP-1 to release histamine from mast cells, and the results were similar to those obtained about the antibacterial activity: ED,, values of GNCP-1, 1.2 pg/ml and pyridylethylated GNCP-1, 0.17 pg/ml, respectively. Pyridylethylated cysteine showed neither antibacterial nor histamine-releasing activity even at such a hrgh concentration as 180 pg/ml. These observations suggest that the intramolecular disulfide cross-linking is not necessarily required for the antibacterial and histamine-releasing activities, but might rather give a suppressive effect on the expression of the biological activities. 3.2. Effect of tryptic or chymotryptic digestion on the antibacterial and histamine-releasing activities of GNCP-1 Next, we proteolytically digested pyridylethylated GNCP-1 to identify the active fragments. After 2 h digestion of pyridylethylated GNCP-1, the biological activities of the digest were examined. As seen in Table 1, the antibacterial and histamine-releasing activities of

o-

0.1

1.0

10.0

Conoentrltlon(#I/II) Fig. 1. Effect of pyridylethylation on the antibacterial and histamine-releasing activities of GNCP-1. (A) Bacteria (5. aureus) were incubated with various concentrations of GNCP-1 (0) or pyridylethylated GNCP-1 (0) in 10 mM sodium phosphate buffer (pH 7.4) for 20 min at 37°C. Antibacterial activity is expressed as hilling (%). (B) Mast cells were incubated with various concentrations of GNCP-1 (0) or pyridylethylated GNCP-1 (0) in Tris-ACM (pH 7.4) for 30 min at 37°C. Data are the means f S.D. of three separate experiments.

pyridylethylated GNCP-1 were almost lost by tryptic digestion. On the other hand, the chymotryptic digest retained both the antibacterial and histamine-releasing activities, suggesting that the active fragments of GNCP-1 can be obtained by chymotrypsin digestion.

Table 1 Effect of chymotryptic and tryptic digestion on the biological pyridylethylated GNCP-1

Undigested pyridylethylated GNCP-1 Chymotrypsin-digested pyridylethylated GNCP-1 Trypsin-digested pyridylethylated GNCP-1

Killing (%)

Histamine

98.3 + 0.5

64.8f8.7

75.5 f 2.0

60.0 + 0.4

10.0 f 2.1

3.9 f 1.4

activities of

release (%I

Antibacterial and histamine-releasing activities were examined using peptides at concentrations of 5.0 pg/ml and 0.2 pg/ml, respectively. Data represent the means + SD. of three separate experiments.

T. Yamashita et al. /Biochimica

et Biophysics Acta 1243 (1995) 295-299 B

GIW-,(I-,,)

0

20

10

Elution time

60

80

(min)

Fig. 2. Reverse-phase HPLC profile of chymotrypsin-digested pyridylethylated GNCP-1. Chymotrypsin-digested pyridylethylated GNCP-1 was loaded onto a C8 column equilibrated in 0.1% trifluoroacetic acid in water. A gradient (- -) of acetonitrile was developed for 80 min at 3.0 ml/min. The effluent was monitored at 220 nm.

3.3. Isolation of active fragments from chymotryptic

digest

Then, the chymotryptic digest was chromatographed by RP-HPLC, and the peptide fragments were isolated. Five main peaks were observed in the elution profile, and numbered according to the order of elution as seen in Fig. 2. The amino acid composition and biological activities of each peak were examined. As shown in Table 2, neither peak I nor peak II had the ability to kill bacteria and release histamine. Amino acid analysis revealed that peak I was the fragment containing amino acid residues at positions 23-27 in GNCP-1 and peak II was the C-terminal tetrapeptide. On the other hand, peaks III, IV and V showed both biological activities. Amino acid analyses showed that peak III was the N-terminal tetradecapeptide, namely GNCP-l(l-14) peptide and peak V was the Cterminal fragment, GNCP-1(15-31) peptide. Peak IV was found to be the fragment in which the C-terminal residues 28-31 were omitted from GNCP-1(15-31) peptide and showed almost the same biological activity as peak V, suggesting that the C-terminal tetrapeptide is not involved

Table 2 Primary structures

and biological

activities of chymotrypsin-digested

3.4. Importance ties

m

CWCP-,(,-I,)

(Z-10

(3-10

of the GNCP-1(X-31)

of Arg-Arg

peptide frag-

sequence for biological actiui-

The three peptides (peaks III, IV and V) possessed both biological activities, and contained the two positively charged amino acid residues, Arg and Arg in the N-terminal sequence, suggesting an important role for the N-terminal sequence in bacteria killing and histamine release. Then, to confirm the importance of the Arg-Arg sequence, peptides with various numbers of Arg residue at the Nterminal sequence, namely GNCP-l(l-141, GNCP-1(214), GNCP-1(3-141, GNCP-1(15-27), GNCP-1(16-27) and GNCP-1(17-27) peptides were synthesized, and their biological activities were determined. As shown in Fig. 3, the antibacterial and histamine-releasing activities reduced to half by the omission of the first Arg residue from GNCP-l(l-14) peptide and were completely lost by omitting two Arg residues at the N-terminus. On the other

peptide fragments

99.6

QNRVY TFCC

Peak II

Pskk

(8-14)

in the active portion ment.

$;;f;lethrlrted 1 RRCI~TTRT~RFPY~RLGT~~~~~~~~~~~C Peak I

(Z-14)

Fig. 3. Biological activities of GNCP-1(1-14), GNCP-1(2-14), and GNCP-1(3-14) peptides. (A) Bacteria were incubated with 100 pg/ml of synthetic peptides. Antibacterial activity is expressed as killing (o/o). ED,, values were 55.6 pg/ml for GNCP-R-14) peptide and > 100 pg/ml for GNCP-1(2-14) peptide, respectively. (B) Mast cells were incubated with 5.0 pg/ml of synthetic peptides. Histamine release is expressed as the percent of total histamine. ED,, values were 3.7 pg/ml for GNCP-l(l-14) peptide and 8.5 pg/ml for GNCP-1(2-14) peptide, respectively. Data represent the meanf SD. of three separate experiments.

RRCICTTRTCRFPY

11.1

4.9

1.4

1.8

0.6

99. I

49.3

Peak IV

RRLGTCIFQNRVY

99,s

66.8

Peak V

RRLCTCIFQNRVYTFCC

91.6

80.1

Antibiological and histamine-releasing activities were examined using peptide fragments pg/ml, respectively. Data shown are representative of three experiments.

isolated by RP-HPLC

at concentrations

of 25 pg/ml

and 1.33

T. Yamashita et al. / Biochimica et Biophysics Acta 1243 (1995) 295-299 100

A

k-L CWcQ-l(16-Ll)(LB-Z1)(11-21)

Fig. 4. Biological activities of GNCP-1(15-27), GNCP-1(16-27), GNCP-1(17-27) peptides. (A) Bacteria were incubated with 20.0 pg/ml of synthetic peptides. Antibacterial activity is expressed as killing (%). ED,, value of GNCP-1(15-27) peptide was 14.0 pg/ml. (B) Mast cells were incubated with 5.0 @g/ml of synthetic peptides. ED,, values were 0.24 pg/ml for GNCP-1(15-271 peptide and 38.2 pg/ml for GNCP1(16-27) peptide, respectively. Data represent the mean+S.D. of three separate experiments.

hand, both biological activities of GNCP-1(15-27) peptide were markedly reduced by the omission of the first Arg residue (Fig. 4). These results suggest that the primary structures containing the positively charged amino acids play an important role in the expression of the biological activities of GNCP. Histamine-releasing activity of bradykinin, a nonapeptide with one arginine residue at the N- and C-terminal positions, is reduced by the removal of the N-terminal or C-terminal Arg residue [14,15]. In addition, the cationicity and hydrophobicity of lysosomal proteins such as cathepsin G has been suggested to be important for their bactericidal action [16,17]. Relcently we have found that the arginine residues but nolt the lysine residue may be involved in the antibacterial activity of guinea-pig eosinophil major basic protein [18]. These reports support our idea that the positively charged amino acids, especially arginine residues, are important for the expression of the antibacterial and histamine-releasing activities of cationic peptide, the most abundant component of neutrophil granules. Our findings may also help the design of clinically useful antibacterial peptides in the future.

299

Acknowledgements This work was supported in part by the Science Research Promotion Fund from the Japan Private School Promotion Foundation.

References HI Klebanoff,

S.J. (1988) in Inflammation, Basic Principles and Clinical Correlates (Gallin, J.I., Goldstein, LM. and Snyderman, R., eds.,) pp. 391-444, Raven Press, New York. 121Elsbach, P. and Weiss, J. (1988) in Inflammation, Basic Principles and Correlates (Gallin, J.I., Goldstein, I.M. and Snyderman, R., eds.), pp. 445-511, Raven Press, New York. [31 Lehrer, RI., Ganz, T., Selsted, M.E., Babior, B.M. and Curnutte, J.T. (1988) Ann. Intern. Med. 109, 127-142. [41 Selsted, M.E., Szklarek, D. and Lehrer, RI. (1984) Infect. Immunol. 45, 150-154. [51 Ganz, T., Selsted, M.E., Szklarek, D., Harwig, S.S.L., Daher, K., Bainton, D.F. and Lehrer, RI. (19851 J. Clin. Invest. 76, 1427-1435. [61Selsted, M.E., Szklarek, D., Ganz, T. and Lehrer, RI. (1985) Infect. Immunol. 49, 202-206. [71 Lehrer, RI., Daher, K., Ganz, T. and Selsted, M.E. (1985) J. Virol. 54, 467-472. [81 Daher, K., Selsted, M.E. and Lehrer, RI. (1986) J. Virol. 60, 1068-1074. [91 Yamashita, T. and Saito, K. (1989) Infect. Immunol. 57,2405-2409. [lOI Pardi, A., Hare, R.D., Selsted, E.M., Morrison, D.R., Bassolino, A.D. and Bach, CA. (1988) J. Mol. Biol. 201, 625-63616. WI Selsted, E.M. and Harwig, S.L.S. (1989) J. Biol. Chem. 264, 4003-4007. WI Corft, L.R. (1980) in Handbook of Protein Sequence Analysis: A Compilation of Amino-acid Sequence of Proteins, pp. 422-453, Wiley, New York. [I31 Selsted, E.M. and Harwig, S.L.S. (1987) Infect. Immunol. 55, 2281-2286. 1141Bueb, J.L., Mousli, M., Bronner, C., Rouot, B. and Landry, Y. (1990) Mol. Pharmacol. 38, 816-822. [151 Devillier, P., Renoux, M., Giroud, J.P. and Regoli, D. (1989) Eur. J. Pharmacol. 117, 89-96. b61 Selsted, M.E., Harwig, S.S.L., Ganz, T., Schilling, J.W. and Lehrer, RI. (1985) J. Clin. Invest. 76, 1436-1439. 1171Shafer, M.W., Pohl, J., Onunka, C.V., Bangalore, N. and Travis, J. (1991) J. Biol. Chem. 266, 112-116. 1181Hashimoto, Y., Nagaoka, I. and Yamashita, T. (1993) Biochim. Biophys. Acta 1203, 236-242.