Primary Structure and Anticandidal Activity of the Major Histatin from Parotid Secretion of the Subhuman Primate, Macaca fascicularis T. XU', E. TELSER', R.F. TROXLER" 2, and F.G. OPPENHEIM" 2,3

'Department of Periodontology and Oral Biology, Goldman School of Graduate Dentistry, and 2Department of Biochemistry, Boston University School of Medicine, 100 East Newton Street, Boston, Massachusetts 02118-2392 A major macaque histatin (M-histatin 1) from the parotid secretion of the subhuman primate, Macaca fascicularis, was isolated by gel filtration on Bio-Gel P-2 and purified to homogeneity by reversed-phase high-performance liquid chromatography on a TSK-ODS C18 column. The complete amino acid sequence of M-histatin 1, determined by automated Edman degradation, is: 1

10

20

Asp-Pse-His-Glu-Glu-Arg-His-His-Gly-Arg-His-Gly-His-His-Lys-Tyr-Gly-Arg-Lys-Phe 21

30

38

His-Glu-Lys-His-His-Ser-His-Arg-Gly-Tyr-Arg-Ser-Asn-Tyr-Leu-Tyr-Asp-Asn

M-histatin 1 contains 38 amino acid residues, a phosphoserine at residue 2, has a molecular weight of 4881.8, a calculated pI of 8.5, and histidine forms 26.3% of the mass. The hydropathicity plot of M-histatin 1 predicts that the molecule is entirely hydrophilic, and Chou-Fasman secondary prediction indicates that the polypeptide is devoid of alpha-helix and betasheet conformation in aqueous solutions but contains a series of beta turns. M-histatin 1 includes a six-amino-acid insert (residue 10-15) not present in human histatins and, with the introduction of gaps to maximize homology, it displays 89% and 91% sequence similarity with human histatins 1 and 3, respectively. M-histatin 1 exhibited fungicidal and fungistatic effects against the dimorphic pathogen, Candida albicans, in three separate bioassays. Its anticandidal effects were comparable with or greater than those of human histatins 1, 3, and 5. M-histatins 2, 3, and 4 were not sequenced directly because insufficient materials were available, but the amino acid composition of M-histatin 3 was nearly identical to that of the Nterminal 20 amino acid residues of M-histatin 1. There appears to be only one major histatin in macaque parotid secretion in contrast to the family of histatins in human parotid and submandibular secretions, and the significance of this in the context of evolution and mechanism of action in anticandidal assays is discussed. J Dent Res 69(11):1717-1723, November, 1990

Introduction. Histatins are a family of histidine-rich polypeptides in human parotid and submandibular gland secretions (Oppenheim et at., 1988, 1989). These proteins are the last group of low-molecular-weight proteins in salivary secretions to be fully characterized (Oppenheim et al., 1986, 1988; Xu et al., 1989; Troxler et al., 1990). Earlier work indicated that partially purified preparations of histidine-rich salivary proteins (now known to be histatins) stimulated glycolytic activity of oral bacteria (Holbrook and Molan, 1975), although the significance of this obReceived for publication April 2, 1990 Accepted for publication August 2, 1990 3To whom correspondence and reprint requests should be addressed This work was supported by National Institutes of Health Grants DE 05672 and DE 07652.

servation is not known. We have shown that histatin 1 plays an important role in the control of mineral-solution interaction in the oral cavity by virtue of its ability to inhibit secondary

precipitation (crystal growth) of calcium phosphate salts from supersaturated solutions (Oppenheim et at., 1986). Histatins have also been shown to exhibit fungicidal and fungistatic activity against the dimorphic pathogenic yeast, Candida albicans (Pollock et at., 1984; Oppenheim et at., 1986, 1988; Santarpia et at., 1988), and certain strains of oral bacteria (MacKay et at., 1984; Xu and Oppenheim, 1990). Recent studies with AIDS patients indicate that histatin levels are increased under immune deficiency conditions (Atkinson et al., 1990). Subhuman primates, specifically macaques, have been used as animal models for the investigation of caries (Bowen, 1981), periodontal disease (Kornman et al., 1981), and oral candidiasis (Austwick et al., 1966; Kaufmann and Quist, 1969; Budtz-Jorgensen, 1975). Relatively little is known concerning the structure and function of proteins in macaque salivary gland secretions. We have isolated and characterized a 42-residue tyrosine-rich phosphoprotein, M-statherin, and a proline-rich phosphoglycoprotein (MPRP) from macaque parotid secretion (Oppenheim et al., 1982, 1985). The primary structure of Mstatherin is highly homologous with that of human statherin (Schlesinger and Hay, 1977). The ability of M-statherin to inhibit both primary (spontaneous) and secondary crystal growth (Oppenheim et at., 1982) was very similar to that of human statherin (Hay et al., 1979, 1984). Significant structural and functional differences occur between MPRP and six major human anionic proline-rich proteins (PRPs) (Oppenheim et al, 1985; Hay et at., 1988). MPRP has five phosphoserines, PRPs have two; MPRP is a glycoprotein, PRPs are not. MPRP inhibits both primary and secondary crystal growth, whereas PRPs inhibit only secondary crystal growth (Oppenheim et al, 1985). Recently, the amino acid sequence of a second macaque statherin from parotid secretion of the stump-tail monkey, Macaca arctoides, was found to differ from M. fasciculaiis M-statherin by two amino acid residues and to display comparable activity in the inhibition of primary and secondary calcium phosphate precipitation (Schlesinger et at., 1989). The present investigation describes the isolation, characterization, amino acid sequence, and anticandidal properties of the major macaque histatin from the parotid secretion of M. fascicularis.

Materials and methods. Materials. -Parotid saliva was collected from M. fasciculanis under sodium pentobarbital anaesthesia with the aid of a Carlson-Crittenden device, as previously described (Oppenheim et at., 1982). Bio-Gel P-2 and electrophoresis reagents were purchased from BioRad Laboratories (Richmond, CA). Pepsin was purchased from Boehringer Mannheim (Indianapolis, IN), and phosphoserine was obtained from Sigma (St. Louis, MO). Acetonitrile was purchased from Baker Chemical Co. (Phillipsburg, NJ), and trifluoroacetic acid was obtained 1717

Downloaded from jdr.sagepub.com at Karolinska Institutets Universitetsbibliotek on May 23, 2015 For personal use only. No other uses without permission.

1718

XU et aL

from Pierce (Rockford, IL). Sequencer chemicals were obtained from Applied Biosystems, Inc. (Foster City, CA). Sabouraud dextrose agar and broth were obtained from Difco (Detroit, MI). RPMI-1640 medium was purchased from Gibco Laboratories (Grand Island, NY). All other chemicals used in experiments were from Fisher Scientific (Pittsburgh, PA), unless otherwise specified. Candida albicans (ATCC 44505) is a well-described strain isolated from human oral cavities (Odds, 1980). Cultures were stored on Sabouraud dextrose agar plate at 4°C until used. Isolation of M-histatins. -Macaque parotid secretion was pooled, dialyzed against distilled water at 4°C, and lyophilized. A 250-mg sample was taken up in 50 mL of 0.05 mol/ L ammonium format, pH 4.0, and chromatographed on a BioGel P-2 column (2.6 x 88 cm) equilibrated in the same buffer, as described previously (Oppenheim et al., 1988). The column was run at a flow rate of 40 mL/h; 20-mL fractions were collected, and eluate was continuously monitored at 230 nm with an LKB UVicord S. Fractions containing macaque histatins (M-histatins) were pooled and lyophilized. Aliquots were taken up in 2% acetic acid and subjected to reversed-phase high-performance liquid chromatography (RP-HPLC) on a TSKODS (120T) C18 column with monitoring at 228 nm. Fractions containing M-histatins were collected with the peak detector module of an LKB SuperRac fraction collector. Solvent A was 0.1% trifluoroacetic acid in water, and solvent B was 0.1% trifluoroacetic acid, 80% acetonitrile, 20% water. The 80-minute gradient was run, with respect to solvent B, as follows: 0-4 min, 0%; 4-14 min, 0-10%; 14-64 min, 10-40%; 64-70

Fig. 1-Cationic PAGE electrophoretograms of human (HPS, 400 jig) and macaque (MPS, 500 ,ug) parotid saliva protein. Human histatins (Histatins 1-8) and macaque histatins (M-histatins 1-4) are present in the cathodic portion of the electrophoretograms.

J Dent Res November 1990

min, 40-100%; 70-75 min, 100%; 75-80 min, 100-0%. Sep-

arated M-histatin was evaporated to dryness in a flash evaporator, re-dissolved in 2% acetic acid, and re-chromatographed isocratically at a concentration of solution B 8% lower than the concentration at which initial gradient elution occurred. Purified M-histatin was evaporated to dryness, dissolved in distilled water, lyophilized, and stored at - 20'C until used. Amino acid analysis.-Samples were hydrolyzed in 6 mol/ L HCl and 1% phenol vapor in a Waters PicoTag station at 108'C for 24 h, or in 4 mol/L HCI with 1% phenol vapor at 100'C for six h for determination of amino sugars. Amino acid analyses were performed on a Beckman System 6300 amino acid analyzer. Cationic gel electrophoresis.-Samples were examined on cationic polyacrylamide (15%) slab gels (cationic PAGE), as described (Baum et al., 1977; Oppenheim et al., 1988). Phosphate analysis.-Measurement of bound phosphate in M-histatin was performed as described (Svanborg and Svannerholm, 1961), with phosphoserine and pepsin as standards. Automated Edman degradation. -M-histatin was subjected to automated Edman degradation with an Applied Biosystems, Inc., 470A gas-phase sequencer equipped with a model 120 PTH analyzer. Bioassays.-C. albicans was grown on Sabouraud dextrose agar plates at 30'C for 18 h. Cells were harvested from the plates, suspended in 0.01 mol/L potassium phosphate buffer, pH 7.4 (suspension buffer), and diluted with the same buffer to 1 x 105 cell/mL for all three assays. Test proteins-including M-histatin 1, and human histatins 1, 3, and 5-were dissolved in suspension buffer and, after appropriate dilution, were added to cells. Cell viability in all three assays was determined colorimetrically with the tetrazolium salt MTT, as described by Levitz and Diamond (1985). (a) Blastospore killing assay.-Fifty-pLL aliquots of C. albicans (1 x 105 cells/mL) in suspension buffer were combined with an equal volume of M-histatin 1 or human histatins 1, 3, and 5 in the same buffer and incubated in a 96-well culture plate on a shaker at 37°C for one h. Controls were made by a combination of 50 ,uL of cell suspension (1 x 105 cells/mL) and 50 pLL of suspension buffer minus test proteins. After incubation, wells were washed three times with 150 pLL of suspension buffer, with gentle aspiration between washes. Aliquots of molten Sabouraud dextrose broth containing 3.5%

Fig. 2-Elution profile of 250 mg of M. fascicularis parotid saliva protein from a Bio-Gel P-2 column (2.6 x 88 cm) equilibrated and developed in 0.05 mol/L ammonium formate at pH 4.0. The flow rate was 40 mL/h, and 20-mL fractions were collected. The eluate was continuously monitored for absorbance at 230 nm. Fractions were pooled as indicated (bars), lyophilized, and subjected to RP-HPLC. An aliquot of each fraction (A-F) was examined in the cationic PAGE system (inset).

Downloaded from jdr.sagepub.com at Karolinska Institutets Universitetsbibliotek on May 23, 2015 For personal use only. No other uses without permission.

STRUCTURE AND ACTIVITY OF MACAQUE SALIVARY HISTATIN

Vol. 69 No. 11

agarose at 45TC were added to each well, and the plate was incubated at 30TC for eight h. A total of 100 cells or colonies was counted under an inverted microscope at 400 x magnification. Biological activity was calculated according to the formula: % blastospore killing = [1 - (colonies in treated sample)/ (colonies in control)] x 100. (b) Genn-tube killing assay. -Cells were suspended at 1 x 105 cells/mL in RPMI-1640 medium containing 0.01 mol/L HEPES, pH 7.4, and 50-pLL aliquots were incubated in the wells of a microtiter plate at 370C for three h. An average of 95% germination was usually obtained per experiment. After germ tubes were washed three times with 150 [LL of suspension buffer, 50 [L of protein test solution plus 50 pLL of suspension buffer were mixed and added to each well. Controls were incubated with 100 FL of suspension buffer without test protein. Following incubation at 37TC for one h, cells were washed three times, as above, and 60 pL of molten Sabouraud dextrose broth containing 3.5% agarose at 450C was added to each well and incubated at 250C for 10 h. The assay consisted of counting a total of 100 non-budding germ tubes or budding germ tubes under an inverted microscope at 400 x magnification and the scoring of non-budding germ tubes as killed and budding germ tubes as not killed. Biological activity was calculated according to the formula: % germ tube killing = [1 - (colonies in treated sample)/(colonies in control)] x 100. (c) Inhibition of germination assay. -Cells were suspended at 1 x 105 cells/mL in RPMI-1640 medium with 0.01 mol/L HEPES, pH 7.4, and 50-pLL aliquots plus 50 pL of suspension buffer with or without test proteins were incubated in the wells of a microtiter plate at 370C for three h. Percent germination was determined by counting 100 germ tubes or ungerminated blastospores per well under an inverted microscope. Controls were obtained by incubation without proteins and always exhibited at least 95% germination. Biological activity was calculated according to the formula: % germination inhibition = [1 - (% germination in treated sample)/(% germination in control)] x 100. Statistical analysis.-Dose-response curves were produced by probit transformation for linear regression analysis from which concentrations of proteins giving 50% killing (LD50) or 50% inhibition of germination (ID50) were obtained (Govindarajulu, 1988). M-histatin 1 and human histatins 1, 3, and 5 i

M-his 3

M-his

I

1719

were tested at each concentration in triplicate in a given experiment, and all assays were repeated at least three times.

Results. Comparison of the electrophoretic patterns of human and parotid salivary proteins in a cationic PAGE system revealed the presence of one prominent protein band (M-histatin 1) with a mobility midway between those of human histatins 1 and 3 (Fig. 1). In addition, three minor bands were also noted (M-histatins 2, 3, and 4, Fig. 1). M-histatin 2 was close to the major band, while M-histatins 3 and 4 represent a second pair of proteins migrating more cathodically and exhibiting mobilities midway between those of histatins 3 and 5 (Fig. 1). Chromatography of 250 mg of macaque parotid salivary protein on Bio-Gel P-2 resolved the mixture into a large void volume peak containing the bulk of salivary proteins (peak A), a salt peak (peak B), and a broad tailing peak (Peaks CF) that eluted between 2 and 4 column volumes (Fig. 2). Electrophoretic analysis of column fractions in a cationic PAGE system revealed that peak C contained primarily M-histatins 3 macaque

and 4. Peaks D-F contained predominantly the major macaque histatin (Fig. 2, inset). The material in peak C-F was pooled, lyophilized, and subjected to RP-HPLC on a TSK-ODS C18 column developed with an acetonitrile gradient. This resolved the mixture into a cluster of small peaks elutingg between 20 and 25% solvent B) and a large peak, consisting of more than 90% of the total protein applied to the column, eluting between 28 and 30% solvent B (Fig. 3). The major peak (M-histatin 1) and the largest of the minor peaks (M-histatin 3) were rechromatographed by isocratic elution and analyzed in the cationic PAGE system (Fig. 3, inset). The amino acid composition of M-histatin 1 and M-histatin 3 is compared with that of human histatins 1 and 3 in Table 1. M-histatin 1 contained ten residues of histidine, five of arginine, four of glycine and tyrosine, and three of aspartic acid/asparagine, glutamic acid/glutamine, serine, and lysine, and lacked threonine, proline, alanine, valine, cysteine, methionine, and isoleucine. Based on ten residues of histidine, M-histatin 1 contained 38.3 amino acid residues. The amino acid composition of M-histatin 3 was similar to that of M-histatin 1 in that it was enriched with respect to histidine, arginine, and glycine, but lacked leucine. Based on six residues of histidine, M-histatin 3 contained 19.7 amino acid residues. The amino acid compositions of both M-histatins 1 and

: imlg

E

c

-60 I-

a, 0-6 t 'd

65

_ 0 m" :04

1

_]40 Q)7

TIME

min)

Fig. 3-Elution profile of fraction C-F (see Fig. 2) on a TSK ODS 120T C18 column (4.6 x 250 mm, pore size 5 pm) developed with an acetonitrile gradient (dotted line) at a flow rate of 1 mL/min. Fractions containing M-his 1 and M-his 3 (bars) were pooled, evaporated to dryness, dissolved in solvent A, and re-chromatographed isocratically. Purified preparations of M-his 1 and M-his 3 were subjected to cationic PAGE (inset). Abbreviations: M-his 1, macaque histatin 1; M-his 3, macaque histatin 3.

TABLE I AMINO ACID COMPOSITION OF MACAQUE AND HUMAN HISTATINS* Amino Acid M-histatin 1 M-histatin 3 Histatin 1+ Histatin 3+ Asx 3.5 (4) 1.3 (1) 5.2 (5) 4.4 (4) Ser 2.8 (3) 1.2 (1) 2.6 (3) 2.8 (3) Glu 3.3 (3) 1.8 (2) 3.2 (3) 0.8 (1) Pro 0 0 (0) (0) 1.1 (1) 0 (0) 4.2 (4) 2.6 (3) Gly 3.2 (3) 2.3 (2) Ala 0 (0) 0 (0) 0 (0) 0.9 (1) Leu 1.1 (1) 0 (0) 1.1 (1) 1.1 (1) 4.1 (4) Tyr 1.2 (1) 4.9 (5) 4.1 (4) Phe 1.2 (1) 0.8 (1) 3.1 (3) 1.3 (1) His 10.0 (10) 6.0 (6) 7.0 (7) 7.0 (7) 3.0 (3) Lys 1.5 (2) 3.1 (3) 4.1 (4) Arg 5.1 (5) 3.3 (3) 4.0 (4) 4.0 (4) Total: 38.3 (38) 19.7 (20) 38.5 (38) 32.8 (32) *Values in parentheses were the residues identified by automated Edman degradation for M-histatin 1 and by amino acid analysis for M-histatin 3. +Data from Oppenheim et al. (1988).

Downloaded from jdr.sagepub.com at Karolinska Institutets Universitetsbibliotek on May 23, 2015 For personal use only. No other uses without permission.

1720

J Dent Res November 1990

XU et al.

10

1 AA

Asp-Ser-His-Clu-Glu-Arg-His-His-Gly-Arg-His-Cly-His-Hls-Lys-Tyr-Gly-Arg

pmol

864 270 208 576 576 144 200 240 436 144 176 400 208 224 196 342 256 134

AA pMol

Lys-Phe-His-Clu-Lys-His-His-Ser-His-Arg-Gly-Tyr-Arg-Ser-Asn-Tyr-Leu-Tyr 296 235 120 192 204 112 144 75 112 72 176 130 75 36 72 91 44 89

AA pMol

38 Gly-Asn 84 41

30

20

Fig. 4-Amino acid sequence of M-histatin 1. The recovery of PTH derivatives is given below each amino acid. The initial yield was 72%, and the repetitive yields calculated for Glu,/Glu22 and Gly,/Gly29 were 93.4% and 95.6%, respectively.

3 were very similar to those of human histatins 1 and 3, respectively. The complete amino acid sequence of M-histatin 1 was determined in a single experiment by automated Edman degradation (Fig. 4). The protein contained 38 amino acid residues, had a molecular weight of 4881.8, a pI of 8.5 (calculated by the IBI Sequence Analysis program), and 26.3% of the mass consisted of histidine. The hydropathicity plot (Kyte and Doolittle, 1982) of M-histatin 1 indicated that the molecule was hydrophilic along its entire length, and a prediction of secondary structure (Chou and Fasman, 1979) indicated that the molecule was devoid of alpha-helices and beta sheets, but instead contained a series of beta turns that would tend to give the protein a random structure in aqueous solutions (data not shown). The experimentally determined sequence was in complete agreement with the amino acid composition of the protein (Table 1). M-histatin 3 was not sequenced directly because insufficient material was available, but the amino acid composition of this peptide was nearly identical to that of the N-terminal 20 amino acid residues of M-histatin 1. This suggests that M-histatin 3 may arise from M-histatin 1 by chymotryptic-like cleavage between Phe20 and His21. Direct measurement of bound phosphate indicated that Mhistatin 1 contained 1 mol of phosphate/mol protein and that pepsin and phosphoserine contained 1.3 and 1.0 mol of phosphate/mol, in agreement with the known phosphate composition of these control compounds (Table 2). Although there were three serine residues in M-histatin 1, dehydroserine was the predominant PTH-derivative observed at cycle 2, whereas serine was the predominant PTH-derivative at cycles 26 and 32. It was concluded that serine at residue 2 is phosphorylated, and that serine at residues 26 and 32 is not, by virtue of the detection of 1 mol bound phosphate/mol protein and observation of PTH-dehydroserine only at cycle 2 during automated Edman degradation (Oppenheim et al., 1986; Schlesinger et al., 1989). Amino acid analyses of M-histatins 1 and 3 gave TABLE 2

PHOSPHATE ANALYSIS OF M-HISTATIN 1 AND STANDARD PROTEINS* Protein Phosphate/Protein Phosphate Sample

(nmol) 19.3 13.2 10.3

(nmol) 20.0

(nmolnmol)

1.0 1.3 10.0 Pepsin 1.0 10.0 Phosphoserine *Each value given represents the average of three determinations.

M-histatin 1

no evidence of the presence of hexosamines in macaque histatins. The antimicrobial activities of M-histatin 1 and human histatins 1, 3, and 5 against C. albicans were compared in three different assays. These assays measure the ability of test compounds to kill blastospores, kill germ tubes, and prevent germination of blastospores. M-histatin 1 was active in all three assays, and clearly exhibited a greater biological activity in killing germ tubes, and tended to exhibit greater activity in inhibiting blastospore germination (Table 3). The LD50 of Mhistatin 1 in the blastospore killing assay was 4.4 nmol/mL, a value roughly equivalent to that of histatin 3, but lower than that of histatin 1 and higher by a factor of 2 than that of histatin 5, which was most active in this assay. The comparison of LD50 values obtained with the assay measuring the killing of germ tubes revealed that M-histatin 1 was twice as effective as histatins 3 and 5, and 12 times as effective as histatin 1. The IDs5 values (50% inhibition dose) of all histatins determined for the inhibition of germination were nearly the same, ranging from 36.3 nmol/mL (M-histatin 1) to 52.2 nmol/mL (histatin 1). It should be noted that the concentrations of histatins to produce a 50% inhibition effect were significantly higher than the concentrations required for 50% killing of blastospores or germ tubes. This is due to the fact that the germination inhibition assay is performed in RPMI-1640 medium exhibiting a high ionic strength that negatively affects the ability of the test proteins to exert their biological activity (Xu and Oppenheim, 1989). Collectively, the results from the three assays show that M-histatin 1 displays an equivalent or even greater biological activity than do human histatins 1, 3, and 5, with respect to anti-microbial properties against C. albicans.

TABLE 3

COMPARISON OF ANTICANDIDAL ACTIVITIES OF M-HISTATIN 1 WITH THOSE OF HUMAN HISTATINS* Histatin 5 Histatin 3 Histatin 1 M-histatin 1 Killing of Blastospores (LD50) 4.2 (3.6-5.0) 2.0 (1.8-2.2) 6.3 (6.1-6.4) 4.4 (4.1-4.8) Killing of Germ Tubes (LD50) 5.7 (5.2-6.3) 72.0 (67.3-81.1) 11.6 (10.8-12.6) 9.7 (8.6-11.2) Inhibition of Germination (ID5o) 36.3 (31.4-39.2) 52.2 (48.2-57.8) 38.9 (35.4-42.7) 49.2 (39.7-53.6) *Values are given in nmol of protein per mL required for 50% killing (LD5,) or 50% inhibition (ID50). Values in parentheses indicate 95% confidence interval of LD50 or ID50.

Downloaded from jdr.sagepub.com at Karolinska Institutets Universitetsbibliotek on May 23, 2015 For personal use only. No other uses without permission.

1721

STRUCTURE AND ACTIVITY OF MACAQUE SALIVARY HISTATIN

Vol. 69 No. 11

I

10

20

M-histatin 1:

Asp-Pse-His-GIu-Clu-Arg-His-His-Gly-Arg-His-Gly-His-His-Lys-Tyr-CI -Arg-Lys-Phe-His-GIu-Lys-

Histatin 1:

Asp-Pse-His-GIu-Lys-Arg-His-His-GIy-------------------------Tyr-Arg-Arg-Lys-Phe-His-Glu-Lys-

Histatin 3:

Asp-Ser-His-AIa-Lys-Arg-His-His-GIy------------------------ Tyr-Lys-Arg-Lys-Phe-His-GIu-Lys24

30

38

His-His-Ser-His-Arg-GIy-------------------------Tyr-Arg-Ser-Asn-Tyr-Leu-Tyr-Asp-Asn Hi s-H i s-Ser-H I s-Arg-C I- Phe-Pro-Phe-Tyr-G I y-Asp-Tyr-G I-Ser-Asn-Tyr-Leu-Tyr-Asp-Asn

His-His-Ser-His-Arg-GIy------------------------- Tyr-Arg-Ser-Asn-Tyr-Leu-Tyr-Asp-Asn Fig. 5-Comparison of the amino acid sequences of M-histatin 1 and human histatin 1 and histatin 3. Dashed lines indicate gaps. Serine at residue 2 of M-histatin 1 and histatin 1 is phosphorylated. Residues in bold-face type indicate amino acid substitutions.

Discussion. The present investigation describes the first isolation, complete amino acid sequence, and functional characterization of a histatin from a sub-human primate. We have shown previously that electrophoresis of human parotid secretion in a cationic PAGE system separates human histatins from other salivary proteins. The former are present as three pairs of cathodically migrating bands represented by histatins 1 and 2, histatins 3 and 4, and histatins 5 and 6 (Oppenheim et al., 1988), in addition to a more rapidly migrating band at the dye front containing a mixture of the minor histatins 7-12 (Troxler et al., 1990). Electrophoretic analysis of macaque parotid secretion in the cationic PAGE system revealed one major band, M-histatin 1, with an electrophoretic mobility midway between those of the pairs of bands represented by human histatins 1 and 2 and human histatins 3 and 4. In addition, three minor bands, M-histatins 2, 3, and 4, were also observed (Fig. 1). The presence of four macaque histidine-rich proteins in parotid secretions of old-world monkeys was noted earlier by use of a cationic starch gel-electrophoresis system (Azen, 1973; Peters et al., 1977). The most abundant of these minor macaque histatins was M-histatin 3. Amino acid analysis indicated that M-histatin 3 was a peptide corresponding to the amino-terminal 20 residues of M-histatin 1. This indicates that M-histatin 3 is derived from M-histatin 1 by proteolysis. M-histatins 2 and 4 exist as trace components, and their structural relationship to M-histatins 1 and 3 is not known. M-histatins 2 and 4 were considered to be histatins on the basis of their chromatographic behavior on Bio-Gel P-2 and electrophoretic properties in the cationic PAGE system. No other proteins in salivary secretions are known which display these properties, and insufficient quantities of M-histatins 2 and 4 were obtained in the present work for a full structural and functional characterization to be carried out. M-histatin 1 is the only major histatin in parotid secretion of M. fascicularis, whereas human parotid secretion contains three major histatins, histatins 1, 3, and 5 (Oppenheim et al., 1988). This is somewhat analogous to the situation with proline-rich proteins (PRPs) in parotid secretion, where there is one major macaque PRP, a phosphoglycoprotein called MPRP (Oppenheim et al., 1982, 1985), and six major acidic human PRPs, designated PRP-1, PRP-2, PRP-3, PRP-4, PIF-s, and PIF-f (Hay et al., 1988; Minaguchi and Bennick, 1989). Thus, for histatins and PRPs, the evolutionary pressures that led on

one hand in M. fascicularis to only one major histatin and one major proline-rich protein, and on the other hand in man to multiple histatins and PRPs, must have been different. The primary structures of M-histatin 1 and human histatins 1 and 3 were compared, with gaps introduced for delineation of inserts that are unique to one, but not the other, of these polypeptides (Fig. 5). M-histatin 1 contains a six-residue sequence, -Arg-His-Gly-His-His-Lys- (residues 10-15), absent from histatins 1 and 3, and histatin 1 contains a six-aminoacid sequence, -Phe-Pro-Phe-Tyr-Gly-Asp- (residues 24-29), not present in either M-histatin 1 or histatin 3. Otherwise, these proteins are remarkably similar. If the residues in M-histatin 1 are numbered as 1-38 (Fig. 5), the same amino acids occur in all three proteins at positions 1-3, 6-9, 16, 18-28, 30, and 32-38. This shows that amino-terminal, middle, and carboxylterminal domains of macaque and human histatins have been conserved in evolution and implies that these domains are functionally relevant. The sequence homology between macaque and human histatins is perhaps illustrated more dramatically by a homology matrix that shows that M-histatin 1 and histatin 1 exhibit 89% sequence similarity, M-histatin 1 and histatin 3 exhibit 91% sequence similarity, and histatins 1 and 3 exhibit 88% sequence similarity, if the six-residue inserts unique to M-histatin 1 and histatin 1, respectively, are not considered (Fig. 6). There is ample evidence that human histatins play an im-

M-histatin 1

Histatin 1

Histatin 3

M-histatin 1

100

89

91

Histatin 1

89

100

88

Histatin 3

91

88

100

Fig. 6-Homology matrix comparing percent

sequence

similarity be-

tween M-histatin 1 and human histatins 1 and 3. Amino acids were scored as identical or dissimilar, according to the alignment shown in Fig. 5. The

six residue inserts in M-histatin 1 and histatin 1 calculation.

Downloaded from jdr.sagepub.com at Karolinska Institutets Universitetsbibliotek on May 23, 2015 For personal use only. No other uses without permission.

were

excluded from the

1722

XU et al.

portant role in the oral cavity by taking part in the host's nonimmune defense system (Pollock et al., 1984; Oppenheim et al., 1986, 1988; Santarpia et al., 1988). The known concentrations of human histatins in healthy adults range between 112 and 2-30 nmol/mL for stimulated and unstimulated secretions, respectively (Atkinson et al., 1990). The concentrations of M-histatin 1 and human histatins exhibiting 50% of maximum activity for killing C. albicans blastospores and inhibiting blastospore germination were comparable and were within the histatin concentration ranges found for salivary secretions of healthy adults. M-histatin 1 was 12 times more effective than histatin 1, and twice as effective as histatins 3 and 5 in killing germ tubes. This suggests that the six-residue insert (residues 10-15) unique to M-histatin 1 and absent from human histatins plays an important role in the activity responsible for killing the mycelial form of C. albicans. Killing of this form of C. albicans is a significant biological activity because there is a direct correlation between germination and infection (Odds, 1979), and it is thought that the mycelial form can invade tissue and escape phagocytosis (Gow and Gooday, 1987). The precise details of the mechanisms by which histatins exert anticandidal activity are not known. It is clear that different processes lead to the killing of blastospores and inhibition of germination. The killing of blastospores, whether germinated or not germinated, is assayed by comparison of the numbers of single cells and cell colonies after a suitable time (Table 3). Colonies arise from single cells mitotically, and single cells not producing colonies are dead, as shown by their inability to produce MT1-formazan (Levitz and Diamond, 1985). Hence, the killing assays do not simply provide an indirect measurement of mitosis inhibition, but actually provide an estimate of cell death following exposure to test proteins. Killing of blastospores has been correlated with loss of cell potassium, suggesting that histatins in some way alter membrane permeability and/or cause membrane damage leading to loss of viability (Pollock et al., 1984). Raj et al. (1990) have presented evidence based on circular dichroism that a synthetic histatin 5 exists in a random structure in water and dilute phosphate buffer, but assumes a helical conformation in non-aqueous solvents. This interesting finding suggests a possible mode of action for histatins in the killing assays. Histatins in random conformation in the oral cavity may bind to a receptor on the blastospore cell membrane, and the resulting histatin-receptor interactions could provide an environment that promotes a transformation from a random structure to a helical conformation. This could favor insertion into the lipid bilayer of the cell membrane, where association of histatins with membrane proteins (e.g., ion translocators) may trigger loss of potassium or other essential ions. Alternatively, the presence of histatins in the cell membrane may initiate any one of a number of intracellular events (e.g., ADP-ribosylation; altered gene expression) resulting in cell death. Inhibition of blastospore germination is an interesting phenomenon because the assay measures the arrest of a developmental process that is incompletely understood but undoubtedly involves a programmed expression of a set of genes required for the switch from the yeast form to the mycelial form of C. albicans. The ability of M-histatin 1 and human histatins to inhibit this process must require binding to the cell membrane, followed by internalization and inhibition of gene expression. The complex processes that ensue are not known, but it seems certain that the molecular events leading to inhibition of germination are ultimately manifested in the nucleus. Histatins could exert their effect by binding to cytosolic proteins, followed by movement into the nucleus. Once in the nucleus, histatins or complexes of histatins and cytosolic or nuclear binding proteins could serve as positive or negative trans-act-

J Dent Res November 1990

ing factors that affect expression of genes required for transformation from the yeast to the mycelial form. Alternatively, histatins could exert either positive or negative control over transcription of other sets of genes whose products are required to promote differentiation. Finally, the fact that M-histatin 1 exhibits anticandidal activity to an extent the same as or greater than do human histatins makes this macaque protein an attractive candidate for studies on the biological role of this family of salivary proteins. Studies are in progress for the molecular basis of the structure/ function relationships of M-histatin 1 and human histatins to be understood, with respect to candidacidal and candidastatic activity in the oral cavity. REFERENCES

ATKINSON, J.C.; YEH, C.; OPPENHEIM, F.G.; BERMUDEZ, D.; BAUM, B.J.; and FOX, P.C. (1990): Elevation of Salivary Antimicrobial Proteins Following HIV-1 Infection, JAIDS 3:4148.

AUSTWICK, P.K.G.; PEPIN, G.; THOMSON, J.C.; and YARROW, D. (1966): Candida albicans and Other Yeasts Associated with Animal Disease. In: Symposium on Candida Infections, H.I. Winner and R. Hurley, Eds., Edinburgh: Livingstone, pp. 89-100. AZEN, E.A. (1973): Properties of Salivary Basic Proteins Showing Polymorphism, Biochem Genet 9:69-86. BAUM, B.J.; BIRD, J.L.; and LONGTON, R.W. (1977): Polyacrylamide Gel Electrophoresis of Human Salivary Histidine-rich Polypeptides, J Dent Res 56:1115-1118. BOWEN, W.H. (1981): Scoring Caries in Primates. In: Proceedings, Animal Models in Cariology, J.M. Tanzer, Ed., Sp. Supp. Microbiology Abstracts, Washington, D.C.: Information Retrieval, Inc., pp. 183-188. BUDTZ-JORGENSEN, E. (1975): Effects of Triamcinolone Acetonide on Experimental Oral Candidiasis in Monkeys, Scand J Dent Res 83:171-178. CHOU, P.Y. and FASMAN, G.D. (1979): Prediction of the Secondary Structure of Proteins from their Amino Acid Sequence, Adv Enzymol Relat Areas Mol Biol 47:45-148. GOVINDARAJULU, Z. (1988): Statistical Techniques in Bioassay, Basel: Karger, pp. 28-64. GOW, N.A.R. and GOODAY, G.W. (1987): Cytological Aspects of Dimorphism in Candida albicans, CRC Crit Rev Microbiol 15:7378. HAY, D.I.; BENNICK, A.; SCHLESINGER, D.H.; MINAGUCHI, K.; MADAPALLIMATTAM, G.; and SCHLUCKEBIER, S.K. (1988): The Primary Structure of Six Human Salivary Acidic Proline-rich Proteins (PRP-1, PRP-2, PRP-3, PRP-4, PIF-s and PIFf), Biochem J 255:15-21. HAY, D.I.; MORENO, E.C.; and SCHLESINGER, D.H. (1979): Phosphoprotein-inhibitors of Calcium Phosphate Precipitation from Salivary Secretions, Inorg Persp Biol Med 2:271-285. HAY, D.I.; SMITH, D.J.; SCHLUCKEBIER, S.K.; and MORENO, E.C. (1984): Relationship between Concentration of Human Salivary Statherin and Inhibition of Calcium Phosphate Precipitation in Stimulated Human Parotid Saliva, J Dent Res 63:857-863. HOLBROOK, I.B. and MOLAN, P.C. (1975): The Identification of a Peptide in Human Parotid Saliva Particularly Active in Enhancing the Glycolytic Activity of the Salivary Micro-organism, Biochem J 149:489-492. KAUFMANN, A.F. and QUIST, K.D. (1969): Thrush in a Rhesus Monkey: Report of a Case, Lab Anim Care 19:526-527. KORNMAN, K.S.; HOLT, S.C.; and ROBERTSON, P.B. (1981): The Microbiology of Ligature-induced Periodontitis in the Cynomolgus Monkey, J Peniodont Res 16:363-371. KYTE, J. and DOOLITTLE, R.F. (1982): A Simple Method for Displaying the Hydropathic Character of a Protein, J Mol Biol 157:105-132. LEVITZ, S.M. and DIAMOND, R.D. (1985): A Rapid Colorimetric Assay of Fungal Viability with the Tetrazolium Salt MTT, J Infect Dis 152:938-945.

Downloaded from jdr.sagepub.com at Karolinska Institutets Universitetsbibliotek on May 23, 2015 For personal use only. No other uses without permission.

Vol. 69 No. 11

STRUCTURE AND ACTIVI7Y OF MA CA QUE SALIVARY HISTA TIN

MacKAY, B.J.; DENEPITIYA, L.; IACONO, V.J.; KROST, S.P.; and POLLOCK, J.J. (1984): Growth Inhibitory and Bactericidal Effects of Human Parotid Salivary Histidine-rich Polypeptides on Streptococcus mutans, Infect Immun 44:695-701. MINAGUCHI, K. and BENNICK, A. (1989a): Genetics of Human Salivary Proteins, J Dent Res 68:2-15. ODDS, F.H. (1979): Candida and Candidiasis, Leicester: Leicester University Press, pp. 29-31. ODDS, F.H. (1980): A Simple System for the Presumptive Identification of Candida albicans and Differentiation of Strains within the Species, Sabouraudia 18:301-317. OPPENHEIM, F.G. (1989): Salivary Histidine-rich Proteins. In: Human Saliva: Clinical Chemistry and Microbiology, Vol. 1, J.O. Tenovuo, Ed., Florida: CRC Press, Inc., pp. 151-160. OPPENHEIM, F.G.; OFFNER, G.D.; and TROXLER, R.F. (1982): Phosphoproteins in the Parotid Saliva from the Subhuman Primate Macaca fascicularis, J Biol Chem 257:9271-9282. OPPENHEIM, F.G.; OFFNER, G.D.; and TROXLER, R.F. (1985): Amino Acid Sequence of a Proline-rich Phosphoglycoprotein from Parotid Secretion of the Subhuman Primate Macaca fascicularis, JBiol Chem 260:10671-10679. OPPENHEIM, F.G.; XU, T.; McMILLIAN, F.M.; LEVITZ, S.M.; DIAMOND, R.D.; OFFNER, G.D.; and TROXLER, R.F. (1988): Histatins, a Novel Family of Histidine-rich Proteins in Human Parotid Secretion. Isolation, Characterization, Primary Structure and Fungistatic Effects on Candida albicans, JBiol Chem 263:74727477. OPPENHEIM, F.G.; YANG, Y.; DIAMOND, R.D.; HYSLOP, D.; OFFNER, G.D.; and TROXLER, R.F. (1986): The Primary Structure and Functional Characterization of the Neutral Histidinerich Polypeptide from Human Parotid Saliva, JBiol Chem 261:11771182. PETERS, E.H.; GOODFRIEND, T.; and AZEN, E.A. (1977): Human Pb, Human Post-Pb, and Nonhuman Primate Pb Proteins: Immunological and Biochemical Relationships, Biochem Genet 15:947-962. POLLOCK, J.J.; DENEPITIYA, L.; MacKAY, B.J.; and IACONO,

1 723

V. (1984): Fungistatic and Fungicidal Activity of Human Parotid Salivary Histidine-rich Polypeptides on Candida albicans, Infect Immun 44:702-707. RAJ, P.A.; EDGERTON, M.E.; and LEVINE, M.J. (1990): Salivary Histatin 5: Dependence of Sequence, Chain Length, and Helical Conformation for Candidacidal Activity, J Biol Chem 265:38983905. SANTARPIA, R.P., III; BRANT, E.C.; LAI, K.; BRASSEUR, M.M.; HONG, A.L.; and POLLOCK, J.J. (1988): A Comparison of the Inhibition of Blastospore Viability and Germ Tube Development in Candida albicans by Histidine Peptides and Ketoconazole, Arch

Oral Biol 33:567-573. SCHLESINGER, D.H. and HAY, D.I. (1977): Complete Covalent Structure of Statherin, a Tyrosine-rich Acidic Peptide, which Inhibits Calcium Phosphate Precipitation from Human Parotid Saliva, J Biol Chem 252:1689-1695. SCHLESINGER, D.H.; HAY, D.I.; and LEVINE, M.J. (1989): Complete Primary Structure of Statherin, a Potent Inhibitor of Calcium Phosphate Precipitation, from the Saliva of the Monkey, Macaca arctoides, Int J Peptide Protein Res 34:374-380. SVANBORG, A. and SVANNERHOLM, L. (1961): Plasma Total Lipid, Cholesterol, Triglycerides, Phospholipids and Free Fatty Acids in a Healthy Scandinavian Population, Acta Med Scand 169:43-49. TROXLER, R.F.; OFFNER, G.D.; XU, T.; VANDERSPEK, J.C.; and OPPENHEIM, F.G. (1990): Structural Relationship between Human Salivary Histatins, J Dent Res 69:2-6. XU, L.; FISCHER, T.; and POLLOCK, J.J. (1989): Sequence Determination of Low Molecular Weight Salivary Histidine-rich Polypeptides from Electroblots, Peptide Res 2:373-375. XU, T. and OPPENHEIM, F.G. (1989): Further Characterization of Anti-Candida Activities of Histatins, J Dent Res 68:405, Abst. No. 1787. XU, T. and OPPENHEIM, F.G. (1990): Antibacterial Effects of Histatins on Different Mutans Streptococci Serotypes, J Dent Res 69:239, Abst. No. 1042.

Downloaded from jdr.sagepub.com at Karolinska Institutets Universitetsbibliotek on May 23, 2015 For personal use only. No other uses without permission.