- Email: [email protected]
Antimicrobial Peptide of Korean Native Goat Lactoferrin and Identification of the Part Essential for This Activity
Biochemical and Biophysical Research Communications 268, 333–336 (2000) doi:10.1006/bbrc.2000.2141, available online at http://www.idealibrary.com on
Antimicrobial Peptide of Korean Native Goat Lactoferrin and Identification of the Part Essential for This Activity Michito Kimura,* Myoung-Soo Nam,* ,1 Yoshifumi Ohkouchi,* Haruto Kumura,* Kei-ichi Shimazaki,* and Dea-Youl Yu† *Dairy Science Laboratory, Department of Animal Product Science, Graduate School of Agriculture, Hokkaido University, Sapporo 060-8589, Japan; and †Animal Molecular Physiology R.U., Korea Research Institute of Bioscience and Biotechnology, Taejeon 305-600, Korea
Received December 14, 1999
The antimicrobial activity of lactoferrin isolated from Korean native goat (KN goat) milk was studied and its antimicrobial domain was identified using synthetic peptides. Antimicrobial activity was assayed by a micro-method using 96-well microplates and a microplate reader. The amino acid sequence of the antimicrobial domain was suggested to be YQWQRRMRKLGAPSIT and this sequence corresponds to amino acid residues 20 to 35 of KN goat lactoferrin. Five peptides with certain amino acid residues deleted were synthesized in an effort to identify the residues essential for antimicrobial activity and it was found that the part with the sequence RRMRK (24 –28) is the region most important for this activity. On the other hand, the conformation of the peptides did not influence the antimicrobial activity. © 2000 Academic Press
Lactoferrin is an antimicrobial component of milk and it contributes to protect the infant from infectious disease (1). Other biological functions attributed to lactoferrin include roles in modulation of the inflammatory response, activation of the immune system, and control of myelopoiesis (2). The antimicrobial effect of lactoferrin is presumably the result of the ability of the protein to deprive bacteria of the iron essential for growth (3). In addition, lactoferrin has been shown to induce the release of substantial amounts of lipopolysaccharide from the surface of Gram-negative bacteria, altering the permeability properties of the outer membrane (4, 5). Bellamy et al. (6) reported the existence of antimicrobial domains near the N-terminus of bovine and human lactoferrins and named the isolated peptides lactoferricin The microbial killing effect of these peptides was ten to one-hundred times stronger than that of undigested lactoferrin. These active peptides 1
Present address: Department of Dairy Science, College of Agriculture, Chungnam National University, Taejon 305-764, Korea.
were found to display broad-spectrum antibacterial properties, having effectiveness against Gram-positive and Gram-negative bacteria, yeast and filamentous fungi (6, 7). Lactoferricin has been shown to have an affinity for cell membranes and may exert its lethal effect by disruption of essential membrane functions. It binds directly to lipopolysaccharide and disrupts the permeability barrier of the outer membrane of Gramnegative bacteria (8). Also, lactoferricin acts to disrupt the permeability properties of the cytoplasmic membrane (7). The existence of peptides possessing the antimicrobial activity of murine and goat lactoferrins was proposed by Vorland et al. (9). We have conducted a study examining the antibacterial activity of lactoferrin isolated from the Korean native goat (KN goat), a domestic animal kept genetically pure and not crossbred for 2,000 years in Korea, against E. coli O111. We tried to obtain an antimicrobial peptide from the hydrolysate generated by pepsin digestion of KN goat lactoferrin. Then, the amino acid residues essential for antimicrobial activity were examined in experiments using chemically synthesized peptides. MATERIALS AND METHODS Lactoferrin and its fragments. Lactoferrin previously isolated from milk of the KN goat (10) was used. Pepsin digestion of lactoferrin was done by the method described by Tomita et al. (11) with slight modifications as follows. Lactoferrin was dissolved in water (5%) and the pH was adjusted to 2.5. Then porcine pepsin (Sigma Chemical Co. Ltd., St. Louis, MO) was added at a final concentration of 0.15% (w/w). The hydrolysis reaction was performed at 37°C for 1 h and terminated by heating at 80°C for 10 min. The reaction mixture was neutralized by the addition of 1 M NaOH and the precipitate formed in the reaction mixture was removed by centrifugation. The supernatant was fractionated by reverse-phase HPLC using a CAPCELL PAK C18 SG300 column (4.6 mm ID ⫻ 250 mm, a product of Shiseido, Tokyo) and a mixture of solvent A (0.1% trifluoroacetic acid in water) and B (0.1% trifluoroacetic acid in acetonitrile). A convex gradient of solvent A and B (0 to 45% B) was applied to the column for 25 min at a flow rate of 1 ml/min and the absorbance of the eluate at 230 or 280 nm was monitored. Fractions were collected
333
0006-291X/00 $35.00 Copyright © 2000 by Academic Press All rights of reproduction in any form reserved.
Vol. 268, No. 2, 2000
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
a difference in only six amino acid residues (5 in the N-lobe and one in the C-lobe), comparing KN goat lactoferrin and Saanen goat lactoferrin (15, 16), the reason for the difference in antimicrobial activity should be confirmed through further experiments.
TABLE 1
Sequences of Chemically Synthesized Peptides and Their Antimicrobial Activity (MIC Value) against E. coli O111
1 2 3 4 5 6
Peptide
MIC (mg/ml)
YQWQRRMRKLGAPSIT YQWQRRM-KLGAPSIT --WQRRMRKLGAPSIT --WQRRM-KLGAPSIT YQWQRRMR-LGAPSIT YQWQ-RMRKLGAPSIT
0.2 na 1.0 na 0.6 na
Antimicrobial Active Peptide from KN Goat Lactoferrin
Note. Each “-” in the sequence indicates a deleted amino acid residue, compared with peptide 1. na, no activity.
and the acetonitrile was removed using a centrifugal evaporator. Then, the samples were dissolved in water, neutralized by adding an alkaline solution, and assayed for antibacterial activity. The amino acid sequence of the peptide showing the most potent activity was determined using an Applied Biosystems gas-phase sequencer 477A and an ABI chromatograph model 120A. Synthetic peptides. Peptides synthesized from Fmoc amino acid active derivatives were purchased from Sigma-Genosys-Japan (Sapporo). Their amino acid sequences are listed in Table 1. Assay of antimicrobial activity. E. coli O111 was used as the target strain to compare the antibacterial activities of lactoferrin and its peptides. The organism was incubated at 37°C in a basal medium of 1% (w/v) Bactopeptone (Difco Laboratories, Detroit, MI), pH 7.0. Stock solutions of the proteins or peptides tested were diluted with water and filter-sterilized and 0.05 ml of the solution was mixed with the same volume of the basal medium. To examine the cell growth inhibition effects, E. coli at an initial cell density of 2 ⫻ 10 5 CFU/ml was cultured at 37°C for 4 to 5 h in wells of a 96-well microplate containing 0.05 ml of the basal medium with or without various amounts of lactoferrin or the peptides. After incubation, turbidimetric measurements were performed using a microplate reader at 600 nm. The MIC value was taken to be the lowest concentration of the test substance that caused complete inhibition of growth of E. coli. Circular dichroic measurements. Circular dichroism (CD) recordings were made on a JASCO Model J-500 dichrograph (Japan Spectroscopic Co. Ltd., Tokyo, Japan). The solvent used for these measurements was 0.05 M sodium phosphate buffer, pH 7.0. All recordings were made at room temperature. Assays of concentration. Lactoferrin concentrations were determined on the basis of the extinction coefficient of bovine lactoferrin at 280 nm, which was taken to be 12.7 (1%, path length 10 mm) (12). Peptide concentrations were determined by Lowry’s method (13). For preparation of the calibration curve, peptide No. 3 (in Table 1), which has one tryptophan residue, was used. The concentration of peptide No. 3 was determined with the molar extinction coefficient taken to be 5,550 at 278 nm (14), due to the indole group of tryptophan.
RESULTS AND DISCUSSION Antimicrobial Activity of KN Goat Lactoferrin The MIC of KN goat lactoferrin was found to be 5 mg/ml, whereas the MIC of bovine and human lactoferrins have been reported to be 2 and 3 mg/ml (6), respectively. From our preliminary experiments, goat (Saanen), sheep and horse lactoferrins showed no activity at the concentration of 7.5 mg/ml (15). As there is
The antibacterial properties of the peptides generated by pepsin digestion of KN goat lactoferrin were examined using E. coli O111 as the target strain. The MIC of KN goat lactoferrin hydrolysate was 100 g/ml and this value is the same as that reported for human lactoferrin hydrolysate (6). On the other hand, the MIC of bovine lactoferrin hydrolysate is reported to be 6 g/ml (6). When the hydrolysate prepared by pepsin digestion of KN goat lactoferrin was fractionated by reverse-phase HPLC, about 30 fractions were obtained as shown in Fig. 1A. The antibacterial activity of each fraction was examined. The fraction showing the most potent activity was eluted at a position corresponding to 35% solvent B and the active constituent was further purified, as shown in Fig. 1B, to determine its amino acid sequence. The sequence of this peptide was found to be YQWQRRMRKLGAPSIT and this sequence corresponds to the sequence of residues 20 to 35 in the N-lobe of KN goat lactoferrin. It has been reported that the sequences of the antimicrobial peptides from human and bovine lactoferrin are VSQPEATKCFQWQRNMRKVRGPPVSCIKRDSPIQCI (12– 47) and FKCRRWQWRMKKLGAPSITCVRRAFA (19 and 36), respectively (6), and in each of these peptides two cysteine residues are cross-linked forming a looped structure. The sequence of the antimicrobial peptide from KN goat lactoferrin showed 75% and 44% similarity with the sequences of the regions between the two cysteine residues of bovine and human lactoferricin, respectively. We then synthesized the peptide YQWQRRMRKLGAPSIT in an effort to confirm whether this peptide has antimicrobial activity. Whereas there is a difference in five amino acid residues in the N-lobe comparing KN goat lactoferrin and Saanen goat lactoferrin (15), none of these substitutions occurs in this region which shows antimicrobial activity. Antimicrobial Activity of Synthetic Peptides The MIC of the synthesized peptide YQWQRRMRKLGAPSIT was 200 g/ml as shown in Fig. 2 and Table 1. In an effort to identify the amino acid residues essential for the antimicrobial activity, five derivatives were synthesized and their MIC values were compared as shown in Fig. 2 and Table 1. Peptides 2, 4 and 6 did not show any antimicrobial activity even at concentrations greater than 1.0 mg/ml. All of these peptides have an arginine residue deleted in the cluster of positively charged residues between the 24th and 28th amino
334
Vol. 268, No. 2, 2000
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 2. Concentration dependence of the antibacterial activity of synthetic peptides corresponding to various regions of KN goat lactoferrin. The symbols E, F, ‚, Œ, 䊐, and ■ indicate peptide 1, peptide 2, peptide 3, peptide 4, peptide 5, and peptide 6, respectively.
Effect of Peptide Conformation on Antimicrobial Activity In an effort to determine whether the antimicrobial activity is dependent on peptide conformation, the gross conformations of the peptides were investigated through circular dichroic spectral measurements. As
FIG. 1. Reverse-phase HPLC fractionation of peptides generated by pepsin hydrolysis of KN goat lactoferrin (A) and chromatographic profile upon reanalysis (B) of the peak shown by an arrow. Chromatographic conditions are described in the text.
acid residues. Therefore, it seems reasonable to conclude that the part with the sequence RRMRK (24 –28) is essential for the activity against the bacterial strain used in this experiment. In the case of bovine lactoferricin, it is reported that the residues RRWQWR (20 – 25) are important for the antimicrobial activity (17). The antimicrobial active peptide derived from KN goat lactoferrin contains a heparin-binding site ( 22WQRRMRKLGA 31) as reported previously (10). Vorland et al. (9) have reported that a linear peptide with the sequence SKCYQWQRRMRKLGAPSITCVRRTS derived from goat lactoferrin showed no antimicrobial activity against E. coli or S. aureus even at concentrations greater than 200 g/ml. The reason for this difference in findings is not clear, but it may be because they used a different strain of E. coli or because the method used to assay the antimicrobial activity was different from ours.
FIG. 3. Circular dichroic spectra of six synthetic peptides in the wavelength range from 200 nm to 250 nm. The symbols E, F, ‚, Œ, 䊐, and ■ indicate the spectra of peptides 1 to 6, respectively. (Inset) The spectrum is that of peptide No. 1 in 50% (v/v) trifluoroethanol. The lateral axis means ellipticity ([], deg 䡠 cm 2 䡠 decimole ⫺1).
335
Vol. 268, No. 2, 2000
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
shown in Fig. 3, the circular dichroic spectra were found to be almost the same for all of the peptides. Peptides No. 3 and 4, each of which has no tyrosine residue, showed no positive peak at 228 nm. Few differences were observed at wavelengths in the range of 200 to 205 nm, none of the peptides showed a helical conformation, and each showed 50 to 70% unordered structure. The secondary structure values were calculated by the method of Yang et al. (18). It seems reasonable that such short chain peptides have little helical structure, but mostly intermolecular -structure and unordered structure. In the presence of trifluoroethanol, a solvent which promotes helix-formation, the CD spectra of the peptides showed 5 to 7% helix content and an example is shown in Fig. 3 (inserted spectrum of peptide 1). Considering the similarity in terms of conformational characteristics, it seems evident that conformational effects did not account for the differences in antimicrobial activity of the synthetic peptides listed in Table 1. Therefore, it is concluded that the positively charged groups are more important for the antimicrobial activity than conformational effects. ACKNOWLEDGMENTS This work was supported by a grant (GG082S) from the Korean Ministry of Science and Technology. We are very grateful to Dr. K. Mikawa and Mr. S. Murata for useful discussions and technical support, and to Ms. H. Matsumoto (Center for Instrumental Analysis, Hokkaido University) for amino acid sequence analysis.
REFERENCES 1. Reiter, B. (1978) Ann. Rech. Vet. 9, 205–224. 2. Brock, J. (1995) Immunol. Today 16, 417– 419.
3. Arnold, R. R., Cole, M. F., and McGhee, J. R. (1977) Science 197, 263–265. 4. Ellison, R. T., Giehl, T. J., and LaForce, F. M. (1988) Infect. Immun. 56, 2774 –2781. 5. Ellison, R. T., LaForce, F. M., Giehl, T. J., Boose, D. S., and Dunn, B. E. (1990) J. Gen. Microbiol. 136, 1437–1446. 6. Bellamy, W., Takase, M., Yamauchi, K., Wakabayashi, H., Kawase, K., and Tomita, M. (1992) Biochim. Biophys. Acta 1121, 130 –136. 7. Bellamy, W., Wakabayashi, H., Takase, M., Kawase, K., Shimamura, S., and Tomita, M. (1993) Med. Microbiol. Immunol. 182, 97–105. 8. Yamauchi, K., Tomita, M., Giehl, T. J., and Ellison, R. T. (1993) Infect. Immun. 61, 719 –728. 9. Vorland, L. H., Ulvatne, H., Andersen, J., Haukland, H., Rekdal, O., Svendsen, J. S., and Gutteberg, T. J. (1998) Scand. J. Infect. Dis. 30, 513–517. 10. Nam, M. S., Shimazaki, K., Kumura, H., Lee, K. K., and Yu, D. Y. (1999) Comp. Biochem. Physiol. B123, 201–208. 11. Tomita, M., Bellamy, W., Takase, M., Yamauchi, K., Wakabayashi, H., and Kawase, K. (1991) J. Dairy Sci. 74, 4137– 4142. 12. Fasman, G. D. (1976) 3rd ed., CRC Press, Cleveland. 13. Lowry, O. H., Rosebrough, N. J., Farr, A. L., and Randall, R. J. (1951) J. Biol. Chem. 193, 265–275. 14. Greenstein, J. P., and Winitz, M. (1961) Chemistry of the Amino Acids, John Wiley & Sons, Inc., New York and London. 15. Lee, T. H., Shimazaki, K., Yu, S. L., Nam, M. S., Kim, S. J., Lee, K. K., and Yu, D. Y. (1997) Anim. Genet. 28, 367–369. 16. Le Provost, F., Nocart, M., Guerin, G., and Martin, P. (1994) Biochem. Biophys. Res. Commun. 203, 1324 –1332. 17. Tomita, M., Takase, M., Bellamy, W., and Shimamura, S. (1994) Acta Paediat. Japonica 36, 585–591. 18. Yang, J. T., Wu, C.-S., and Martinez, H. M. (1986) in Methods in Enzymology (Hirs, C. H. W., and Timasheff, S. N., Eds.), Vol. 130, pp. 208 –269, Academic Press, New York.
336
Antimicrobial Peptide of Korean Native Goat Lactoferrin and Identification of the Part Essential for This Activity Michito Kimura,* Myoung-Soo Nam,* ,1 Yoshifumi Ohkouchi,* Haruto Kumura,* Kei-ichi Shimazaki,* and Dea-Youl Yu† *Dairy Science Laboratory, Department of Animal Product Science, Graduate School of Agriculture, Hokkaido University, Sapporo 060-8589, Japan; and †Animal Molecular Physiology R.U., Korea Research Institute of Bioscience and Biotechnology, Taejeon 305-600, Korea
Received December 14, 1999
The antimicrobial activity of lactoferrin isolated from Korean native goat (KN goat) milk was studied and its antimicrobial domain was identified using synthetic peptides. Antimicrobial activity was assayed by a micro-method using 96-well microplates and a microplate reader. The amino acid sequence of the antimicrobial domain was suggested to be YQWQRRMRKLGAPSIT and this sequence corresponds to amino acid residues 20 to 35 of KN goat lactoferrin. Five peptides with certain amino acid residues deleted were synthesized in an effort to identify the residues essential for antimicrobial activity and it was found that the part with the sequence RRMRK (24 –28) is the region most important for this activity. On the other hand, the conformation of the peptides did not influence the antimicrobial activity. © 2000 Academic Press
Lactoferrin is an antimicrobial component of milk and it contributes to protect the infant from infectious disease (1). Other biological functions attributed to lactoferrin include roles in modulation of the inflammatory response, activation of the immune system, and control of myelopoiesis (2). The antimicrobial effect of lactoferrin is presumably the result of the ability of the protein to deprive bacteria of the iron essential for growth (3). In addition, lactoferrin has been shown to induce the release of substantial amounts of lipopolysaccharide from the surface of Gram-negative bacteria, altering the permeability properties of the outer membrane (4, 5). Bellamy et al. (6) reported the existence of antimicrobial domains near the N-terminus of bovine and human lactoferrins and named the isolated peptides lactoferricin The microbial killing effect of these peptides was ten to one-hundred times stronger than that of undigested lactoferrin. These active peptides 1
Present address: Department of Dairy Science, College of Agriculture, Chungnam National University, Taejon 305-764, Korea.
were found to display broad-spectrum antibacterial properties, having effectiveness against Gram-positive and Gram-negative bacteria, yeast and filamentous fungi (6, 7). Lactoferricin has been shown to have an affinity for cell membranes and may exert its lethal effect by disruption of essential membrane functions. It binds directly to lipopolysaccharide and disrupts the permeability barrier of the outer membrane of Gramnegative bacteria (8). Also, lactoferricin acts to disrupt the permeability properties of the cytoplasmic membrane (7). The existence of peptides possessing the antimicrobial activity of murine and goat lactoferrins was proposed by Vorland et al. (9). We have conducted a study examining the antibacterial activity of lactoferrin isolated from the Korean native goat (KN goat), a domestic animal kept genetically pure and not crossbred for 2,000 years in Korea, against E. coli O111. We tried to obtain an antimicrobial peptide from the hydrolysate generated by pepsin digestion of KN goat lactoferrin. Then, the amino acid residues essential for antimicrobial activity were examined in experiments using chemically synthesized peptides. MATERIALS AND METHODS Lactoferrin and its fragments. Lactoferrin previously isolated from milk of the KN goat (10) was used. Pepsin digestion of lactoferrin was done by the method described by Tomita et al. (11) with slight modifications as follows. Lactoferrin was dissolved in water (5%) and the pH was adjusted to 2.5. Then porcine pepsin (Sigma Chemical Co. Ltd., St. Louis, MO) was added at a final concentration of 0.15% (w/w). The hydrolysis reaction was performed at 37°C for 1 h and terminated by heating at 80°C for 10 min. The reaction mixture was neutralized by the addition of 1 M NaOH and the precipitate formed in the reaction mixture was removed by centrifugation. The supernatant was fractionated by reverse-phase HPLC using a CAPCELL PAK C18 SG300 column (4.6 mm ID ⫻ 250 mm, a product of Shiseido, Tokyo) and a mixture of solvent A (0.1% trifluoroacetic acid in water) and B (0.1% trifluoroacetic acid in acetonitrile). A convex gradient of solvent A and B (0 to 45% B) was applied to the column for 25 min at a flow rate of 1 ml/min and the absorbance of the eluate at 230 or 280 nm was monitored. Fractions were collected
333
0006-291X/00 $35.00 Copyright © 2000 by Academic Press All rights of reproduction in any form reserved.
Vol. 268, No. 2, 2000
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
a difference in only six amino acid residues (5 in the N-lobe and one in the C-lobe), comparing KN goat lactoferrin and Saanen goat lactoferrin (15, 16), the reason for the difference in antimicrobial activity should be confirmed through further experiments.
TABLE 1
Sequences of Chemically Synthesized Peptides and Their Antimicrobial Activity (MIC Value) against E. coli O111
1 2 3 4 5 6
Peptide
MIC (mg/ml)
YQWQRRMRKLGAPSIT YQWQRRM-KLGAPSIT --WQRRMRKLGAPSIT --WQRRM-KLGAPSIT YQWQRRMR-LGAPSIT YQWQ-RMRKLGAPSIT
0.2 na 1.0 na 0.6 na
Antimicrobial Active Peptide from KN Goat Lactoferrin
Note. Each “-” in the sequence indicates a deleted amino acid residue, compared with peptide 1. na, no activity.
and the acetonitrile was removed using a centrifugal evaporator. Then, the samples were dissolved in water, neutralized by adding an alkaline solution, and assayed for antibacterial activity. The amino acid sequence of the peptide showing the most potent activity was determined using an Applied Biosystems gas-phase sequencer 477A and an ABI chromatograph model 120A. Synthetic peptides. Peptides synthesized from Fmoc amino acid active derivatives were purchased from Sigma-Genosys-Japan (Sapporo). Their amino acid sequences are listed in Table 1. Assay of antimicrobial activity. E. coli O111 was used as the target strain to compare the antibacterial activities of lactoferrin and its peptides. The organism was incubated at 37°C in a basal medium of 1% (w/v) Bactopeptone (Difco Laboratories, Detroit, MI), pH 7.0. Stock solutions of the proteins or peptides tested were diluted with water and filter-sterilized and 0.05 ml of the solution was mixed with the same volume of the basal medium. To examine the cell growth inhibition effects, E. coli at an initial cell density of 2 ⫻ 10 5 CFU/ml was cultured at 37°C for 4 to 5 h in wells of a 96-well microplate containing 0.05 ml of the basal medium with or without various amounts of lactoferrin or the peptides. After incubation, turbidimetric measurements were performed using a microplate reader at 600 nm. The MIC value was taken to be the lowest concentration of the test substance that caused complete inhibition of growth of E. coli. Circular dichroic measurements. Circular dichroism (CD) recordings were made on a JASCO Model J-500 dichrograph (Japan Spectroscopic Co. Ltd., Tokyo, Japan). The solvent used for these measurements was 0.05 M sodium phosphate buffer, pH 7.0. All recordings were made at room temperature. Assays of concentration. Lactoferrin concentrations were determined on the basis of the extinction coefficient of bovine lactoferrin at 280 nm, which was taken to be 12.7 (1%, path length 10 mm) (12). Peptide concentrations were determined by Lowry’s method (13). For preparation of the calibration curve, peptide No. 3 (in Table 1), which has one tryptophan residue, was used. The concentration of peptide No. 3 was determined with the molar extinction coefficient taken to be 5,550 at 278 nm (14), due to the indole group of tryptophan.
RESULTS AND DISCUSSION Antimicrobial Activity of KN Goat Lactoferrin The MIC of KN goat lactoferrin was found to be 5 mg/ml, whereas the MIC of bovine and human lactoferrins have been reported to be 2 and 3 mg/ml (6), respectively. From our preliminary experiments, goat (Saanen), sheep and horse lactoferrins showed no activity at the concentration of 7.5 mg/ml (15). As there is
The antibacterial properties of the peptides generated by pepsin digestion of KN goat lactoferrin were examined using E. coli O111 as the target strain. The MIC of KN goat lactoferrin hydrolysate was 100 g/ml and this value is the same as that reported for human lactoferrin hydrolysate (6). On the other hand, the MIC of bovine lactoferrin hydrolysate is reported to be 6 g/ml (6). When the hydrolysate prepared by pepsin digestion of KN goat lactoferrin was fractionated by reverse-phase HPLC, about 30 fractions were obtained as shown in Fig. 1A. The antibacterial activity of each fraction was examined. The fraction showing the most potent activity was eluted at a position corresponding to 35% solvent B and the active constituent was further purified, as shown in Fig. 1B, to determine its amino acid sequence. The sequence of this peptide was found to be YQWQRRMRKLGAPSIT and this sequence corresponds to the sequence of residues 20 to 35 in the N-lobe of KN goat lactoferrin. It has been reported that the sequences of the antimicrobial peptides from human and bovine lactoferrin are VSQPEATKCFQWQRNMRKVRGPPVSCIKRDSPIQCI (12– 47) and FKCRRWQWRMKKLGAPSITCVRRAFA (19 and 36), respectively (6), and in each of these peptides two cysteine residues are cross-linked forming a looped structure. The sequence of the antimicrobial peptide from KN goat lactoferrin showed 75% and 44% similarity with the sequences of the regions between the two cysteine residues of bovine and human lactoferricin, respectively. We then synthesized the peptide YQWQRRMRKLGAPSIT in an effort to confirm whether this peptide has antimicrobial activity. Whereas there is a difference in five amino acid residues in the N-lobe comparing KN goat lactoferrin and Saanen goat lactoferrin (15), none of these substitutions occurs in this region which shows antimicrobial activity. Antimicrobial Activity of Synthetic Peptides The MIC of the synthesized peptide YQWQRRMRKLGAPSIT was 200 g/ml as shown in Fig. 2 and Table 1. In an effort to identify the amino acid residues essential for the antimicrobial activity, five derivatives were synthesized and their MIC values were compared as shown in Fig. 2 and Table 1. Peptides 2, 4 and 6 did not show any antimicrobial activity even at concentrations greater than 1.0 mg/ml. All of these peptides have an arginine residue deleted in the cluster of positively charged residues between the 24th and 28th amino
334
Vol. 268, No. 2, 2000
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 2. Concentration dependence of the antibacterial activity of synthetic peptides corresponding to various regions of KN goat lactoferrin. The symbols E, F, ‚, Œ, 䊐, and ■ indicate peptide 1, peptide 2, peptide 3, peptide 4, peptide 5, and peptide 6, respectively.
Effect of Peptide Conformation on Antimicrobial Activity In an effort to determine whether the antimicrobial activity is dependent on peptide conformation, the gross conformations of the peptides were investigated through circular dichroic spectral measurements. As
FIG. 1. Reverse-phase HPLC fractionation of peptides generated by pepsin hydrolysis of KN goat lactoferrin (A) and chromatographic profile upon reanalysis (B) of the peak shown by an arrow. Chromatographic conditions are described in the text.
acid residues. Therefore, it seems reasonable to conclude that the part with the sequence RRMRK (24 –28) is essential for the activity against the bacterial strain used in this experiment. In the case of bovine lactoferricin, it is reported that the residues RRWQWR (20 – 25) are important for the antimicrobial activity (17). The antimicrobial active peptide derived from KN goat lactoferrin contains a heparin-binding site ( 22WQRRMRKLGA 31) as reported previously (10). Vorland et al. (9) have reported that a linear peptide with the sequence SKCYQWQRRMRKLGAPSITCVRRTS derived from goat lactoferrin showed no antimicrobial activity against E. coli or S. aureus even at concentrations greater than 200 g/ml. The reason for this difference in findings is not clear, but it may be because they used a different strain of E. coli or because the method used to assay the antimicrobial activity was different from ours.
FIG. 3. Circular dichroic spectra of six synthetic peptides in the wavelength range from 200 nm to 250 nm. The symbols E, F, ‚, Œ, 䊐, and ■ indicate the spectra of peptides 1 to 6, respectively. (Inset) The spectrum is that of peptide No. 1 in 50% (v/v) trifluoroethanol. The lateral axis means ellipticity ([], deg 䡠 cm 2 䡠 decimole ⫺1).
335
Vol. 268, No. 2, 2000
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
shown in Fig. 3, the circular dichroic spectra were found to be almost the same for all of the peptides. Peptides No. 3 and 4, each of which has no tyrosine residue, showed no positive peak at 228 nm. Few differences were observed at wavelengths in the range of 200 to 205 nm, none of the peptides showed a helical conformation, and each showed 50 to 70% unordered structure. The secondary structure values were calculated by the method of Yang et al. (18). It seems reasonable that such short chain peptides have little helical structure, but mostly intermolecular -structure and unordered structure. In the presence of trifluoroethanol, a solvent which promotes helix-formation, the CD spectra of the peptides showed 5 to 7% helix content and an example is shown in Fig. 3 (inserted spectrum of peptide 1). Considering the similarity in terms of conformational characteristics, it seems evident that conformational effects did not account for the differences in antimicrobial activity of the synthetic peptides listed in Table 1. Therefore, it is concluded that the positively charged groups are more important for the antimicrobial activity than conformational effects. ACKNOWLEDGMENTS This work was supported by a grant (GG082S) from the Korean Ministry of Science and Technology. We are very grateful to Dr. K. Mikawa and Mr. S. Murata for useful discussions and technical support, and to Ms. H. Matsumoto (Center for Instrumental Analysis, Hokkaido University) for amino acid sequence analysis.
REFERENCES 1. Reiter, B. (1978) Ann. Rech. Vet. 9, 205–224. 2. Brock, J. (1995) Immunol. Today 16, 417– 419.
3. Arnold, R. R., Cole, M. F., and McGhee, J. R. (1977) Science 197, 263–265. 4. Ellison, R. T., Giehl, T. J., and LaForce, F. M. (1988) Infect. Immun. 56, 2774 –2781. 5. Ellison, R. T., LaForce, F. M., Giehl, T. J., Boose, D. S., and Dunn, B. E. (1990) J. Gen. Microbiol. 136, 1437–1446. 6. Bellamy, W., Takase, M., Yamauchi, K., Wakabayashi, H., Kawase, K., and Tomita, M. (1992) Biochim. Biophys. Acta 1121, 130 –136. 7. Bellamy, W., Wakabayashi, H., Takase, M., Kawase, K., Shimamura, S., and Tomita, M. (1993) Med. Microbiol. Immunol. 182, 97–105. 8. Yamauchi, K., Tomita, M., Giehl, T. J., and Ellison, R. T. (1993) Infect. Immun. 61, 719 –728. 9. Vorland, L. H., Ulvatne, H., Andersen, J., Haukland, H., Rekdal, O., Svendsen, J. S., and Gutteberg, T. J. (1998) Scand. J. Infect. Dis. 30, 513–517. 10. Nam, M. S., Shimazaki, K., Kumura, H., Lee, K. K., and Yu, D. Y. (1999) Comp. Biochem. Physiol. B123, 201–208. 11. Tomita, M., Bellamy, W., Takase, M., Yamauchi, K., Wakabayashi, H., and Kawase, K. (1991) J. Dairy Sci. 74, 4137– 4142. 12. Fasman, G. D. (1976) 3rd ed., CRC Press, Cleveland. 13. Lowry, O. H., Rosebrough, N. J., Farr, A. L., and Randall, R. J. (1951) J. Biol. Chem. 193, 265–275. 14. Greenstein, J. P., and Winitz, M. (1961) Chemistry of the Amino Acids, John Wiley & Sons, Inc., New York and London. 15. Lee, T. H., Shimazaki, K., Yu, S. L., Nam, M. S., Kim, S. J., Lee, K. K., and Yu, D. Y. (1997) Anim. Genet. 28, 367–369. 16. Le Provost, F., Nocart, M., Guerin, G., and Martin, P. (1994) Biochem. Biophys. Res. Commun. 203, 1324 –1332. 17. Tomita, M., Takase, M., Bellamy, W., and Shimamura, S. (1994) Acta Paediat. Japonica 36, 585–591. 18. Yang, J. T., Wu, C.-S., and Martinez, H. M. (1986) in Methods in Enzymology (Hirs, C. H. W., and Timasheff, S. N., Eds.), Vol. 130, pp. 208 –269, Academic Press, New York.
336