Potent Antibacterial Peptides Generated by Pepsin Digestion of Bovine Lactoferrin MAMORU TOMITA, WAYNE BELLAMY, MITSUNORI TAKASE, KOJi YAMAUCHI, HIAOYUKI WAKABAYASHI, and KOUZOU KAWASE Nutritional Science Laboratory Morinaga Milk industry Go. Ltd. 183-5 Higashihara. Zama City, Kanagawa 228, Japan
mals, is a main antimicrobial component of colostrum and milk that contributes to protect the infant from infectious disease (8, 9, 10). The antimicrobial activity of lactofenin is known to be dependent on its iron-free state and is commonly attributed to its ability to bind and sequester iron, producing an irondeficient environment that limits microbial growth (8, 9). However, several studies have suggested that iron-free lactoferrin has bactericidal activity and kills sensitive bacteria by a mechanism distinct from sequestering of iron that involves its direct interaction with the bacterial cell surface (2, 3, 4, 5). Because few investigators have examined the antimicrobial properties of enzyme-hydrolyzed lactoferrin (6, 14), and because no active peptides have been reported previously, it is uncertain whether the entire lactoferrin molecule is required for this bactericidal effect. In the present study, to determine whether active peptides are produced from bovine lactoferrin, various commercially available proteases were employed to digest this protein, and the hydrolysates were assayed for antibacterial activity.
ABSTRACT
The antibacterial properties of enzymatic hydrolysates of bovine lactofenin were examined to determine whether active peptides are produced from this protein. Hydrolysates prepared by cleavage of lactoferrin with porcine pepsin, cod pepsin, or acid protease from Penicillium duponti showed strong activity against Escherichia coli 0 1 11, whereas hydrolysates produced by trypsin, papain, or other neutral proteases were much less active. Low moIecular weight peptides generated by porcine pepsin cleavage of lactoferrin showed broad-spectrum antibacterial activity, inhibiting the growth of a number of Gram-negative and Gram-positive species, including strains that were resistant to native lactoferrin. The antibacterial potency of the hydrolysate was at least eightfold greater than that of undigested lactoferrin with all strains tested. The active peptides retained their activity in the presence of added iron, unlike native lactofenin. The effect of the hydrolysate was bactericidal as indicated by a rapid loss of viability of E. coli 0111. The lactoferrin hydrolysate described in the present study has oommercial value as a natural preservative agent for use in foods and cosmetics, and as a functional component of new clinical foods for prevention or treatment of gastrointestinal disease. (Key words: lactoferrin, antibacterial peptides, pepsin digestion)
MATERIALS AND METHODS Preparation of Lactoterrin
Bovine lactofenin was prepared from fresh skim milk (Morinaga Milk Industry Co.,Zama City, Jpn.) by the method of Law and Reiter (12). The iron content of the lactofenin was 15% of saturation as determined by atomic absorption spectrometry.
INTRODUCTION
Lactofenin, an iron-binding glycoprotein present in most mucosal secretions of mam-
Received April 29. 1991. Accepted July 1, 1991. 1991 J Dairy Sci 74:4137-4142
Proteolytic Digestion of Lactoferrln
For hydrolysis of lactofenin, the following enzymes were used: porcine pepsin (EC 3.4.23.1; 10 units/mg; Difco Laboratories, Detroit, Ml), porcine trypsin (EC 3.4.21.4; loo0 units/mg; Sigma Chemical Co., St.
4137
4138
TOMITA
Louis, MO), PD enzyme (Penicillium duponri) (10 unitdmg; Seishin Pharmaceutical Co., Noda City, Jpn), Protease A (10 units/mg; Amano Pharmaceutical Co., Nagoya, Jpn), F'rotease P (10 units/mg; Amano Pharmaceutical Co.,Nagoya, Jpn), Actinase AS (250 units/ mg; K a k a Pharmaceutical Co.,Tokyo, Jpn), cod pepsin (EC 3.4.23.1; 10 units/mg; Nagase Biochemicals Co., Tokyo, Jpn), papain (EC 3.4.22.2; 50 unitdmg; Nagase Biochemicals Co., Tokyo, Jpn), and Bioprase (20 UnitS/mg; Nagase Biochemicals Co., Tokyo, Jpn). Bovine lactofenin was dissolved in distilled water at 5% (wthol), and the pH was adjusted to 2.5 or 7.0. Each protease was added at a final concentration of 3% (wt/wt of substrate). The hydrolysis reaction was performed at 37'C for 4 h, unless otherwise indicated, and terminated by heating at 80'C for 15 min. Reaction mixtures containing pepsin or PD enzyme were neutralized by addition of 1N NaOH. The precipitate of insoluble peptides formed (less than 10% of total protein) in each reaction mixture was removed by centrifugation at 15,000 x g, and the supernatant was retained and freezedried. The resulting lyophilized powder was used in the experiments. Cultures and Growth Condltlons
Escherichia coli 0111, a pathogenic intestinal bacterium obtained from the Institute of Medical Sciences at the University of Tokyo, Japan, was used routinely to investigate the antibacterial activity of lactoferrin and its derivatives. The organism was cultured at 37'C in a basal medium of 1% (wt/vol) Bactopeptone (Difco Laboratories, Detroit, MI), pH 6.8, or in peptone-yeast-glucose medium (1% Bactopeptone, 1% glucose, .05% yeast extract), pH 6.8. Stock solutions of lactoferrin or lactoferrin hydrolysate were prepared in distilled water and filter-sterilized before addition to these media. To examine the effect of added iron on antibacterial activity, the basal medium was supplemented with .1 mM FeSO4. Other bacterial strains were cultured in a similar manner. The extent of bacterial growth was determined by monitoring the absorbance of cultures at 660 nm (model U-3200 spectrophotometer, Hitachi, Jpn). Journal of Dairy Science Vol. 74, No. 12, 1991
ET AL. Assays of
Antlbacterlal Actlvlty
For determination of minimal inhibitory concentration, each strain was cultured for 16 to 20 h in a series of tubs containing 2.0 ml of peptone-yeast-glucose medium with various concentrations of lactoferrin (0 to 8.0 m d d ) or lactoferrin hydrolysate (0 to 1.6 mg/ml). A standard inoculum of logarithmic phase cells was added to each tube at a final concentration of 106 cfu/ml. The minimal inhibitory concentration was taken as the lowest concentration of lactoferrin or lactoferrin hydrolysate that caused complete inhibition of growth. For assay of bactericidal activity, suspensions of logarithmic phase E. coli 0111 (106 cfu/d) in basal medium with or without lactoferrin hydrolysate were incubated at 37'C in a shakerincubator water bath. After the indicated time, serial 10-fold dilutions were prepared in basal medium and plated onto plate count agar (Eiken Chemical Co.,Tokyo, Jpn) for determination of colony-forming units. Electrophoresls
Sodium dodecyl sulfate-PAGE was performed using a precast 10 to 25% linear gradient polyacrylamide gel and Laemmli buffer kit under conditions as recommended by the manufacturer (Daiichi Chemical Co.,Tokyo, Jpn). Molecular weight markers used were galactosidase (1 16,000), phosphorylase B (93,000), bovine s e ru m albumin (66,000), ovalbumin (45,000), soybean trypsin inhibitor (22,000), and lysozyme (14,000). HPLC Fractionation of Peptides
The antibacterial hydrolysate prepared by porcine pepsin cleavage of bovine lactoferrin was fractionated by reversephase HPLC using a column of TSK-GEL12Or (6.0 x 150 mm; Tosoh, Tokyo, Jpn) and a mixture of eluents A (.05% trifluoroacetic acid) and B (90% acetonitrile in .05% trifluoroacetic acid). Initially, a mix8020 of eluents A B was applied to the column for 10 min, followed by a linear gradient from 80:20 to 40:60 of eluent A. eluent B for 30 min at a flow rate of .8 ml/min. Fractions representing each of the peptide peaks were collected, dried under vacuum in a centrifugal evaporator, and assayed for antibacterial activity.
ANTIBACTERIAL HYDROLYSATE
RESULTS AND DISCUSSION
4139
OF LACTOFERRIN
Pepsln Hydrolysis of Lactoferrin
To examine the susceptibility of bovine lactoferrin to gastric pepsin cleavage, the protein was treated at pH 2.5 with porcine pepsin for 4 In a preliminary experiment, to determine h at 37'C, and samples were removed at whether antibacterial peptides are generated by 30-min intervals. Under the conditions emproteolytic cleavage of bovine lactofenin, hy- ployed, the reaction appeared to be complete drolysates prepared using various commer- within 30 min, as indicated by the SDS-PAGE cially available enzymes were tested for activ- profile of the reaction products, and by this ity against E. coli 0111 (Table 1). Undigested time most peptides in the hydrolysate had lactoferrin exhibited no antibacterial activity at molecular weights of less than 6000 (Figure 1). .5 mg/ml, however, at the same concentration, These results are consistent with previous hydrolysates produced at acid pH by porcine reports indicating that both bovine (13) and pepsin, cod pepsin, or the aspartic protease of human (7, 13) lactoferrin are readily cleaved P. duponti profoundly reduced the viability of by gastric pepsin under acidic conditions. the test strain. Hydrolysates produced at neutral pH by trypsin and other proteases, except Antibacterlal Properties papain, also caused some inhibition of growth, of the Lactoferrin Hydrolysate but the antibacterial effect was much weaker The low molecular weight peptides of bothan observed with the pepsin hydrolysate. vine lactoferrin obtained after 30 min of pepsin These observations suggest that potent antibac- treatment completely inhibited the growth of terial peptides of bovine l a c t o f e e are gener- E. coli 0111 at concentrations of .25 mdml or ated by enzymes having a cleavage site speci- more. The same minimal inhibitory concentraficity similar to porcine pepsin A (EC tion was observed after 4 h of pepsin treat3.4.23.1), which preferentially cleaves at the ment, indicating that the active peptides are carboxyl terminus of phenylalanine and leu- resistant to further cleavage by this enzyme. In cine residues in the substrate (15). It is possi- contrast, the minimal inhibitory concentration ble that the other proteases tested cleave within of undigested lactoferrin against this strain was the region of the substrate that is essential for 2.0 mg/ml. This activity of undigested lactoferthe potent antibacterial effect, but additional rin against E. coli 0111 was completely abolstudies are required to confirm this hypothesis. ished in the presence of .1 mM FeSO4, which Screening of Hydrolysates for Antibacterial Activity
TABLE 1. Effect of enzymatic hydrolysates of bovine laaoferrin on
Enzyme (source)
Pepsin (pig) Pepsin (cod) PD enzyme (Penicillium worm] Trypsin big)
pap&
(Papaya) Actioase AS (Streptomyces griseus)
Rotease P (Aspergillus melleus) Rotease A (Aspergillus oryzae) Bioprase (Bacillus subtilis) Control, no lactofenin Control, undigested lactofenin
the viability of Escherichia coli 0111.
pH of enzymatic reaction
25 2.5 2.5 7.0 7.0 7.0
7.0 7.0 7.0
... ...
Viability' (CfJml) 3 mo without by disruption of essential membrane functions. any decrease in potency, and no loss of activity was observed after lyophilization of the hyHPLC Fractlonatlon drolysate (data not shown). The hydrolysate prepared by porcine pepsin Bactericldal Effect cleavage of lactofenin was fractionated by The effect of the hydrolysate was bacteri- reversephase HPLC, and antibacterial activity cidal as indicated by a rapid loss of viability of was detected in only one peak (Figure 3). The E. coli 0 1 1 1 (Figure 2). The active peptides active I k t i o n contained several peptides, apparently represent a bactericidal domain of however, as determined by SDS-PAGE(data lactoferrin released by pepsin cleavage. By not shown). It remains uncertain whether a l l of pointing to the existence of a such a domain, the activity can be attributed to a single pepour observations lend support to previous work tide. The partially purified peptides in this of Arnold et al. (2, 3, 4, 5) suggesting that fraction exhibited a minimal inhibitory concen-
Journal of Dairy Science Vol. 74, No. 12, 1991
4142
TOMITA ET AL. ACKNOWLEDGMENTS
We thank Y. Enomoto for expert technical assistance. REFERENCES
Time (min) Figure 3. Reverse-phase HPLC fractionation of pep tides generated by porcine pepsin hydrolysis of bovine lactofedn. Antibacterial activity was detected in a single peptide peak, as indicated by the BROW.
tration of .02 mdml against E. coli 0111, a level similar to the potency of many clinically useful antibiotics. If such potent antibacterial peptides are naturally produced in vivo, by gastric pepsin cleavage of lactofenin in milk and saliva, these peptides may affect the species composition of the gut microflora and have an important role in prevention of infectious disease. Further studies now in progress are aimed at purification and characterization of the active peptides. Elucidation of their primary structure is expected to provide valuable insight into the antimicrobial mechanism of lactoferrin. CONCLUSIONS
The lactoferrin hydrolysate described in the present study has potent broad-spectnun antibacterial propeaies, and considerable potential exists for its widespread commercial use. Industrial-scale processes are already established for the recovery of bovine lactofenin from cheese whey in a nondenatued functional form, and we think that the lactoferrin hydrolysate can be manufactured economically for use as a safe and effective natural preservative agent. Commercial interest in such materials is increasing as the result of a rising consumer demand for " ~ t u r a l "foods and cosmetics that contain no manmade preservatives. Furthermore, the lactoferrin hydrolysate should be useful as a functional component of new clinical foods for prevention or treatment of gastrointestinal infections, especially in neonates.
Journal of Dairy Science VoL 74. No. 12, 1991
IAbe, H.,H. Saito, H. Miyakxwa, Y. Tamura, S. Shimamura, E. Nagao, and M. Tomita 1991. Heat stability of bovine lactoferrin at acidic pH. J. Dairy Sci. 74:65. LArnold, R. R, U Brewer, and J. J. Gauthier. 1980. Bactericidal activity of human 1actofemir.c sensitivity a f a variety of microorganisms. Infect. Jmmun. 28: 893. 3 Arnold, R. R., M. P.Cole, and J. R. McGhee. 1977. A bactericidal role for lactoferrin. Science 197263. 4Amold, R. R, J. E. Russell, W. J. Champion, M. Brewer, and J. J. Gauthier. 1982. Bactericidal activity of human lactoferrin: differentiation from the stasis of iron deprivation. Infect. Immun. 35:792. 5 Arnold, R R, J. E. Russell, W. J. Champion, and J. J. Gauthier. 1981. Bactericidal activity of human lact o f d intluence of physical conditions and metabolic state of the target microorgauism.Infect. Immun. 32655. 6Brines, R. D., and 1. H.Brock. 1983. The effect of trypsin and chymotrypsin on the in vitro aniimicrobial and iron-binding properties of lactoferrin in human milk and bovine colostrum. Unusual resistance of human lactoferrin to proteolytic dqestioa Biochim. Biophys. Acta 759:229. 7 britt^^, J. R., and 0. KOldOVSky. 1989. Gastric I d nal digestion of lactofenin and transferrin by preterm infaats. Early H L ~Dev. . 19:127. 8 Brock, J. H.1980. Lactofenin in human m& its role in iron absorption and protection against enteric infection in the newborn infant. Arch. Dis. Child. 55:417. 9 Bnllen, J. J. 1981. The significance of iron in S e c tion Rev. Infect. Dis. 3:1127. lOBnllen, I. J., H.J. Rogers, and L.Leigh. 1972. Ironbinding proteins in milk and resistance to E. coli infection in infants. Br. Med. J. 1:69. llElliSon, R. T.. T.1. Gkhl, and P. M.LaForce. 1988. Damage of the outer membrane of enteric Gramnegative bacteria by lactofexrin and transferria Infect. h u n . 562774. 12Law, B. A., and B. Reiter. 1977. The isolation and bacteriostatic properties of lactofenin from bovine milk whey. I. Dairy Res. 44:595. 13 Swuki, T.,M.Nonaka, I. Kiyosawa, and K. Ogasa. 1978. Peptic digestion of human and bovine lactoferrin. J. J p Soc. Nutr. Food Sci. 31:395. 14Suzuki, T.,Y. Yamaachi, K. Kawase, M. Tomita, I. Kiand S.Okonogi. 1989. propertics of peptictryptic digestion products of human and bovine lactofurins. J. Jpn Soc. Nutr. Food Sci. 4213. 15 Webb, E.C. 1984. Enyme nomenclature 1984. Academic Ress, Inc., San Diego, CA.
mals, is a main antimicrobial component of colostrum and milk that contributes to protect the infant from infectious disease (8, 9, 10). The antimicrobial activity of lactofenin is known to be dependent on its iron-free state and is commonly attributed to its ability to bind and sequester iron, producing an irondeficient environment that limits microbial growth (8, 9). However, several studies have suggested that iron-free lactoferrin has bactericidal activity and kills sensitive bacteria by a mechanism distinct from sequestering of iron that involves its direct interaction with the bacterial cell surface (2, 3, 4, 5). Because few investigators have examined the antimicrobial properties of enzyme-hydrolyzed lactoferrin (6, 14), and because no active peptides have been reported previously, it is uncertain whether the entire lactoferrin molecule is required for this bactericidal effect. In the present study, to determine whether active peptides are produced from bovine lactoferrin, various commercially available proteases were employed to digest this protein, and the hydrolysates were assayed for antibacterial activity.
ABSTRACT
The antibacterial properties of enzymatic hydrolysates of bovine lactofenin were examined to determine whether active peptides are produced from this protein. Hydrolysates prepared by cleavage of lactoferrin with porcine pepsin, cod pepsin, or acid protease from Penicillium duponti showed strong activity against Escherichia coli 0 1 11, whereas hydrolysates produced by trypsin, papain, or other neutral proteases were much less active. Low moIecular weight peptides generated by porcine pepsin cleavage of lactoferrin showed broad-spectrum antibacterial activity, inhibiting the growth of a number of Gram-negative and Gram-positive species, including strains that were resistant to native lactoferrin. The antibacterial potency of the hydrolysate was at least eightfold greater than that of undigested lactoferrin with all strains tested. The active peptides retained their activity in the presence of added iron, unlike native lactofenin. The effect of the hydrolysate was bactericidal as indicated by a rapid loss of viability of E. coli 0111. The lactoferrin hydrolysate described in the present study has oommercial value as a natural preservative agent for use in foods and cosmetics, and as a functional component of new clinical foods for prevention or treatment of gastrointestinal disease. (Key words: lactoferrin, antibacterial peptides, pepsin digestion)
MATERIALS AND METHODS Preparation of Lactoterrin
Bovine lactofenin was prepared from fresh skim milk (Morinaga Milk Industry Co.,Zama City, Jpn.) by the method of Law and Reiter (12). The iron content of the lactofenin was 15% of saturation as determined by atomic absorption spectrometry.
INTRODUCTION
Lactofenin, an iron-binding glycoprotein present in most mucosal secretions of mam-
Received April 29. 1991. Accepted July 1, 1991. 1991 J Dairy Sci 74:4137-4142
Proteolytic Digestion of Lactoferrln
For hydrolysis of lactofenin, the following enzymes were used: porcine pepsin (EC 3.4.23.1; 10 units/mg; Difco Laboratories, Detroit, Ml), porcine trypsin (EC 3.4.21.4; loo0 units/mg; Sigma Chemical Co., St.
4137
4138
TOMITA
Louis, MO), PD enzyme (Penicillium duponri) (10 unitdmg; Seishin Pharmaceutical Co., Noda City, Jpn), Protease A (10 units/mg; Amano Pharmaceutical Co., Nagoya, Jpn), F'rotease P (10 units/mg; Amano Pharmaceutical Co.,Nagoya, Jpn), Actinase AS (250 units/ mg; K a k a Pharmaceutical Co.,Tokyo, Jpn), cod pepsin (EC 3.4.23.1; 10 units/mg; Nagase Biochemicals Co., Tokyo, Jpn), papain (EC 3.4.22.2; 50 unitdmg; Nagase Biochemicals Co., Tokyo, Jpn), and Bioprase (20 UnitS/mg; Nagase Biochemicals Co., Tokyo, Jpn). Bovine lactofenin was dissolved in distilled water at 5% (wthol), and the pH was adjusted to 2.5 or 7.0. Each protease was added at a final concentration of 3% (wt/wt of substrate). The hydrolysis reaction was performed at 37'C for 4 h, unless otherwise indicated, and terminated by heating at 80'C for 15 min. Reaction mixtures containing pepsin or PD enzyme were neutralized by addition of 1N NaOH. The precipitate of insoluble peptides formed (less than 10% of total protein) in each reaction mixture was removed by centrifugation at 15,000 x g, and the supernatant was retained and freezedried. The resulting lyophilized powder was used in the experiments. Cultures and Growth Condltlons
Escherichia coli 0111, a pathogenic intestinal bacterium obtained from the Institute of Medical Sciences at the University of Tokyo, Japan, was used routinely to investigate the antibacterial activity of lactoferrin and its derivatives. The organism was cultured at 37'C in a basal medium of 1% (wt/vol) Bactopeptone (Difco Laboratories, Detroit, MI), pH 6.8, or in peptone-yeast-glucose medium (1% Bactopeptone, 1% glucose, .05% yeast extract), pH 6.8. Stock solutions of lactoferrin or lactoferrin hydrolysate were prepared in distilled water and filter-sterilized before addition to these media. To examine the effect of added iron on antibacterial activity, the basal medium was supplemented with .1 mM FeSO4. Other bacterial strains were cultured in a similar manner. The extent of bacterial growth was determined by monitoring the absorbance of cultures at 660 nm (model U-3200 spectrophotometer, Hitachi, Jpn). Journal of Dairy Science Vol. 74, No. 12, 1991
ET AL. Assays of
Antlbacterlal Actlvlty
For determination of minimal inhibitory concentration, each strain was cultured for 16 to 20 h in a series of tubs containing 2.0 ml of peptone-yeast-glucose medium with various concentrations of lactoferrin (0 to 8.0 m d d ) or lactoferrin hydrolysate (0 to 1.6 mg/ml). A standard inoculum of logarithmic phase cells was added to each tube at a final concentration of 106 cfu/ml. The minimal inhibitory concentration was taken as the lowest concentration of lactoferrin or lactoferrin hydrolysate that caused complete inhibition of growth. For assay of bactericidal activity, suspensions of logarithmic phase E. coli 0111 (106 cfu/d) in basal medium with or without lactoferrin hydrolysate were incubated at 37'C in a shakerincubator water bath. After the indicated time, serial 10-fold dilutions were prepared in basal medium and plated onto plate count agar (Eiken Chemical Co.,Tokyo, Jpn) for determination of colony-forming units. Electrophoresls
Sodium dodecyl sulfate-PAGE was performed using a precast 10 to 25% linear gradient polyacrylamide gel and Laemmli buffer kit under conditions as recommended by the manufacturer (Daiichi Chemical Co.,Tokyo, Jpn). Molecular weight markers used were galactosidase (1 16,000), phosphorylase B (93,000), bovine s e ru m albumin (66,000), ovalbumin (45,000), soybean trypsin inhibitor (22,000), and lysozyme (14,000). HPLC Fractionation of Peptides
The antibacterial hydrolysate prepared by porcine pepsin cleavage of bovine lactoferrin was fractionated by reversephase HPLC using a column of TSK-GEL12Or (6.0 x 150 mm; Tosoh, Tokyo, Jpn) and a mixture of eluents A (.05% trifluoroacetic acid) and B (90% acetonitrile in .05% trifluoroacetic acid). Initially, a mix8020 of eluents A B was applied to the column for 10 min, followed by a linear gradient from 80:20 to 40:60 of eluent A. eluent B for 30 min at a flow rate of .8 ml/min. Fractions representing each of the peptide peaks were collected, dried under vacuum in a centrifugal evaporator, and assayed for antibacterial activity.
ANTIBACTERIAL HYDROLYSATE
RESULTS AND DISCUSSION
4139
OF LACTOFERRIN
Pepsln Hydrolysis of Lactoferrin
To examine the susceptibility of bovine lactoferrin to gastric pepsin cleavage, the protein was treated at pH 2.5 with porcine pepsin for 4 In a preliminary experiment, to determine h at 37'C, and samples were removed at whether antibacterial peptides are generated by 30-min intervals. Under the conditions emproteolytic cleavage of bovine lactofenin, hy- ployed, the reaction appeared to be complete drolysates prepared using various commer- within 30 min, as indicated by the SDS-PAGE cially available enzymes were tested for activ- profile of the reaction products, and by this ity against E. coli 0111 (Table 1). Undigested time most peptides in the hydrolysate had lactoferrin exhibited no antibacterial activity at molecular weights of less than 6000 (Figure 1). .5 mg/ml, however, at the same concentration, These results are consistent with previous hydrolysates produced at acid pH by porcine reports indicating that both bovine (13) and pepsin, cod pepsin, or the aspartic protease of human (7, 13) lactoferrin are readily cleaved P. duponti profoundly reduced the viability of by gastric pepsin under acidic conditions. the test strain. Hydrolysates produced at neutral pH by trypsin and other proteases, except Antibacterlal Properties papain, also caused some inhibition of growth, of the Lactoferrin Hydrolysate but the antibacterial effect was much weaker The low molecular weight peptides of bothan observed with the pepsin hydrolysate. vine lactoferrin obtained after 30 min of pepsin These observations suggest that potent antibac- treatment completely inhibited the growth of terial peptides of bovine l a c t o f e e are gener- E. coli 0111 at concentrations of .25 mdml or ated by enzymes having a cleavage site speci- more. The same minimal inhibitory concentraficity similar to porcine pepsin A (EC tion was observed after 4 h of pepsin treat3.4.23.1), which preferentially cleaves at the ment, indicating that the active peptides are carboxyl terminus of phenylalanine and leu- resistant to further cleavage by this enzyme. In cine residues in the substrate (15). It is possi- contrast, the minimal inhibitory concentration ble that the other proteases tested cleave within of undigested lactoferrin against this strain was the region of the substrate that is essential for 2.0 mg/ml. This activity of undigested lactoferthe potent antibacterial effect, but additional rin against E. coli 0111 was completely abolstudies are required to confirm this hypothesis. ished in the presence of .1 mM FeSO4, which Screening of Hydrolysates for Antibacterial Activity
TABLE 1. Effect of enzymatic hydrolysates of bovine laaoferrin on
Enzyme (source)
Pepsin (pig) Pepsin (cod) PD enzyme (Penicillium worm] Trypsin big)
pap&
(Papaya) Actioase AS (Streptomyces griseus)
Rotease P (Aspergillus melleus) Rotease A (Aspergillus oryzae) Bioprase (Bacillus subtilis) Control, no lactofenin Control, undigested lactofenin
the viability of Escherichia coli 0111.
pH of enzymatic reaction
25 2.5 2.5 7.0 7.0 7.0
7.0 7.0 7.0
... ...
Viability' (CfJml) 3 mo without by disruption of essential membrane functions. any decrease in potency, and no loss of activity was observed after lyophilization of the hyHPLC Fractlonatlon drolysate (data not shown). The hydrolysate prepared by porcine pepsin Bactericldal Effect cleavage of lactofenin was fractionated by The effect of the hydrolysate was bacteri- reversephase HPLC, and antibacterial activity cidal as indicated by a rapid loss of viability of was detected in only one peak (Figure 3). The E. coli 0 1 1 1 (Figure 2). The active peptides active I k t i o n contained several peptides, apparently represent a bactericidal domain of however, as determined by SDS-PAGE(data lactoferrin released by pepsin cleavage. By not shown). It remains uncertain whether a l l of pointing to the existence of a such a domain, the activity can be attributed to a single pepour observations lend support to previous work tide. The partially purified peptides in this of Arnold et al. (2, 3, 4, 5) suggesting that fraction exhibited a minimal inhibitory concen-
Journal of Dairy Science Vol. 74, No. 12, 1991
4142
TOMITA ET AL. ACKNOWLEDGMENTS
We thank Y. Enomoto for expert technical assistance. REFERENCES
Time (min) Figure 3. Reverse-phase HPLC fractionation of pep tides generated by porcine pepsin hydrolysis of bovine lactofedn. Antibacterial activity was detected in a single peptide peak, as indicated by the BROW.
tration of .02 mdml against E. coli 0111, a level similar to the potency of many clinically useful antibiotics. If such potent antibacterial peptides are naturally produced in vivo, by gastric pepsin cleavage of lactofenin in milk and saliva, these peptides may affect the species composition of the gut microflora and have an important role in prevention of infectious disease. Further studies now in progress are aimed at purification and characterization of the active peptides. Elucidation of their primary structure is expected to provide valuable insight into the antimicrobial mechanism of lactoferrin. CONCLUSIONS
The lactoferrin hydrolysate described in the present study has potent broad-spectnun antibacterial propeaies, and considerable potential exists for its widespread commercial use. Industrial-scale processes are already established for the recovery of bovine lactofenin from cheese whey in a nondenatued functional form, and we think that the lactoferrin hydrolysate can be manufactured economically for use as a safe and effective natural preservative agent. Commercial interest in such materials is increasing as the result of a rising consumer demand for " ~ t u r a l "foods and cosmetics that contain no manmade preservatives. Furthermore, the lactoferrin hydrolysate should be useful as a functional component of new clinical foods for prevention or treatment of gastrointestinal infections, especially in neonates.
Journal of Dairy Science VoL 74. No. 12, 1991
IAbe, H.,H. Saito, H. Miyakxwa, Y. Tamura, S. Shimamura, E. Nagao, and M. Tomita 1991. Heat stability of bovine lactoferrin at acidic pH. J. Dairy Sci. 74:65. LArnold, R. R, U Brewer, and J. J. Gauthier. 1980. Bactericidal activity of human 1actofemir.c sensitivity a f a variety of microorganisms. Infect. Jmmun. 28: 893. 3 Arnold, R. R., M. P.Cole, and J. R. McGhee. 1977. A bactericidal role for lactoferrin. Science 197263. 4Amold, R. R, J. E. Russell, W. J. Champion, M. Brewer, and J. J. Gauthier. 1982. Bactericidal activity of human lactoferrin: differentiation from the stasis of iron deprivation. Infect. Immun. 35:792. 5 Arnold, R R, J. E. Russell, W. J. Champion, and J. J. Gauthier. 1981. Bactericidal activity of human lact o f d intluence of physical conditions and metabolic state of the target microorgauism.Infect. Immun. 32655. 6Brines, R. D., and 1. H.Brock. 1983. The effect of trypsin and chymotrypsin on the in vitro aniimicrobial and iron-binding properties of lactoferrin in human milk and bovine colostrum. Unusual resistance of human lactoferrin to proteolytic dqestioa Biochim. Biophys. Acta 759:229. 7 britt^^, J. R., and 0. KOldOVSky. 1989. Gastric I d nal digestion of lactofenin and transferrin by preterm infaats. Early H L ~Dev. . 19:127. 8 Brock, J. H.1980. Lactofenin in human m& its role in iron absorption and protection against enteric infection in the newborn infant. Arch. Dis. Child. 55:417. 9 Bnllen, J. J. 1981. The significance of iron in S e c tion Rev. Infect. Dis. 3:1127. lOBnllen, I. J., H.J. Rogers, and L.Leigh. 1972. Ironbinding proteins in milk and resistance to E. coli infection in infants. Br. Med. J. 1:69. llElliSon, R. T.. T.1. Gkhl, and P. M.LaForce. 1988. Damage of the outer membrane of enteric Gramnegative bacteria by lactofexrin and transferria Infect. h u n . 562774. 12Law, B. A., and B. Reiter. 1977. The isolation and bacteriostatic properties of lactofenin from bovine milk whey. I. Dairy Res. 44:595. 13 Swuki, T.,M.Nonaka, I. Kiyosawa, and K. Ogasa. 1978. Peptic digestion of human and bovine lactoferrin. J. J p Soc. Nutr. Food Sci. 31:395. 14Suzuki, T.,Y. Yamaachi, K. Kawase, M. Tomita, I. Kiand S.Okonogi. 1989. propertics of peptictryptic digestion products of human and bovine lactofurins. J. Jpn Soc. Nutr. Food Sci. 4213. 15 Webb, E.C. 1984. Enyme nomenclature 1984. Academic Ress, Inc., San Diego, CA.
