Antibacterial activity

The antibacterial properties of milk have been observed for a long time. Most of the relevant literature consists of observations that various pathogenic and saprophytic bacteria are killed or their growth temporarily inhibited by cow's milk. Bacteriostasis was the widely characterized inhibitory mechanism of LF with well-documented data. Over the past three decades, various laboratories have identified LF as a broad-spectrum antimicrobial and reported a variety of inhibitory mechanisms on both Gram-positive and Gramnegative bacteria (Table 3).

1. Gram-positive bacteria. In 1967, Reiter and Oram reported the antibacterial effects of LF against Bacillus stearothermophilus and B. subtilis. This study also observed that apo-LF was unable to inactivate bacterial spores but could inhibit then germination. A decade later, Arnold and co-workers (1977) reported cidal activity of LF against Streptococcus mutans and other oral streptococci.

The occurrence of LF in saliva has initiated many studies on antimicrobial activity against oral streptococci and contr ol of caries. Apo-LF could cause a potent in vitro growth-inhibition of Streptococcus mutans and this effect could be reversed by iron (Visca et al., 1989). Furthermore, LF seems to reduce the adsorption of S.mutans cells to hydroxyapatite. This adherence-inhibiting effect of apo-LF together with bacteriostasis activity towards S. mutans suggests a possible patho-biological significance of caries control in the oral cavity in vivo. However, apo-bLF seems to elicit a low degree antimicrobial effect on mastitis-associated streptococci in bovine mammary secretions (Todhunter et al., 1985).

Naidu and co-workers (1990; 1991) have identified specific LF-binding proteins in Staphylococcus aureus isolated from human and animal infections as well as among various species of coagulase-negative staphylococci causing bovine mastitis. Apo-bLF at concentrations of 0.1%-0.4% could convert compact colonies of Staphylococcus haemolyticus transient to diffused in soft agar (Godo et al., 1997). This surface-active property of LF has prevented autoaggregation of cocci in compact ball-like colonies by hydrophobic interaction or trypsin-sensitive proteins. In vivo anti-staphylococcal activity of hLF, bLF and bLF hydrolysate was reported in an experimental mouse model (Bhimani et al., 1999). All the LF preparations demonstrated weak in vitro antibacterial activity while holo-LFs showed no activity. LF-treated mice (1 mg, i.v.) when injected with 106 staphylococci, showed 30-50% reduction in kidney infections, and viable bacterial counts in the kidney decreased 5- to 12-fold. The inhibitory effect was dose-dependent up to 1 mg LF. The LF preparations were effective when given 1 day prior to the bacterial challenge, after which there was no significant effect even at doses up to 5 mg. Apo- and Fe-saturated forms of hLF and bLF were all equally effective, while bLF hydrolysate was not protective. Different degrees of iron-saturation did not alter the in vivo antimicrobial property of either native LF preparation. Feeding mice with 2% bLF in drinking water also reduced the kidney infections by 40-60%, and viable bacterial counts, 5-12-fold.

Human LF was shown to be bactericidal in vitro for Micrococcus luteus but not for other Micrococcus species (M. radiophilus, M. roseus and M. varians) (de Lillo et al., 1997). A correlation between the binding of LF to the bacterial surface and the antimicrobial action was observed. Viability assays showed that ferric, but not ferrous, salts prevented binding and consequently M. luteus was not killed. The unsaturated form of LF showed a greater affinity than that of the iron-saturated molecule for lipomannan, a lipo-glycan present on the cell wall of M. luteus, supporting the role for lipomannan as one of the possible binding sites of LF on M. luteus.

Custer and Hansen (1983) found that LF fragments could react with nitrite and cause inhibition of Bacillus cereus spore outgrowth. LF and lysozyme were shown to inhibit the growth of Bacillus stearothermophilus var. calidolactis spores (Carlsson et al., 1989). The growth of Bacillus cereus could be inhibited by LF and this effect could be reversed by the addition of iron (Sato et al., 1999). The growth inhibition was also reversed by the addition of erythrocytes and hemoglobin. B. cereus seems to use heme or heme-protein complex (hemoglobin-haptoglobin and hematin-albumin complexes) as iron sources in iron deficient conditions.

Oral administration of bLF with milk has been reported to inhibit various species of Clostridia including C. ramosum, C. paraputrificum and C. perfringens in an experimental mouse model (Teraguchi et al., 1995).

Table 3. Inhibitory spectrum of hLF, bLF and LFcins against various bacteria.

Bacterial species

Form

Dose

Effect

Reference

Actino. actinomycetemcomitans

hLF

2 fiM

Cidal (2-log reduction)

Kalmar & Arnold, 1988

Aeromonas hydrophila

bLF

0.1%

Adhesion-blockade (47%)

Paulsson et al., 1993

Bacillus cereus

bLF ein

6 nM

Cidal (4-log, 100%)

Hoek et al., 1997

Bacillus circulans

bLFcin

0.006%

Cidal (6-log, 100%)

Bellamy et al., 1992

Bacillus natto IF03009

bLFcin

0.002%

Cidal (6-log, 100%)

Bellamy et al., 1992

Bacillus stearothermophilus

bLF

1:20

Stasis

Reiter & Oram, 1967

Bacillus subtil is

bLF

1:20

Stasis

Reiter & Oram, 1967

Bacillus subtilis ATCC6633

bLFcin

0.002%

Cidal (6-log, 100%)

Bellamy et al., 1992

Bifidobacterium longum

bLF

0.1%

Agglutination

Tomita et al., 1994

Corynebacterium diphtheriae

bLFcin

0.018%

Cidal (6-log, 100%)

Bellamy et al., 1992

Coryne. ammoniagenes

bLFcin

0.003%

Cidal (6-log, 100%)

Bellamy et al., 1992

Coryne. renale

bLFcin

0.001%

Cidal (6-log, 100%)

Bellamy et al., 1992

Clostridium innocuum

bLF

0.1%

Agglutination

Tomita et al., 1994

Clostridium perfringens

bLFcin

0.024%

Cidal (6-log, 100%)

Bellamy et al., 1992

Clostridium paraputrificum

bLFcin

0.003%

Cidal (6-log, 100%)

Bellamy et al., 1992

Enterococcus faecalis

bLFcin

0.06%

Cidal (6-log, 100%)

Bellamy et al., 1992

Escherichia coli E386

bLF

0.1%

Stasis (24-h, 100%)

Naidu et al., 1993

Escherichia coli

hLF

42 nM

Cidal (6-log reduction)

Arnold etal., 1980

Escherichia coli H10407

bLF

0.1%

Adhesion-blockade (50%)

Paulsson et al., 1993

Escherichia coli IID-861

bLFcin

10 nM

Cidal (3-log reduction)

Bellamy et al., 1992

Escherichia coli HB101

hLF

0.2%

Invasion-inhibition

Longhi et al., 1993

Escherichia coli CL99

bLF

20 nM

LPS release, OM damage

Yamauchi et al., 1993

Klebsiella pneumoniae

bLFcin

10 |jM

Cidal (3-log reduction)

Bellamy et al., 1992

Lactobacillus casei

bLFcin

0.012%

Cidal (6-log, 100%)

Bellamy et al., 1992

Legionella pneumophila

hLF

0.03%

Cidal (4-Iog reduction)

Bortner et al., 1986

Listeria monocytogenes

bLFcin

10 |iM

Cidal (4-log reduction)

Bellamy et al., 1992

L. monocytogenes NCTC7973

bLFcin

2 fiM

Cidal (4-log, 100%)

Hoek et al., 1997

Micrococcus luteus

bLF

0.1%

Agglutination

Tomita et al., 1994

Proteus vulgaris JCM1668T

bLFcin

0.012%

Cidal (6-log, 100%)

Bellamy et al., 1992

Pseudomonas aeruginosa

hLF

42 |iM

Cidal (7-log, 100%)

Arnold et al., 1980

Ps. aeruginosa IF03446

bLFcin

10 jiM

Cidal (3-log reduction)

Bellamy et al., 1992

Pseudomonas fluorescens

bLFcin

8 |iM

Cidal (4-log, 100%,)

Hoek et al., 1997

Salmonella abony

bLF

0.8%

Stasis (24-h, 100%,)

Naidu & Arnold, 1994

Salmonella dublin

bLF

0.2%

Stasis (24-h, 100%,)

Naidu & Arnold, 1994

Salmonella enteritidis

bLFcin

0.012%

Cidal (6-log, 100%)

Bellamy et al., 1992

Salmonella hartford

bLF

0.8%

Stasis (24-h, 100%)

Naidu & Arnold, 1994

Salmonella kentucky

bLF

0.2%

Stasis (24-h, 100%)

Naidu & Arnold, 1994

Salmonella panama

bLF

0.1%

Stasis (24-h, 100%)

Naidu & Arnold, 1994

Salmonella pullorum

bLF

0.2%

Stasis (24-h, 100%)

Naidu & Arnold, 1994

Salmonella rostock

bLF

0.2%

Stasis (24-h, 100%)

Naidu & Arnold, 1994

Salmonella salford

bLFcin

4 jliM

Cidal (4-log, 100%)

Hoek et al., 1997

Salmonella montevideo

bLF

20 fiM

LPS release, OM damage

Yamauchi et al., 1993

Salmonella thompson

bLF

0.1%

Stasis (24-h, 100%)

Naidu & Arnold, 1994

Salmonella typhimurium Rd

bLF

0.5%

Stasis (64%)

Naidu et al., 1993

Salm. typhimurium RIO

bLF

0.1%

Adhesion-blockade (68%)

Paulsson et al., 1993

Sahn, typhimurium SL696

bLF

20 (iM

LPS release, OM damage

Yamauchi et al., 1993

Salmonella virchow

bLF

0.8%

Stasis (24-h, 100%)

Naidu & Arnold, 1994

Shigella flexneri

bLF

0.1%

Adhesion-blockade (30%)

Paulsson et al., 1993

Staphylococcus albus

bLF

0.5%

Stasis

Masson et al., 1966

Staphylococcus aureus

bLF

0.1%

Adhesion-blockade (54%)

Paulsson et al., 1993

Staph, aureus JCM215I

bLFcin

10 (jM

Cidal (3-log reduction)

Bellamy et al., 1992

Staphylococcus epidermidis

bLFcin

0.006%

Cidal (6-log, 100%)

Bellamy et al., 1992

Staphylococcus haemolyticus

bLFcin

0.001%

Cidal (6-log, 100%)

Bellamy et al., 1992

Staphylococcus hominis

bLFcin

0.003%

Cidal (6-log, 100%)

Bellamy et al., 1992

Bacterial species

Form Dose Effect

Reference

Streptococcus bovis Streptococcus cremoris Streptococcus lactis Streptococcus mitior Streptococcus mutans AHT Strep, mutans LM-7 Strep, mutans ATCC25175 Streptococcus pneumoniae Streptococcus salivarius Streptococcus thermophilics Vibrio cholerae 569B

hLF 0.01% Agglutination hLF 42 pM Cidal (7-log, 100%)

Bellamy et al, 1992 Bellamy et al., 1992 Bellamy et al., 1992 Arnold et al, 1980 Arnold et al, 1977 Arnold et al, 1980 Soukka et al, 1993 Arnold et al, 1980 Arnold et al, 1980 Bellamy et al, 1992 Arnold et al, 1977

Groenink and co-workers (1999) reported a potent antimicrobial activity of synthetic cationic peptides derived from the N-terminal domain that comprises an amphi-pathic a-helix in hLF (hLF 18-31 and hLF 20-38) and bLF (bLF 17-30 and bLF 19-37). Peptide bLF 17-30, containing the largest number of positively charged amino acids, elicited the highest inhibitory spectrum against both Gram-positive and Gram-negative bacteria.

2. Gram-negative bacteria. Many studies have shown the antimicrobial activity of LF against Gram-negative bacteria, E. coli, in particular. Various antimicrobial effects of LF were demonstrated against E.coli and different mechanisms were postulated to elucidate these effects. LF elicits a bacteriostatic effect on E. coli. Based on the time required for E.coli 0111 to reach one-half maximal cell density, Stuart and co-workers (1984) indicated that the in vitro effects of LF on the growth of E. coli was kinetic rather than bacteriostatic. Compared to a control, added apo-LF (0.25-1.0 mg/ml) produced only a delay effect indicating that these concentrations are probably within the sub-inhibitory concentration range. The kinetic delay effect of apo-LF also increased steadily in the presence of Zn2+ and Cu2+ cations. Cu2+, Zn2+ and nitrilotriacetate did not affect the growth rate of this organism in the absence of LF compared to the control. These studies indicate that the mechanism by which LF alters the bacterial growth of E. coli Olll is more complex than simple iron deprivation.

Rainard (1987) reported the antibacterial activity of milk against a virulent strain of E. coli using milk fractions from normal or inflamed glands. Whey obtained from mastitis milk exhibited either bactericidal or bacteriostatic activities, depending on whether bacteria were enumerated by the pour plate technique or by surface plating onto sheep blood agar. The cidal activity, however, was not due to LF, even when assayed in distilled water. Milk whey ultra-filtrate was used to assay the ability of normal and mastitis milk to support the antibacterial activities of LF against E. coli. The addition of purified LF to ultra-filtrate from mastitis whey resulted in bacteriostasis, whereas LF was without effect in ultra-filtrate from normal whey. It was suggested that LF could inhibit the growth of LF-sensitive bacteria during mastitis depending on plasma exudation during mastitis. Dionysius et al. (1993) reported that the antibacterial activity of apo-LF depends on bacterial inoculum size and the addition of holo-LF or cytochrome-C could diminish the effect. Furthermore, the resistance to inhibition by LF was not related to the production of bacterial siderophores in E.coli.

Ellison et al. (1988) hypothesized that the iron-binding proteins could affect the Gram-negative outer membrane in a manner similar to that of the chelator EDTA. Further, both the whole protein and a cationic N-terminus peptide fragment directly damage the outer membrane of Gram-negative bacteria suggesting a mechanism for the supplemental effects. Several groups have also shown that LF could synergistically interact with immunoglobins, complement, and neutrophil cationic proteins against Gram-negative bacteria.

Klebanoff and Waltersdorf (1990) found that Fe2+ and apo-LF could generate hydroxy radicals via an H202 intermediate with toxicity to E.coli, and hypothesized that such a mechanism could possibly contribute to the microbicidal activity of phagocytes.

LF binds to surface structures expressed in E. coli K-12 strains grown under iron limitation (Visca et al., 1990). Both apo and holo forms of LF yielded a maximum of 1.6 X 105 bound molecules/is. coli K-12 cell. The amount of LF bound was independent of the expression of iron-regulated outer membrane proteins. However, LF did not bind to E. coli clinical isolates. Apo-LF (500 iig/ml ) in a chemically defined medium inhibited the growth of E. coli K-12 strains but not of clinical isolates. These findings suggested that the antibacterial activity of the protein could be associated to its binding to the cell surface. Enterotoxigenic strains demonstrate higher LF interaction than enteropathogenic, enteroinvasive, enterohemorrhagic strains or normal intestinal E. coli isolates (Naidu et al., 1991). Also the enteropathogenic strains belonging to serotypes 044 and 0127 demonstrate higher LF binding compared to 026, 055, Ol 11, 0119 and 0126 serotypes. No significant differences in the degree of hLF or bLF binding were noticed between aer-obactin-producing and non-producing strains. In later studies, Naidu and co-workers have identified and characterized porins in the outer membrane of Gram-negative bacteria as the specific receptors for LF interaction (Tigyi et al., 1992; Naidu et al., 1993; Erdei et al., 1994)

Using an in vitro model, Gutteberg and co-workers (1990) reported the early effect of E. coli, Streptococcus agalactiae (group B streptococci, GBS) and recombinant tumor necrosis factor alpha (TNF) on the release of LF and the generation of interleukin-1 (IL-1) due to E. coli, using heparinized whole blood from healthy full-term newborns. In a final concentration of 107 per ml both bacteria increased the release of LF markedly. The response to E. coli was immediate. GBS was a less potent stimulant than E. coli and the response was only apparent after 20 minutes. TNF in a concentration of 10 ng/ml as well as 1 ng/ml increased the release of LF significantly, whereas a concentration of 0.1 ng/ml had no effect. Whole blood incubated with different preparations of LF for 20 minutes did not increase the LF levels. No significant changes in IL-1 levels were observed. LF had bacteriostatic but no bactericidal effect on GBS and Streptococcus mutans.

Payne et al. (1990) demonstrated that apo-bLF had bacteriostatic activity against four strains of L. monocytogenes and an E.coli at concentrations of 15 to 30 mg/ml, in UHT milk. At 2.5 mg/ml the compound has no activity against S. typhimurium, P. fluo-rescens and limited activity against E.coli 0157:H7 or L. monocytogenes VPHI (Payne et al., 1994).

Human LF and free secretory component (fSC) were shown to inhibit the hemagglutination induced by E. coli CFA1+ (Giugliano et al., 1995). The lowest concentrations of purified fSC and hLF to inhibit the hemagglutination by E. coli strain TR50/3 CFA1+ were 0.06 mg/ml and 0.1 mg/ml, respectively.

Table 4: Lactoferrin - Antiviral effects

Viral pathogen Antiviral effect Reference

Spleen focus forming virus (SFFV) Decreases multiplication Hangoc et al., 1987

Human influenza virus Inhibits viral hemagglutination Kawasaki et al., 1993

Human cytomegalovirus (HCMV) Inhibits infection & replication Hasegawa et al., 1994

Inhibits MT4 cell cytopathy Harmsen et al., 1995

Human herpes simplex virus (HSV-1) Inhibits adsorption & penetration Hasegawa et al., 1994

Prevents plaque formation Fujihara & Hayashi, 1995

Human immunodeficiency virus (HIV) Inhibits MT4 cytopathic effect Harmsen et al., 1995

Inhibits vero cell cytopathy Marchetti et al., 1996

Feline immunodeficiency virus (FIV) Effects clinical outcome Sato et al., 1996

Respiratory syncytial virus Inhibits viral multiplication Grover et al., 1997

Hepatitis 0 virus (HCV) Binds El and E2 envelope proteins Yi et al.. 1997

Rotavirus Inhibits HT-29 cell infection Superti et al., 1997

The antimicrobial activities of bLF, and bLFcin against four clinical isolates of enterohemorrhagic E. coli 0157:H7 were reported (Shin et al., 1998). The MICs against these isolates were 3 mg/'ml for bLF, and 8-10 jig/ml for bLFcin in 1% Bacto-peptone broth. Transmission electron microscopy findings suggested that bLFcin acts on the bacterial surface and affects cytoplasmic contents. Furthermore, bLFcin affected the levels of verotoxins in the culture supernatant fluid of an E. coli 0157:H7 strain.

The antimicrobial effect of LF against Salmonella typhimurium was tested by measuring conductance changes in the cultivation media by using a Malthus-AT system (Naidu et al., 1993). Conductance measurement studies revealed that the low-LF-binding strain 395MS with smooth LPS was relatively insusceptible to LF, while the high-LF-binding mutant Rd with rough LPS was more susceptible to LF suggesting an LPS shielding of antimicrobial effect. Later studies have led to the identification of porins as LF-binding outer membrane proteins in various species of Salmonellae (Naidu & Arnold, 1994).

Antimicrobial effects of LF against various Gram-negative bacterial pathogens, including Aeromonas hydrophila, Yersinia enterocolitica, Campylobacter jejuni, Helicobacteri pylori, Pseudomonas aeruginosa, Vibrio sp., have also been reported (Arnold et al., 1977; Paulsson et al., 1993; Tomita et al., 1994).

Was this article helpful?

0 0

Post a comment