Previous Article | Next Article ![]()
Antimicrobial Agents and Chemotherapy, July 2005, p. 2583-2588, Vol. 49, No. 7
0066-4804/05/$08.00+0 doi:10.1128/AAC.49.7.2583-2588.2005
Copyright © 2005, American Society for Microbiology. All Rights Reserved.
María T. Andrés,2,
and
José F. Fierro1,2*
Department of Functional Biology (Microbiology), Faculty of Medicine,1 Laboratory of Oral Microbiology, School of Stomatology, University of Oviedo, 33006 Oviedo, Spain2
Received 15 December 2004/ Returned for modification 18 January 2005/ Accepted 4 April 2005
|
|
|---|
|
|
|---|
Human lactoferrin (hLf) is a major protein of mucosal innate host defenses (11) that includes different antimicrobial amino acid sequences, such as human lactoferricin (lactoferricin H) and kaliocin-1 (4, 33). The first reported bactericidal hLf-derived peptide was lactoferricin H (residues 1 to 47), obtained by enzymatic hydrolysis of lactoferrin, which exhibits antimicrobial activity greater than that of the intact protein (3, 4). Furthermore, a variety of shorter antimicrobial synthetic peptides (lactoferricins) derived from the original sequence of lactoferricin H have been described (3, 11, 20, 24, 25). For example, Lfpep (residues 18 to 40 of hLf) is a lactoferricin H derived-peptide that possess giardicidal and bactericidal effects and the ability to permeabilize bacterial membranes, thus leading to cell death (1, 31). We recently reported on a novel antimicrobial hLf-derived peptide (residues 153 to 183 of hLf), termed kaliocin-1. This peptide differs from lactoferricin H (Table 1) and exhibits a bactericidal effect at low salt concentrations without causing membrane permeabilization (33).
|
View this table: [in a new window] |
TABLE 1. Amino acid sequences of human lactoferricin, Lfpep, and kaliocin-1
|
|
|
|---|
Peptides. The human lactoferrin-derived peptides kaliocin-1 and Lfpep were synthesized according to standard 9-fluorenylmethoxy carbonyl chemistry by Bethyl Laboratories (Montgomery, TX), and their purities (>95%) were checked by reverse-phase high-pressure liquid chromatography. The peptides were allowed to refold by air oxidation for 18 h at room temperature, as reported previously (33). The compositions and concentrations of the purified peptides were verified by mass spectrometry and amino acid analysis. Table 1 shows the amino acid sequences of both peptides. Magainin 2 was purchased from Bachem Bioscience (King of Prussia, PA).
Yeast strains and growth conditions. The strains used for determination of candidacidal activity included Candida albicans ATCC 10231 (American Type Culture Collection, Manassas, VA) and clinical isolates obtained from M. Alvarez (Hospital Universitario Central de Asturias, Oviedo, Spain). The cells were grown to stationary phase in SDB and subcultured (1:400) in the same medium to the mid-logarithmic growth phase at 37°C. After the blastoconidia were centrifuged and washed twice with the same buffer or medium used to perform the assay, their concentrations were adjusted as appropriate.
Candidacidal assays. The anti-Candida effects of the peptides were monitored by using cell suspensions (105 cells/ml) of C. albicans ATCC 10231 in 10 mM Tris-HCl (pH 7.4) incubated with the peptides at 37°C. At the indicated times after peptide addition, serial dilutions prepared in the same buffer were spread onto SDA plates. Colonies were counted after 24 h of incubation at 37°C, and the results were expressed as the percentages of the colony counts of bacteria not exposed to the peptides.
To determine the optimal pH value, candidacidal assays were performed in Tris buffer at different pHs (range, 5.5 to 7.5) by incubation of the cells with Lfpep (50 µM) or kaliocin-1 (150 µM) at 37°C for 1 h. Similar assays were performed in Tris buffer (pH 7.4) to determine the differences in the activity at 4°C and 37°C.
Broth microdilution assays. Cell suspensions (105 cells/ml) of Candida spp. were suspended in PG broth (0.3% Bacto peptone, 2% glucose) in 96-well plates. The growth-inhibitory concentration was determined by using twofold serial dilutions of Lfpep (range, 0.6 to 75 µM), kaliocin-1 (range, 0.6 to 300 µM), rhLf (range, 0.125 to 25 µM), and magainin 2 (range, 0.6 to 150 µM). The plates were incubated at 37°C for 24 h, and the growth-inhibitory concentration was the one at which there was no visible growth. Resistance to fluconazole (>50 µg/ml) and amphotericin B (>2 µg/ml) was determined in RPMI medium following CLSI (formerly NCCLS) guidelines (23).
Permeability of cytoplasmic membrane. The permeabilization of the membrane of C. albicans cells was investigated by flow cytometric analysis by using PI, as described previously (20). Cell suspensions (105 cells/ml) in Tris-HCl (pH 7.4) were incubated at 37°C with the peptide, and samples taken at different times were incubated with 9 µM PI for 5 min. Cell fluorescence was recorded with a flow cytometer (Cytoron Absolute; Ortho Diagnostics Systems Inc., Raritan, NJ).
Measurement of ions. Determinations of K+ release were performed as described previously (2, 34). Briefly, cell suspensions were centrifuged (2,000 x g, 2 min) and then washed twice with ice-cold MgCl2 (20 mM) containing an isosmotic concentration of sorbitol. The cells (109 cells/ml) were resuspended in 5 mM sodium phosphate buffer (pH 7.4), and the peptides or nystatin (100 µg/ml) was added. The mixture was incubated at 37°C, samples (0.5 ml) were taken at different times (15, 30, 45, and 60 min) and centrifuged, and the supernatant was collected to determine the extracellular K+ content by flame photometry. The total K+ content of the cellular suspension was estimated after K+ was released from the cells by treatment with 0.5% (vol/vol) perchloric acid and heat (95°C for 1 h) and centrifugation to remove the cell debris. The supernatant was then analyzed for the intracellular K+ content. The percentages of K+ were calculated on the basis of the total cellular K+ content of untreated cells.
Measurement of electrical potential in C. albicans cells. Membrane potential was measured with the membrane potential-sensitive fluorescent probe DiOC5(3), as described previously (17). Cells (105 cells/ml) in Tris-HCl (pH 7.4) were incubated with the peptide at 37°C. The samples were reincubated for an additional 10 min with 0.5 µM DiOC5(3) and were immediately analyzed by cytofluorometry.
Sequence analysis. Comparison of amino acid sequences was performed by using the ClustalW software (30). The similarity of the amino acid sequences was calculated by use of the program EMBOSS-Align at the homepage of the European Bioinformatics Institute (http://www.ebi.ac.uk/emboss/align).
Statistical analysis. The data are presented as the means ± standard errors of the means from at least three sets of independent experiments. The Mann-Whitney U test was performed by using GraphPad InStat v. 3.06 (GraphPad Software, San Diego, CA). A P value <0.05 was considered statistically significant.
|
|
|---|
60 mM NaCl, and maximum inhibition (100%) was observed with about 100 mM NaCl. The inhibitory effect of NaCl on the killing activity was not reversed by increased concentrations of kaliocin-1 (up to 300 µM) or Lfpep (up to 150 µM). In addition, the candidacidal effect of neither peptide was significantly modified in the presence of 0.32 M sorbitol (data not shown).
![]() View larger version (15K): [in a new window] |
FIG. 1. Candidacidal activities of kaliocin-1 and Lfpep. (A and B) Dose- and time-dependent reductions in the number of viable C. albicans cells (105 cells/ml) incubated with kaliocin-1 (A) or Lfpep (B) in 10 mM Tris-HCl (pH 7.4) at 37°C. (C) Effect of NaCl on the candidacidal activities of 150 µM kaliocin-1 and 50 µM Lfpep tested for 1 h in 10 mM Tris-HCl (pH 7.4) containing different concentrations of NaCl. The percentage of viable cells was determined relative to that for cells incubated without peptides. The results are the means ± standard deviations of at least three separate experiments.
|
Comparative growth inhibition. Concentrations over a range of 0.6 to 150 µM of kaliocin-1 were inactive against Candida spp. when standard MIC determinations were performed in RPMI 1640 or PG broth (containing 1% peptone) (data not shown). A similar inhibitory effect, caused by medium components, was previously observed when other anticandidal peptides were tested and was solved by using nonstandardized synthetic media (26, 27). However, these alternative media were not suitable for testing the susceptibility of the Candida strains to kaliocin-1 due to the presence of an inhibitory salt concentration. Consequently, the assays were performed in PG broth prepared with 0.3% peptone that contained low Na+ and K+ concentrations, as determined by flame photometry (data not shown). In this medium, the growth of all Candida strains was supported and the killing activity of the peptide was retained, thus allowing the comparative assessment of the in vitro antifungal efficacies of the peptides. Lfpep and kaliocin-1 were active against all Candida strains, including fluconazole- and amphotericin B-resistant clinical isolates (Table 2). Lfpep was more active than kaliocin-1 against C. albicans (8-fold) and Candida spp. (8- to 30-fold). The corresponding values for rhLf and the antimicrobial peptide magainin 2 are also reported for the sake of comparison.
|
View this table: [in a new window] |
TABLE 2. Susceptibilities of Candida spp. to lactoferrin-derived peptides, lactoferrin, and magainin 2
|
Cytoplasmic membrane permeabilization. To determine the effects of kaliocin-1 and Lfpep on cell membrane integrity, C. albicans cells that had been treated or not treated with either of the peptides for different times (15, 30, 60, 90, and 120 min) were exposed to the fluorescent probe PI and analyzed by flow cytometry. PI fluorescently stains the nucleic acids of cells when cells suffer a loss of membrane integrity. Figure 2A shows typical results for C. albicans cells exposed to a candidacidal concentration of kaliocin-1 (150 µM) for 30 min, indicating the inability of this peptide to permeabilize the cytoplasmic membrane of C. albicans cells. Similar results were obtained after cell incubation with kaliocin-1 for 60, 90, and 120 min. In contrast, cells incubated with Lfpep (50 µM) showed a rapid increase in fluorescence due to intracellular PI accumulation, indicative of membrane permeabilization (Fig. 2A). A similar permeabilization was observed in control assays with the membrane-perturbing antifungal drug amphotericin B.
![]() View larger version (18K): [in a new window] |
FIG. 2. Effects of kaliocin-1 and Lfpep on the cytoplasmic membrane of C. albicans cells. The fluorescence profiles of C. albicans cells (105 cells/ml) treated for 30 min with 50 µM Lfpep, 150 µM kaliocin-1 (Kal-1), or 16 µg/ml amphotericin B (AmB) or untreated cells (C; control) stained with propidium iodide to determine membrane permeabilization (A) or DiOC5(3) to measure the membrane potential (B) are shown. The fluorescence distribution was analyzed by flow cytometry.
|
Effect of rhLf-derived peptides on the C. albicans membrane potential. Peptide-induced changes in membrane potential were measured with the potentiometric dye DiOC5(3), a cationic membrane-permeant probe that accumulates inside the cell on polarized membranes. The fluorescence of the probe thus decreases upon membrane depolarization. Addition of candidacidal concentrations of kaliocin-1 (150 µM) or Lfpep (50 µM) to a C. albicans suspension caused an immediate decrease in fluorescence intensity, indicative of membrane depolarization. A similar decrease of the membrane potential was observed at 15, 30, 60, 90, and 120 min. Figure 2B shows representative results obtained with cells exposed to the indicated candidacidal concentrations of both peptides and to amphotericin B (positive control) for 30 min.
|
|
|---|
Kaliocin-1 was inactive (at 300 µM) under reducing conditions, suggesting that the disulfide bridges have an important effect on the candidacidal activity of this peptide. However, the activity of Lfpep under similar reducing conditions was not changed, in agreement with previous reports showing that the absence of intramolecular disulfide bonds is not critical for the antimicrobial activities of lactoferricins (3). Work is being conducted with linearized synthetic peptides to determine the role of disulfide bonds on the antimicrobial activity of kaliocin-1.
We have previously shown that Lfpep and kaliocin-1 are able to permeabilize liposomes (1, 33) and planar membranes (unpublished data). Although this fact suggests that cell death might be a consequence of the disturbance of membrane functions, this mode of action is probably true only for Lfpep, due to the loss of cell viability correlated with the rapid intracellular accumulation of propidium iodide, high K+ release (
90%), and collapse of the membrane potential. These effects are similar to those previously reported for other lactoferricin peptides with permeabilizing activities (11). In contrast, the killing mediated by kaliocin-1 was not accompanied by membrane permeabilization for propidium iodide, and a lower percentage (
20%) of the intracellular K+ released was observed. Moreover, the dissociation between the loss of viability and membrane depolarization suggests that these effects are unrelated. Although we cannot draw a definitive conclusion with regard to the mechanism underlying the killing effect of kaliocin-1, the ability of this peptide to permeabilize artificial membranes (34), although not cell membranes, differs from that of pore-forming peptides and hypothetically suggests a different mechanism of action for kaliocin-1, such as the putative interaction of kaliocin-1 with a structural element of the membrane (e.g., protein), which induces K+ efflux and membrane depolarization, as was reported for the candidacidal peptide histatin 5 (17), among other possible mechanisms.
Interestingly, kaliocin-1 contains the recently reported amino acid sequence of the "multidimensional antimicrobial signature," including the characteristic CXG motif (Table 3), which corresponds to a structure present only in cysteine-containing peptides that exhibit antimicrobial activity (36). Since this model may be used to predict the unforeseen antimicrobial activities of peptides from phylogenetically distant organisms, the observed antibacterial (33) and antifungal activities of kaliocin-1 support that suggestion.
|
View this table: [in a new window] |
TABLE 3. Alignments of the amino acid sequences of kaliocin-1 and brevinin-1Sa
|
37%) and similarity (
47%) in their amino acid sequences (Table 3). Furthermore, kaliocin-1 contains a "Rana box"-like sequence that is also located in the C terminus of the peptide, including the characteristic Thr residue at the fourth amino acid position, and that is limited by two pairs of amino acids identical to the limiting residues present in the "Rana box" of brevinin-1Sa (Table 3). Although the antimicrobial mechanism of action of brevinins is not yet understood, these natural peptides and kaliocin-1 exhibit certain common activities, such as a broad antimicrobial spectrum (bactericidal and fungicidal effects) related to a non-pore-forming mechanism, K+ release, and the absence of hemolytic activity (12, 33), suggesting a functional resemblance between both types of peptides. These structural and functional similarities could be a starting point for the creation of hybrid peptides with enhanced antifungal effects. In this sense, the possibility of modifying the candidacidal activities of brevinin-1 peptides by substitution of a single amino acid has recently been reported (8). We have previously reported that sequences homologous to the kaliocin-1 sequence are present at the molecular surface of other transferrin family proteins (33). These sequences also contain the "multidimensional antimicrobial signature" (36) and even the motif CXG and include a primary structure pattern similar to the "Rana box." Moreover, kaliocin-1 partially retains the cidal effect of lactoferrin only in a low ionic environment, as described for transferrins (33). Overall, these observations are compatible with our previous hypothesis in which we suggested that sequences homologous to the kaliocin-1 sequence present in transferrins may be involved in the well-known antimicrobial activities of these evolutionarily ancient proteins (33).
In conclusion, two novel human lactoferrin-derived peptides with different anti-Candida activities are described. These results suggest an attractive approach to the search for cell targets for antifungals and novel antimicrobial synthetic peptides containing conserved sequences throughout evolution.
M.V.-D. and M.T.A. contributed equally to this report. ![]()
|
|
|---|
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»