Altering the Composition of Caseicins A and B as a Means of Determining the Contribution of Specific Residues to Antimicrobial Activity

Abstract

Caseicin A (IKHQGLPQE) and caseicin B (VLNENLLR) are antimicrobial peptides generated through the bacterial fermentation of sodium caseinate, and on the basis of this and previous studies, they are active against many Gram-negative pathogens (Cronobacter sakazakii, Cronobacter muytjensii, Salmonella enterica serovar Typhimurium, Escherichia coli, Klebsiella pneumoniae, and Pseudomonas fluorescens) as well as the Gram-positive organism Staphylococcus aureus. Here we describe further studies with the aim of establishing the importance of specific (charged and nonpolar aliphatic) residues within the caseicin peptides and the effects that they have on the bacteria listed above. In order to achieve our objective, we created four derivatives of each caseicin (A1 to A4 and B1 to B4) in which specific residues were altered, and results obtained with these derivatives were compared to wild-type caseicin activity. Although conversion of cationic residues to alanine in caseicins B1 (R8A change), A1 (K2A), A2 (H3A), and A3 (K2A-H3A) generally resulted in their activity against microbial targets being reduced or unaltered, C. sakazakii DPC6440 was unusual in that it displayed enhanced sensitivity to three peptides (caseicins A1, A3, and B2) in which positively charged residues had been eliminated. While the replacement of leucine with alanine in selected variants (B3 and B4) resulted in reduced activity against a number of strains of Cronobacter and, in some cases, S. Typhimurium, these changes enhanced the activities of these peptides against DPC6440 and a number of S. aureus strains. It is thus apparent that the importance of specific residues within the caseicin peptides is dependent on the strain being targeted.

In recent years, the milk protein casein has been identified as a rich source of antimicrobial peptides, many of which have been characterized extensively. The casein-derived caseicin A and B peptides, which are the subjects of this study, were originally identified following a bacterial fermentation of sodium caseinate and were previously shown to exhibit antibacterial activity against strains of Cronobacter sakazakii and Escherichia coli (15). These and a number of other casein-derived antimicrobials are primarily cationic in nature, and thus it was plausible that these functioned in a manner similar to that for classical cationic antimicrobial peptides (cAMPs). As well as being cationic, cAMPs are amphipathic and hydrophobic α-helical peptides which are active against a range of microbial targets as a consequence of permeabilization of the cytoplasmic membrane leading to loss of the proton gradient (6, 21, 25, 37). The importance of the charged residues within cAMPs is apparent from studies on peptide derivatives with reduced cationic charges, which were shown to exhibit reductions in antimicrobial activity (19). These observations prompted the initiation of corresponding investigations, described here, to assess the importance of a number of residues, including cationic amino acids, in caseicins A and B.

The antimicrobial properties of peptides derived from milk proteins have long been recognized. In 1930, Simms and Jones isolated the first antimicrobial peptide, lactenin, which was generated following the treatment of milk with rennin (20) and is active against pathogenic streptococci (42). A number of antimicrobial peptides have also been generated specifically from casein. These include casecidin, an antibacterial glycopeptide which inhibits the growth of Escherichia coli (16), and casocidin I, a highly cationic peptide which also inhibits E. coli as well as Staphylococcus carnosus (45). Another such example is the positively charged peptide isracidin, which corresponds to the 23 N-terminally located residues of bovine α_s1-casein B and which inhibits Staphylococcus aureus, Listeria monocytogenes, and Candida albicans (23). The positive charge of these peptides suggests that they may be comparable to cationic antimicrobial peptides such as cecropin (17), indolicidin (36), cathelicidins (26, 43, 44), and defensins (5, 13, 38), all of which rely on charge to facilitate an interaction with the negatively charged membranes of target bacteria. However, the importance of charged residues in casein-derived peptides has not been assessed before.

The positively charged antimicrobial peptide caseicin A and the neutral peptide caseicin B were recently found to inhibit strains of C. sakazakii, E. coli, Listeria innocua, and Streptococcus mutans (15). Both peptides were derived from bovine α_s1-casein B, with caseicin A (IKHQGLPQE) and caseicin B (VLNENLLR) corresponding to amino acids 6 to 14 and 15 to 22, respectively, of isracidin. In this study, the activities of caseicins A and B against an even wider range of pathogens were tested, and the importance of specific residues was investigated through the creation and assessment of the antimicrobial activities of variant peptides.

MATERIALS AND METHODS

Bacterial strains and culture media.

Bacterial strains used in this study are listed in Table 1. All microbes were cultured aerobically at 37°C in LB medium (Oxoid, Basingstoke, United Kingdom).

TABLE 1.

Strains used in this study

Strain	Description	Source or referencea
Gram-negative strains
C. muytjensii ATCC 51329	Unknown source	30; ATCC
C. sakazakii BAA894	Infant isolate	ATCC
C. sakazakii DSM4485 (ATCC 2954)	Child's throat isolate	ATCC
C. sakazakii DPC6440 (NCTC08155)	Tin of dried milk	NCTC
E. coli NCIMB 11843	Laboratory strain	NCIMB
S. Typhimurium LT2	Laboratory strain	UCC
K. pneumoniae NCIMB 13218	Laboratory strain	NCIMB
P. fluorescens NCIMB 9046	Laboratory strain	NCIMB
Gram-positive strains (S. aureus strains)
Sa113 (ATCC 35556)	Derivative of NCTC8325	18
RN4220	Restriction-defective derivative of NCTC8325-4	22
Newman	Osteomyelitis isolate	8
8325-4	Laboratory strain	32
RF122 (ET3-1)	Bovine mastitis-causing isolate	12

^aATCC, American Type Culture Collection; NCTC, National Collection of Type Cultures; NCIMB, National Collection of Industrial, Food and Marine Bacteria; UCC, University College Cork.

Peptide synthesis.

Caseicin peptides were synthesized chemically using microwave-assisted solid-phase peptide synthesis (MW-SPPS) performed on a Liberty CEM microwave peptide synthesizer. Caseicin A was made on H-Glu(OtBu)-HMPB-ChemMatrix resin, and caseicin B was made on H-Arg(Pbf)-HMPB-ChemMatrix resin (PCAS Biomatrix Inc., Quebec, Canada). Synthetic peptides were purified using reversed-phase high-performance liquid chromatography (RP-HPLC) on a Vydac C₁₈ (10 μm by 300 Å) preparative column (Grace Davison Discovery Sciences, CA) developed in a gradient of 10 to 20% acetonitrile-0.1% trifluoroacetic acid (TFA) over 25 min for caseicin A and 15 to 35% acetonitrile-0.1% TFA over 25 min for caseicin B. Fractions containing peptides of the correct molecular mass were identified using matrix-assisted laser desorption ionization-time of flight (MALDI-TOF) mass spectrometry, pooled, and lyophilized on a Genevac HT 4X (Genevac Ltd., Ipswich, United Kingdom) lyophilizer. The peptides were stored at −20°C. In advance of use, the peptides were resuspended in LB broth and stored at 4°C.

MIC determination.

MICs were determined as described previously (41). Briefly, serial 2-fold dilutions of the antimicrobial peptides were made in LB broth and added to the wells of a 96-well plate. All strains were grown in LB broth to an optical density at 600 nm (OD₆₀₀) of 0.5. Bacteria were added to give a final inoculum of 10⁵ CFU per milliliter in a final volume of 200 μl. After incubation for 16 h at 37°C, the MIC was read as the lowest peptide concentration that prevented visible growth. Results are given as the mean values for three independent determinations.

Assessment of the growth of target strains in the presence of caseicin peptides and variants thereof.

Bacterial growth in the presence or absence of antimicrobial peptides was assessed using a 96-well plate spectrophotometer (MDS Inc., CA). Target bacteria were inoculated from an overnight culture (16 h) into fresh medium and incubated at 37°C until an OD₆₀₀ of 0.5 was reached. Twenty microliters of this culture was added to 1 ml of fresh broth, and following 1/10 dilution, 100 μl was added to each of the 96 wells. The relevant concentration of reconstituted peptides was added (100 μl), and the plate was incubated for 23 h at the temperature of choice. Growth was assessed spectrophotometrically (OD₆₀₀) at 60-min intervals. All experiments were performed in triplicate.

RESULTS

Design and synthesis of caseicin A and B variants.

Caseicin A is a cationic peptide with an overall positive charge of +1, which is a consequence of the presence of two positively charged amino acids, i.e., K2 and H3, as well as the negatively charged amino acid E9, while caseicin B is a neutral peptide possessing only 1 positively charged residue, i.e., R8, which is neutralized by the negatively charged glutamate residue, E4. To understand the importance of these and other amino acids, a number of variants of these peptides were designed. Thus, caseicin A1 was designed such that a K2A change was made, the caseicin A2 variant contained an H3A alteration, and both changes, i.e., K2A-H3A, were incorporated into caseicin A3 (Table 2). Similarly, the caseicin B1 variant was designed to contain an R8A change, while caseicin B2 also lacked R8, but in this instance no substitution was made (Table 2). A number of additional variants in which selected nonpolar aliphatic residues were manipulated were also generated. This involved the removal of such a residue (i.e., ΔI1 mutation in caseicin A4) or its replacement with alanine (i.e., L7A and L6A mutations in caseicins B3 and B4, respectively). All peptides were synthesized and purified, and mass spectrometry was employed to confirm that peptides of the correct mass were generated.

TABLE 2.

Sequences and masses of caseicin variants

Caseicin	Amino acid sequencea	Mass (Da)
		Predicted	Actual
A (wild-type control)	IKHQGLPQE	1,049	1,049.19
A1 variant	IAHQGLPQE	992	992.09
A2 variant	IKAQGLPQE	983	983.13
A3 variant	IAAQGLPQE	926	926.03
A4 variant	(ΔI)KHQGLPQE	918	936.03
B (wild-type control)	VLNENLLR	970	970.13
B1 variant	VLNENLLA	885	885.02
B2 variant	VLNENLL(ΔR)	796	813.94
B3 variant	VLNENLAR	928	928.05
B4 variant	VLNENALR	928	928.05

^aAmino acid changes are indicated in bold. Δ, removal of an amino acid without its replacement.

Activities of caseicin and caseicin variants against a range of target pathogens.

MIC determination assays were carried out to quantify the activities of caseicins A and B against 8 Gram-negative pathogens, i.e., C. sakazakii DPC6440, DSM 4485, and BAA894, Cronobacter muytjensii ATCC 51329, Salmonella enterica serovar Typhimurium LT2, E. coli NCIMB 11843, Klebsiella pneumoniae NCIMB 13218, and Pseudomonas fluorescens NCIMB 9046. The MIC of caseicin A against C. sakazakii DSM4485, DPC6440, and BAA894 and C. muytjensii ATCC 51329 was 0.625 mM in each case, whereas Salmonella LT2, P. fluorescens, K. pneumoniae, and E. coli were 2-fold more resistant (i.e., MIC of 1.25 mM) (Table 3). These targets were equally sensitive to caseicin B, with the exception of C. sakazakii DPC6440, which was 2-fold more resistant to caseicin B (MIC of 1.25 mM).

TABLE 3.

MICs of caseicins against Gram-negative and Gram-positive targets

Strain	MIC (mM)
	A (WT)	B (WT)	A1	A2	A3	A4	B1	B2	B3	B4
C. sakazakii DPC6440	0.625	1.25	0.312	0.625	0.312	1.25	2.5	0.312	0.625	0.625
C. sakazakii DSM 4485	0.625	0.625	2.5	2.5	2.5	2.5	2.5	2.5	2.5	2.5
C. sakazakii BAA894	0.625	0.625	2.5	2.5	2.5	2.5	2.5	2.5	2.5	2.5
C. muytjensii ATCC 51329	0.625	0.625	2.5	2.5	2.5	2.5	2.5	2.5	2.5	2.5
S. Typhimurium LT2	1.25	1.25	1.25	1.25	1.25	1.25	1.25	1.25	2.5	2.5
E. coli NCIMB 11843	1.25	1.25	1.25	1.25	1.25	1.25	2.5	1.25	1.25	1.25
K. pneumoniae NCIMB 13218	1.25	1.25	1.25	1.25	1.25	1.25	2.5	1.25	1.25	1.25
P. fluorescens NCIMB 9046	1.25	1.25	1.25	1.25	1.25	1.25	2.5	1.25	1.25	1.25
S. aureus 8325-4	1.25	1.25	1.25	1.25	1.25	1.25	1.25	1.25	0.625	1.25
S. aureus Sa113	1.25	1.25	1.25	1.25	1.25	0.625	1.25	1.25	0.625	0.625
S. aureus RF122	1.25	1.25	1.25	1.25	1.25	1.25	1.25	1.25	1.25	1.25
S. aureus Newman	1.25	1.25	1.25	1.25	1.25	1.25	1.25	1.25	1.25	0.625
S. aureus RN4220	1.25	1.25	1.25	1.25	1.25	1.25	1.25	1.25	1.25	1.25

MIC-based assessments of the antimicrobial activities of the caseicin variants were also undertaken with a view to identifying peptides with reduced activity, thereby highlighting amino acids which make important contributions to the activities of the corresponding wild-type peptides. It was noted that the lysine-to-alanine change in caseicin A1 resulted in at least a 2-fold reduction in activity (to an MIC of 2.5 mM) against two (DSM4485 and BAA894) of the three C. sakazakii strains and against the C. muytjensii strain. In contrast, activity was not reduced against the other Gram-negative targets, and activity was increased against C. sakazakii strain DPC6440. The histidine-to-alanine change in A2 and the combination of the individual A1- and A2-associated mutations in A3 also had the same impacts on the Gram-negative targets, with the exception of C. sakazakii DPC6440. Against this strain, A2 retained the original level of activity (0.625 mM) and A1 and A3 were unexpectedly more active (0.312 mM) than the caseicin A control. Finally, caseicin A4, the only caseicin A variant in which the positively charged residues were not altered, exhibited diminished activity against three of the Cronobacter strains but retained original levels of activity against the other Gram-negative targets and against C. sakazakii DPC6440.

Of the caseicin B peptides, the B1 variant, in which the C-terminal arginine was converted to an alanine, was unique in that it was the only variant that exhibited reduced activity (MIC of 2.5 mM) against seven of the eight Gram-negative targets (with the exception being S. Typhimurium LT2). This loss of activity against E. coli NCIMB 11843, K. pneumoniae NCIMB 9046, and P. fluorescens NCIMB 13218 cannot be attributed solely to the loss of the C-terminal arginine, as caseicin B2, in which this residue was removed but not replaced, retained wild-type levels of activity against this target. Thus, it appears that the introduction of a C-terminal alanine, as occurred in caseicin B1, has a specifically negative impact. Caseicin B2 did, however, exhibit reduced activity against three of the Cronobacter strains, although DPC6440 was again atypical in that, as with caseicins A1 and A3, the absence of charged residues resulted in the peptide having enhanced antimicrobial activity (MIC of 0.312 mM) against this target. Finally, caseicins B3 and B4 were similar with respect to the nature of the changes made and possessed identical levels of activity. They were exceptional in that they were the only variants that exhibited reduced activity against S. Typhimurium LT2. Activity was also reduced against the Cronobacter strains, with the exception of DPC6440 yet again, but was unaltered against E. coli NCIMB 11843, K. pneumoniae NCIMB 13218, and P. fluorescens NCIMB 9046.

The relative impacts of a specific concentration (1.25 mM) of the antimicrobial peptides against a variety of the targets, i.e., C. sakazakii DPC6440 and BAA894, C. muytjensii ATCC 51329, E. coli NCIMB 11843, Klebsiella pneumoniae NCIMB 13218, and Pseudomonas fluorescens NCIMB 9046, were also determined by assessing growth over a 23-h period (Fig. 1 a to f). These were consistent with MIC-based experiments in that the relatively reduced potency of caseicin B1 against C. sakazakii DPC6440, E. coli NCIMB 11843, K. pneumoniae NCIMB 13218, and P. fluorescens NCIMB 9046 was again very evident, while none of the variants retained antimicrobial activity against C. sakazakii BAA894 or C. muytjensii ATCC 51329.

FIG. 1.

Impacts of exposure of C. sakazakii DPC6440 (a) and BAA894 (b), C. muytjensii ATCC 51329 (c), E. coli NCIMB 11843 (d), K. pneumoniae NCIMB 13218 (e), and P. fluorescens NCIMB 9046 (f) to 1.25 mM caseicin peptide, or different forms thereof, at 37°C in LB broth. Blue diamonds, control culture with no added peptide; red squares, wild-type caseicin A; green triangles, wild-type caseicin B; purple x's, variant A1; light blue stars, variant A2; orange circles, variant A3; light blue vertical lines, variant A4; pink horizontal lines, variant B1; green horizontal lines, variant B2; purple diamonds, variant B3; light blue squares, variant B4. Results are the averages for experiments carried out in triplicate. Error bars represent standard deviations.

Activities of caseicin variants against S. aureus.

MIC determination assays were carried out to quantify the activities of caseicins A and B against a number of strains of the Gram-positive pathogen Staphylococcus aureus, i.e., Sa113, 8325-4, RF122, Newman, and RN4220. Caseicins A and B were equally active (MIC of 1.25 mM) against all S. aureus strains tested (Table 3). An assessment of the antimicrobial activities of the caseicin variants against these S. aureus targets revealed that none of the variants exhibited reduced activity against these targets. There were, however, a number of examples of variants with 2-fold enhanced activity (MIC of 0.625 mM) against specific targets, all of which resulted from the manipulation of nonpolar aliphatic amino acids (i.e., caseicins A4, B3, and B4). Caseicin A4 was more active than caseicin A against one target, strain Sa113; caseicin B3 was more active (MIC of 0.625 mM) against two targets, Sa113 and 8325-4; and caseicin B4 exhibited enhanced activity against Sa113 and Newman.

DISCUSSION

The antimicrobial activity of milk and milk proteins such as casein has been recognized widely since as far back as 1930 (20). Casein is a milk protein which contains numerous antimicrobial peptides encrypted within its protein sequence, many of which have been characterized by research groups worldwide (4, 15, 23, 27-29, 34, 45). In a previous study, the casein-derived peptides caseicin A and caseicin B were shown to exhibit antimicrobial activity against certain pathogenic strains of E. coli and C. sakazakii (15). In this study, a range of caseicin variants, A1 to A4 and B1 to B4, were designed and generated, with the aims of identifying essential amino acids within the peptide sequences and ascertaining the consequences of altering the charges of these peptides. In addition, the range of target microorganisms tested in this study was larger than that in the original study (15), with a view of determining if the caseicin peptides are active against a broad range of Gram-negative targets as well as a representative Gram-positive target. While all of the targets are pathogens, the inclusion of C. sakazakii and S. aureus as targets is particularly important in light of their significance with respect to the food industry.

Cronobacter sakazakii (previously known as Enterobacter sakazakii) is a Gram-negative, peritrichously motile, non-spore-forming bacterium which has been the cause of many serious diseases in humans, including meningitis, hydrocephalus, necrotizing enterocolitis, septicemia, and brain abscesses (3, 31, 40). Infants, especially those who are immunocompromised or of low birth weight, are a major risk group for infection, whereas adults are rarely infected (7, 14, 24, 33, 35). There have been a total of 120 cases of C. sakazakii infection, with 27 deaths being reported for infants alone (9). C. sakazakii is a particularly significant concern for the powdered infant milk formula industry as a consequence of its ability to withstand extreme stresses. The resistance of the species to stress, combined with the fact that traditional preservatives cannot be added to powdered infant milk formula (11), means that the development of novel strategies to control this pathogen in powdered infant milk formula is highly desirable. Although the amount of peptide needed to inhibit C. sakazakii is quite high in comparison to those for classical antimicrobials, combined with the low water activity (a_w) of powdered infant formula and the presence of a number of other constituents, such as high sugar levels, it is anticipated that the presence of caseicins will have beneficial consequences. The other pathogen of note, S. aureus, is a Gram-positive, non-spore-forming, facultative anaerobe which causes staphylococcal food poisoning (2, 39). It is found mainly on the skin of warm-blooded animals but has been isolated from dairy products, eggs and egg products, seafood, meat, and meat products (10). The fact that caseicins have the ability to inhibit the growth of a range of S. aureus strains is thus potentially of great significance.

The results of our study showed that caseicin B1 was impacted most dramatically by the nature of the amino acid change made, i.e., replacement of the C-terminal arginine of caseicin B with an alanine. The B1 variant exhibited consistently poor activity against the Gram-negative strains tested, including DPC6440 and NCIMB 11843, against which other variants were consistently more active; however, removal of the positive charge had no effect on the activity against the Gram-positive strains tested. The consequences of removing but not replacing the C-terminal arginine in caseicin B2 were not as severe, resulting in its activity being reduced against C. sakazakii DSM 4485 and BAA894 and C. muytjensii ATCC 51329 but not against LT2, NCIMB 11843, or NCIMB 13218. This suggests that while the loss of this cationic residue alone impacts negatively on activity against some Cronobacter targets, this effect is further amplified through the introduction of an alanine. Furthermore, B2 surprisingly exhibited quite potent activity against C. sakazakii DPC6440. This was only one of a number of MIC-based assays in which DPC6440 showed an unusual sensitivity to all caseicin variants except caseicin B1. The basis for the exquisite sensitivity of this strain is not clear, as it has not been assessed in depth before now. It is worth noting, however, that the various sensitivities of different strains of C. sakazakii to environmental stresses such as heat (1) have been noted previously. The impacts on caseicin A arising from the removal of charged residues (A1 to A3) were similar to those seen for B2, in that while activity against some Cronobacter targets was quite reduced, activity against other Gram-negative indicator strains and S. aureus was not affected. This suggests that the mechanisms of action of the caseicins may differ in a target-specific manner. The fact that the changes made to generate A1, A3, and B2 had surprisingly beneficial impacts on relative activities against C. sakazakii DPC6440 again highlights the atypical behavior of this target. There were also a few instances where changes which did not impact charged residues, i.e., the changes in variants A4, B3, and B4, had beneficial impacts with respect to activities against specific S. aureus targets, i.e., Sa113 (A4, B3, and B4), 8325-4 (B3), and Newman (B4). Other than these isolated instances, the consequences of manipulating the (iso) leucine residues in these peptides were quite similar to those resulting when charged residues were manipulated, thereby indicating that the latter do not have an especially important role with respect to the antimicrobial activity of the caseicin A and B peptides.

In conclusion, in the process of demonstrating that caseicins A and B are not typical cationic antimicrobial peptides, we have highlighted the potential benefits of generating derivatives of these peptides. While derivatives with enhanced activity may not be commercially relevant as a consequence of lacking food-grade status, the creation of these and large collections of variants will enable researchers to gain a deeper insight into the importance of specific amino acids within these peptides. The generation of caseicin derivatives will also be important with respect to elucidating the mechanism of action of these and related peptides, which in turn will lead to the development of approaches to optimize their application as food-grade antimicrobials capable of controlling C. sakazakii and other pathogens in infant formula and other foods.