Scientific Publications - Work Done by Microbiology Reader Bioscreen C

Cellular and Molecular Life Sciences (CMLS), Volume 61, Number 10, Date: May 2004, Pages: 1229-1237

Periplasmic lysozyme inhibitor  contributes to lysozyme  resistance in Escherichia coli

D. Deckers, B. Masschalck, A. Aertsen, L. Callewaert, C. G. M. Van Tiggelen, M. Atanassova and C. W. Michiels

ABSTRACT

TAbstract The product of the Escherichia coli ORFan gene ykfE was recently shown to be a strong inhibitor of C-type lysozyme in vitro. The gene was correspondingly renamed ivy (inhibitor of vertebrate lysozyme), but its biological function in E. coli remains unknown. In this work, we investigated the role of Ivy in the resistance of E. coli to the bactericidal effect of lysozyme in the presence of outer-membrane-permeabilizing treatments. Both in the presence of lactoferrin (3.0 mg/ml) and under high hydrostatic pressure (250 MPa), the lysozyme resistance of E. coli MG1655 was decreased by knock-out of Ivy, and increased by overexpression of Ivy. However, knock-out of Ivy did not increase the lysozyme sensitivity of an E. coli MG1655 mutant previously described to be resistant to lysozyme under high pressure. These results indicate that Ivy is one of several factors that affect lysozyme resistance in E. coli, and suggest a possible function for Ivy as a host interaction factor in commensal and pathogenic E. coli.

Key words: Lysozyme - lysozyme inhibitor - Escherichia coli - defense mechanism - Ivy

INTRODUCTION

Lysozyme is the common name of a group of enzymes that can hydrolyze the bacterial cell wall polymer pepti­doglycan by virtue of an N-acetylmuramoylhydrolase ac­tivity (E.C. 3.2.1.17). Lysozymes are commonly found in the major taxa of prokaryotes and eukaryotes, including the bacteria themselves, and also in bacteriophages, but they may have different roles in different organisms. For example, animal and plant lysozymes are important play­ers in the defense against bacterial invaders [1–5], phage lysozymes play a role in phage penetration into, and/or release from the host cell [6, 7], while some bacterial lysozymes, called autolysins, allow controlled hydrolysis of the cell wall at sites of cell growth or cell division [8, 9]. The widespread natural occurrence of lysozyme, even in the human body, and its specificity against bacteria (peptidoglycan occurs exclusively but almost universally in the phylum of the Bacteria), has fueled the develop­ment of applications of this enzyme as an antibacterial agent in foods, pharmaceuticals and cosmetics, and even for treating bacterial infections [10, 11]. One of the ma­jor limitations of lysozyme in this type of application is its poor efficiency against some bacteria. In particular, Gram-negative bacteria are generally insensitive to lysozyme because their peptidoglycan is protected by the surrounding outer membrane layer. However, two differ­ent approaches have proven useful to improve the effi­ciency of lysozyme against Gram-negative bacteria [12]. First, lysozyme can be structurally modified such that it can cross the outer membrane, probably by a mechanism related to self-promoted uptake [13, 14]. This has been accomplished by conjugation with a variety of molecules including fatty acids [15, 16], and also by extension of the polypeptide with a stretch of hydrophobic amino acids using a recombinant genetic approach [17]. Alternatively, the barrier function of the outer membrane can be dis­rupted to allow passage of (unmodified) lysozyme. This can be achieved by substances that complex or displace the stabilizing Ca²⁺ and Mg²⁺ ions from the outer leaflet of the outer membrane, such as EDTA or the polycation lactoferrin [2, 18–20], but also by some physical treat­ments, in particular high pressure [21–23]. Nevertheless, for example under high pressure, not all bacteria in a pop­ulation appear to become sensitive to lysozyme [24, 25]. Studies with Salmonella enterica serovar Typhimurium and Pseudomonas aeruginosa outer-membrane mutants suggested that variations in outer membrane composition in bacterial populations may explain these differences in lysozyme susceptibility under high pressure, but other factors may be involved as well, such as variations in pep­tidoglycan structure, or the presence of different levels of lysozyme inhibitors [26].

The existence of a lysozyme inhibitor in Escherichia coli has been discovered only recently. When the E. coli or­phan gene ykfE was fused to a His tag and overexpressed to investigate its gene product, lysozyme, used to lyse the cells, was co-purified with the YkfE protein [27]. YkfE was subsequently found not only to bind specifically and strongly to lysozyme, but also to inhibit its enzymatic ac­tivity. Since inhibition was specific for C-type lysozymes (including besides hen egg white, also human lysozyme), ykfE was renamed ivy (inhibitor of vertebrate lysozyme). It was the first well-characterised proteinaceous lysozyme inhibitor to be described. Open reading frames with vari­ous levels of homology to the E. coli ivy gene exist in a number of other proteobacteria belonging to the genera Shigella, Yersinia, Pseudomonas, Chromobacterium and Burkholderia [unpublished results], but their lysozyme­inhibiting activity has not been demonstrated. In view of the widespread occurrence of lysozyme and its important role in antibacterial defense in animals including humans, a hypothetical function of lysozyme inhibitors in com­mensal or pathogenic strains of these Gram-negative bac­teria could be to provide an extra layer of defense against the host lysozyme, when the outer membrane becomes permeabilized by components of the innate defense such as lactoferrin and defensins. Other possible functions in­clude protection against bacteriophages, or regulation of autolysin activity during cell growth and cell division. A contribution of lysozyme inhibitors to bacterial lysozyme resistance would also have important consequences for the development of applications of lysozyme in food, pharmaceuticals, cosmetics and medicine. In this work, we investigated the role of Ivy in lysozyme resistance of E. coli, using Ivy knock-out and overexpression strains, and using high-pressure treatment or lactoferrin to per­meabilize the outer membrane.

Material and methods

Construction of an arabinose-inducible Ivy overexpressor strain

A PCR- (Platinum Pfx DNA polymerase; Invitrogen, Merelbeke, Belgium) amplified fragment containing araC and the arabinose-inducible _PBAD promoter of the araBAD operon [28] of E. coli MG1655 was cloned up­stream of promoterless gfp in pFPV25 [29] cut with EcoRI and XbaI. The gfp gene was subsequently removed by cleavage with XbaI and HindIII and replaced by the MG1655 ivy gene, amplified with the primers: ivyF (5'-gatctctagacgccaggctttaggagg-3) and ivyR (5'-gat­caagcttcggagccgaaaggctcc-3) and cut with the same en­zymes. All restriction enzymes were purchased from Roche Diagnostics Belgium (Vilvoorde, Belgium). The resulting construct contains the ivy gene under control of the _PBAD promoter and was designated pAA410. This plasmid was transformed into E. coli MG1655 and the resulting strain was designated as E. coli MG1655 (pAA410).

Construction of ivy knock-out mutants

E. coli TE2680, containing a chromosomal ivy::Kan knock-out was kindly donated by the laboratory of J. M. Claverie (CNRS Institute of Structural Biology and Microbiology, Marseille, France). From this strain, a ± 1.7-kb ivy::Kan fragment was obtained by PCR, using primers ivyF and ivyR, and transformed to MG1655 con­taining the pKD46 plasmid which strongly enhances the rate of allelic exchange with linear double-stranded DNA fragments [30]. A Kan-resistant transformant was puri­fied and Kan resistance was transduced by P1vir [31] to MG1655 and LMM1010 (a high-pressure-resistant deriv­ative of MG1655, isolated by Hauben et al. [22]. Re­placement of ivy by the ivy::Kan allele was confirmed by PCR and the resulting strains were designated as E. coli MG1655 ivy::Kan and LMM1010 ivy::Kan.

Growth conditions

Luria Bertani (LB, 10 g/l tryptone, 5 g/l yeast extract, 5 g/l NaCl) agar master plates were inoculated from –80°C stocks and contained ampicillin (100 )lg/ml) for MG1655 (pAA410) and kanamycin (50 )ig/ml) for MG1655 ivy::Kan and LMM1010 ivy::Kan. All antibi­otics were purchased from Sigma-Aldrich (Bornem, Bel­gium). Cultures for lysozyme sensitivity experiments were grown overnight at 37°C in tryptone soy broth (TSB; Oxoid, Basingstoke, U. K.) without antibiotics, di­luted 1/100 in fresh TSB and incubated further at 37°C. At an optical density _(OD600) of 0.4, 0.2% (w/v) L-(+)-arabinose (Fluka, Buchs, Switzerland) was added and growth was continued at 37°C until an _OD600 around 0.9 (early stationary phase, approx. 5 x 10⁸ CFU/ml). Al­though arabinose served only to induce ivy expression from plasmid pAA410, it was also added to cultures of strains not carrying this plasmid to ensure identical cul­ture conditions for all strains in an experiment. Addition of arabinose was omitted in experiments not involving pAA410.

Lysozyme inhibition assay

The production of Ivy by the different E. coli strains was analyzed by measuring the inhibition of lysozyme enzy­matic activity by E. coli cell extracts and by periplasmic fractions (see below for preparation). Cell extracts were made by sonication (2 ¥ 5 min, amplitude 40%, pulse 5 s on/5 s off; Sonics & Materials SMC, Danbury, Conn.) of early stationary phase cells (5 ¥ 10⁸ CFU/ml) in potas­sium phosphate buffer (10 mM, pH 7.0), followed by cen­trifugation (24,000 g, 30 min) to remove cell debris. The extracts were stored at –20°C. The inhibition of lysozyme activity was measured using an adapted lysozyme activ­ity assay of Weisner [32], which is based on the decrease in turbidity of a culture of lyophilized Micrococcus lysodeikticus. First, freeze-dried M. lysodeikticus ATCC 4698 (two different batches were used, Sigma, St. Louis, Mo.) were resuspended at 0.5 mg/ml in the cell extracts or in different dilutions of the periplasmic fractions of E. coli or in potassium phosphate buffer (10 mM, pH 7.0) for the control. Thirty microliters of 10 mg/ml hen egg white lysozyme (66,000 U/mg; Fluka) in potassium phos­phate buffer (10 mM, pH 7.0) was added to 270 ml of M. lysodeikticus cell suspension (final working concentra­tion 1.0 mg/ml lysozyme), and cell lysis was followed during 4 h at 20°C as the decrease in optical density (^OD600) using a Bioscreen C Microbiology Reader (Lab­systems Oy, Helsinki, Finland). Inhibition was expressed as the percent reduction in the initial enzymatic reaction rate (decrease in ^OD600^/h)^.

Isolation of periplasmic proteins by osmotic shock

An E. coli culture grown for 21 h in 50 ml LB broth at 37°C, was diluted 100-fold in 350 ml LB broth and in­duced with 0.2% arabinose when the _OD600 reached 0.5. After 4 h of further incubation, 200 ml of the culture was harvested by centrifugation (10 min, 2900 g, room temper­ature). The cell pellet was resuspended in 25 ml 30 mM Tris-HCl (pH 8) with 20% (w/v) sucrose. After addition of 2.5 ml 10 mM EDTA and 10 min shaking at room temper­ature, the sample was centrifuged (10 min, 6350 g, 4°C), and the cell pellet resuspended in 25 ml ice-cold 5 mM MgSO₄, and shaken for 10 min in an ice-waterbath. This suspension was centrifuged again (10 min at 16,800 g, 4°C), and the supernatant, corresponding to the periplas­mic fraction, was stored at –20°C until further analysis.

SDS-polyacrylamide gel electrophoresis

SDS-PAGE was conducted according to the Laemmli method [33], with a 12% separating gel and a 4% stacking gel. Samples of the periplasmic protein extracts were mixed with an equal volume of denaturation buffer con­taining b-mercaptoethanol and SDS and boiled for 3 min in a water bath. Fifty microliters of the samples was loaded per lane, and flanked by a lane with molecular­weight markers (low-molecular-weight calibration kit, Amersham Biosciences, Uppsala, Sweden). After elec­trophoresis, gels were stained overnight with Coomassie blue R 250 [300 ml/l ethanol, 50 ml/l acetic acid and 5 mg/l Coomassie blue R 250 (Sigma-Aldrich)] followed by a series of destaining steps in a solution with 300 ml/l ethanol and 50 ml/l acetic acid. For higher sensitivity, gels were subsequently fully destained and silver-stained by the procedure of Heukenshoven and Dernick [34].

Amino acid sequencing by mass spectrometry

A protein band suspected to correspond to Ivy was sam­pled from a Coomassie-stained gel and trypsin-digested according to the method of Shevchenko et al. [35]. Tryp­tic digests were then analyzed by electrospray tandem mass spectrometry on a LCQ Classic (ThermoFinnigan, San Jose, Calif.) ion trap mass spectrometer equipped with a nano-liquid chromatography column switching system and a nanoelectrospray device. Tandem mass spectrometry data were searched using MASCOT (Ma­trix Sciences, London, U. K.) and SEQUEST (Ther­moFinnigan) against the GenBank non-redundant protein database.

Sensitivity of E. coli to lysozyme under high pressure Cultures were grown to early stationary phase and in­duced with arabinose as described above, except for the experiment involving strain LMM1010, in which growth was for 21 h (stationary phase, approx. 1 ¥ 10⁹ CFU/ml) because this strain was reported to be resistant to lysozyme under high pressure when it was grown under these conditions [21]. The cells were harvested by cen­trifugation (3800 g, 5 min) and resuspended in the same volume of potassium phosphate buffer (10 mM, pH 7.0). Although the pH of phosphate buffer is more pressure de­pendent than that of some other buffers [36], we preferred to use phosphate buffer because it is widely used in bac­terial inactivation studies and is not harmful to E. coli. Portions of this cell suspension (270 ml), supplemented with 30 ml of lysozyme (final concentration 1.0 or 10.0 µg/ ml) or 30 ml of buffer for the controls, were sealed in a sterile polyethylene pouch after exclusion of air bubbles and subjected at 20°C to high-pressure treatment at 250 or 600 MPa in an 8-ml high-pressure vessel (Resato, Ro­den, The Netherlands), which was thermostatically con­trolled with a water jacket. Compression and decompres­sion were achieved with a manual spindle pump at a rate of approximately 100 MPa per minute. Under these con­ditions, the adiabatic temperature increase in the vessel increased to 38°C upon pressurization to 600 MPa. The high-pressure transmission fluid was TP1, a mixture of glycols (Van Meeuwen, Weesp, The Netherlands). After decompression, the samples were aseptically removed from the pouches, serially diluted in sterile potassium phosphate buffer (10 mM, pH 7.0) and, along with un­treated control cultures, plated on tryptone soy agar (Ox­oid) with a spiral plater (Spiral Systems Inc., Cincinnati, Ohio). The plated volume was 50.0 ml, and hence the lower detection limit was 20 CFU/ml. Colonies were al­lowed to develop for 24 h at 37°C. Inactivation was ex­pressed as a logarithmic viability reduction factor, ^log10 N_o/N, where N_o and N are the colony counts of control and treated samples, respectively.

Sensitivity of E. coli to lysozyme in the presence of lactoferrin

Cultures were grown as described above, harvested by centrifugation (3800 g, 5 min) and resuspended in the same volume of potassium phosphate buffer (10 mM, pH 7.0) without and with lactoferrin (gift from Morinaga Milk Industries, Kanagawa, Japan; 3.0 mg/ml final con­centration) and/or lysozyme (50 mg/ml). After 7 h of in­cubation at 20°C, the number of survivors was enumer­ated using the same plate count procedure as for pressure­treated cell suspensions. Inactivation was expressed as a viability reduction factor, N_o/N, where N_o and N are the colony counts of control and treated samples, respec­tively.

Statistical analysis

All data shown are averages ± standard deviations (SD) from three experiments with three independent cultures of each strain. The possible effect of high pressure/lacto­ferrin and lysozyme was investigated using a two-factor ANOVA model. The main effects of high pressure/lacto­ferrin and lysozyme as well as their interaction were in­cluded in the model. Following significant main effects or interactions, post hoc comparisons of treatment means were carried out using a Bonferroni multiple-comparison procedure with a 5% family significance level [37]. The SAS software (SAS version 8.2; SAS Institute Inc., Cary, N. C.) was used throughout all analyses.

Results

Overexpression and deletion of Ivy in E. coli MG1655 Ivy knock-out and overexpression strains [respectively MG1655 ivy::Kan and MG1655 (pAA410)] were con­structed as described in Materials and methods, and sub­jected to a rapid evaluation by comparing the lysozyme inhibition activity of their sonicated cell extracts with those obtained from wild-type strain MG1655. Figure 1A shows the progression of M. lysodeikticus lysis by lysozyme in the presence of these cell extracts. The presence of cell extracts increased the initial _OD600 by 0.05. The rate of cell lysis by lysozyme (0.50 _OD600/h in the ab­sence of E. coli cell extracts) was reduced by 74% by cell extract from the Ivy knock-out strain (0.13 _OD600/h), by 84% by the wild-type extract (0.08 _OD600/h) and com­pletely inhibited by the extract from the Ivy overexpres­sor strain.

Figure 1. Lysis (expressed as OD₆₀₀ against time) of M. lysodeikti­cus cell suspension (0.5 mg/ml) by 1.0 mg/ml lysozyme in the ab­sence (dotted line) and presence of undiluted sonicated cell extracts of E. coli MG1655 (diamonds), MG1655 (pAA410) (triangles) and MG1655 ivy::Kan (crosses) (A) and periplasmic fractions of twofold-diluted E. coli MG1655 (diamonds), twofold-diluted MG1655 (pAA410) (triangles) and undiluted MG1655 ivy::Kan (crosses) (B). The control sample (full line) consisted of phosphate buffer instead of lysozyme solution added to M. lysodeikticus. The differences in initial optical density between A and B was caused by the use of different batches of M. lysodeikticus.

These results indicated that knock-out and overexpres­sion of Ivy had been successful and, therefore, the pro­duction of Ivy protein in the three strains was further studied in periplasmic cell fractions, because Ivy is a periplasmic protein [27]. Figure 1B shows the progres­sion of M. lysodeikticus lysis by lysozyme in the presence of these periplasmic fractions. The rate of cell lysis by lysozyme (0.58 _OD600/h) was not reduced by the periplasmic fraction from the Ivy knock-out strain (0.59 ^OD600^/h), was reduced by 41% by the twofold-diluted wild-type periplasmic extract (0.34 _OD600/h) and com­pletely inhibited by the twofold-diluted periplasmic frac­tion from the Ivy overexpressor strain and by the undi­luted sample from the wild-type (latter data not shown). Lysozyme inhibitory activity of wild-type and overex­pressor strain periplasmic extracts became undetectable upon 4-fold and 27-fold dilution respectively, indicating that the overexpressor strain contained about 7-fold­higher levels of Ivy in its periplasm. These results were confirmed by a silver-stained SDS-PAGE gel (fig. 2), where a band corresponding to the molecular weight of Ivy (15.0 kDa) was present in the periplasmic fractions of MG1655 wild-type and of the Ivy overexpressor MG1655 (pAA410) (highest intensity in the latter), while the same band was completely lacking in Ivy knock-out strain MG1655 ivy::Kan. The same pattern was observed on Coomassie-stained gels, but the Ivy band in the wild-type was only very weak (data not shown). Tandem mass spec­trometry analysis identified the 15-kDa band as mature Ivy protein with high confidence by finding 92.2% amino acid sequence coverage. The band of 31.2 kDa, which is only present in MG1655 (pAA410) (fig. 2), most likely corresponds to b-lactamase encoded by the bla gene of the pAA410 plasmid, since the molecular weight matches and the same band was also seen in periplasmic cell fractions of E. coli MG1655 containing a plasmid with the same bla gene but lacking the ivy gene and the arabinose promoter (data not shown).

Figure 2. Silver-stained SDS-PAGE gel of periplasmic fractions of E. coli MG1655 (lane 1), Ivy overexpressor strain MG1655 (pAA410) (lane 2) and MG1655 ivy::Kan (lane 3). The position of Ivy is indicated by an arrow and the positions of the molecular­weight markers by lines.

Figure 3. Inactivation (log₁₀N₀/N) of E. coli MG1655, MG1655 (pAA410) and MG1655 ivy::Kan at atmospheric pressure (3 h) in the presence of lysozyme (hatched bars), and under high pressure (250 MPa, 15 min) in the absence (black bars) and presence (open bars) of lysozyme (1.0 mg/ml). Mean values ± SD (error bars) are shown. Lysozyme treatments resulting in significant differences (p < 0.05) compared to the same treatments without lysozyme are marked with an asterisk.

Since no other changes in periplasmic protein profile were observed, and since knock-out or overexpression of Ivy did not affect the growth of the E. coli strains or their morphology in phase contrast microscopy (data not shown), we can reasonably assume that overexpression or knock-out of Ivy does not have a major impact on cell viability and physiology. Consequently, the constructed strains can be used to assess the role of Ivy in cellular re­sistance to lysozyme. Lysozyme up to 10 mg/ml did not inhibit the growth in LB broth of any of the strains in the absence of an outer-membrane-permeabilizing treatment (data not shown). Therefore, in the next experiments, we tested the lysozyme sensitivity of the strains using either hydrostatic pressure or lactoferrin to permeabilize the outer membrane.

Role of Ivy in protection of E. coli MG1655 against lysozyme under high pressure

Suspensions of early stationary phase wild-type, ivy knock-out and ivy overexpression strains that had been induced with arabinose were treated with and without 1.0 or 10.0 mg/ml lysozyme at ambient pressure (3 h at 20°C) and under high pressure (15 min, 250 MPa at 20°C), and survivors were enumerated. At ambient pressure, none of the strains was significantly sensitive to lysozyme, indi­cating that the outer membrane was an effective lysozyme barrier in all strains under these conditions. High pressure alone caused only a low level of inactivation for all the strains (0.5–0.9 log units), but for 1.0 mg/ml lysozyme, it induced lysozyme sensitivity in an Ivy-dependent man­ner (fig. 3): the Ivy knock-out strain was strongly sensi­tized (2.6 log extra inactivation compared to high pres­sure alone), the wild-type strain partly (1.4 log units extra inactivation) and the Ivy overexpressor strain not at all. However, at a level of 10.0 mg/ml lysozyme, all three strains became equally sensitive to lysozyme under high pressure (data not shown). These results showed that the sensitivity of E. coli to lysozyme depended on the periplasmic Ivy content of the cells.

Role of Ivy in protection of E. coli MG1655 against lysozyme in the presence of lactoferrin

Suspensions of early stationary phase wild-type, ivy knock-out and ivy overexpression strains that had been induced with arabinose were treated with 3.0 mg/ml lactoferrin and/or 50 mg/ml lysozyme, and survivors were enumerated (fig. 4). The levels of inactivation achieved in this experiment were lower than in the high-pressure ex­periment, probably because lactoferrin is a weaker outer­membrane permeabilizer than high pressure, at least for our E. coli strains. That is also the reason why a higher lysozyme concentration was chosen in this experiment. Nevertheless, sensitization to lysozyme by lactoferrin was again Ivy dependent: strong sensitization occurred in the Ivy knock-out strain, intermediate sensitization in the wild-type strain and no sensitization in the Ivy over­expressor. Treatments with either of the antimicrobial proteins alone had little effect on any of the strains (1.7 < No/N < 3.7).

Contribution of Ivy in lysozyme resistance of E. coli LMM1010

E. coli LMM1010 is a mutant of E. coli MG1655 that was originally selected for high-pressure resistance [22], but which was later found to also have increased lysozyme re­sistance under high pressure [25], unlike two other pres­sure-resistant mutants that were isolated with the same approach. The finding that Ivy affects lysozyme resis­tance in E. coli now raised the question whether the lysozyme resistance of strain LMM1010 could be due to overexpression of Ivy. This was investigated by subject­ing E. coli LMM1010 and its ivy knock-out mutant to treatments with high pressure and lysozyme (fig. 5), and with lactoferrin and lysozyme (fig. 6), in the same way as described above for MG1655 and its derivative strains. Strain LMM1010 was more resistant to high pressure than MG1655 and was not sensitized to lysozyme under high pressure as expected, even at 600 MPa. However, knock-out of Ivy did not render LMM1010 sensitive to lysozyme at 250 or 600 MPa. Similar results were ob­tained with lactoferrin as a permeabilizer: neither the parental strain LMM1010 nor its Ivy knock-out deriva­tive were sensitized to lysozyme by lactoferrin.

Figure 4. Inactivation (No/N) of E. coli MG1655, MG1655 ivy::Kan and MG1655 (pAA410) after 7 h of incubation with 50 mg/ml lysozyme (hatched bars), 3.0 mg/ml lactoferrin (black bars) and lysozyme and lactoferrin together (open bars). Mean val­ues ± SD (error bars) are shown. Lysozyme treatments resulting in significant differences (p < 0.05) compared to the same treatments without lysozyme are marked with an asterisk.

Figure 5. Inactivation (log₁₀N₀/N) of E. coli LMM1010 and LMM1010 ivy::Kan in the presence of 1.0 mg/ml lysozyme at, from left to right, atmospheric pressure (3 h), 250 MPa (15 min) in the absence and in the presence of lysozyme, and at 600 MPa (15 min) in the absence and in the presence of lysozyme. Mean values ± SD (error bars) are shown. No significant differences (p > 0.05) were observed between treatments with and without lysozyme.

Figure 6. Inactivation (N_o/N) of E. coli LMM1010 and LMM1010 ivy::Kan after 7 h of incubation with 50 mg/ml lysozyme (hatched bars), 3.0 mg/ml lactoferrin (black bars) and both lysozyme and lactoferrin (open bars). Mean values ± SD (error bars) are shown. No significant differences (p > 0.05) were observed between treat­ments with and without lysozyme.

Discussion

The discovery that the E. coli orphan gene ykfE encodes a strong and specific lysozyme inhibitor [27] has raised speculation about the cellular and/or ecological role of this protein. In our study, we investigated the possible role of Ivy in the defense of E. coli against exogenous lysozyme. Since lysozymes are important components of the antibacterial defense systems of many plants and ani­mals, a lysozyme inhibitor could enhance the survival of commensal or pathogenic bacteria in their host. If this were indeed the case, one would expect to find lysozyme inhibitors to be widely distributed in bacteria. However, a BLAST search in the GenBank (http://www.ncbi.nlm. nih.gov/) and TIGR (http://www.tigr.org/) databases re­vealed putative Ivy orthologs only in a few other Gram­negative bacteria: P. aeruginosa PAO1 (two sequences with 47 and 44% similarity), Burkholderia mallei (45% similarity), Yersinia pestis KIM and CO92 (both 51% similarity), Shigella flexneri 2a (99% similarity), Chro­mobacterium violaceum ATCC12472 (44% similarity) and E. coli 0157:H7 and CFT073 (both 100% similarity). On the other hand, the limited distribution of Ivy does not necessarily imply that less well conserved lysozyme in­hibitors are absent in other bacteria. Secreted proteina­ceous lysozyme inhibitors, different from Ivy in molecu­lar weight or composition, have indeed been reported ear­lier, but they have been poorly characterized: a putative lipoprotein was reported in Bacillus subtilis [38, 39], two proteins in P. aeruginosa [40] and a glycoprotein in En­terobacter cloacae [41]. We have constructed an E. coli MG1655 derivative that overexpresses Ivy from an arabi­nose-inducible promoter and one in which the ivy gene was knocked out by insertion of a Kan resistance cassette. These strains overproduced Ivy or failed to produce it, re­spectively, as shown by the inhibition of lysozyme activ­ity by cellular extracts (fig. 1A) and by periplasmic frac­tions (fig. 1B) and by the levels of Ivy in periplasmic fractions (fig. 2), but their growth was otherwise unaf­fected (data not shown).

Using these Ivy knock-out and overexpressor strains, we were able to demonstrate that Ivy, at wild-type level, pro­vides some protection against certain levels of lysozyme when the outer membrane is permeabilized by high pres­sure or by lactoferrin, and that an increased or decreased production of Ivy resulted in an increased and decreased lysozyme resistance, respectively (figs 3, 4). Although our study was conducted with hen egg white lysozyme, human lysozyme belongs to the same structural class of lysozymes called C-type lysozymes and was previously shown also to be inhibited by Ivy [27]. Therefore, our findings suggest the existence of a novel interaction mechanism that may play a role in many bacteria-host in­teractions. Although the functionality of Ivy in a real in vivo situation where E. coli interacts with a vertebrate host is not yet proven, the concentrations of lysozyme that we used (1 and 50 mg/ml for the high-pressure and lacto­ferrin experiments, respectively) were within the range of naturally occurring concentrations. For example, the con­centration of human lysozyme ranges from 0.77 mg/ml in mixed respiratory secretions to 60 mg/ml in preterm hu­man colostrum [42, 43]. Furthermore, although we had to use a higher than natural concentration of lactoferrin to accomplish outer membrane permeabilization, in vivo, multiple components are known to act together in a syn­ergistic manner, such as lactoferrin, bactericidal perme­ability-inducing protein, cathelicidins and defensins [44]. With respect to the pathogens in which ivy orthologs have been identified, of interest will be to investigate whether these orthologs are functional and contribute to viru­lence.

Several lines of evidence suggest that lysozyme possesses a bactericidal mode of action that is independent of its en­zymatic activity and that does not cause cell lysis. This non-enzymatic inactivation mechanism has been attributed to cytoplasmic membrane destabilization due to the cationic properties of lysozyme, resulting in cell leakage, or to the induction of autolysins [45–47]. Therefore, an interesting question to be resolved in future work is whether this non-enzymatic mode of action is also inhibited by Ivy. Although our results provide evidence for a role of Ivy in the lysozyme resistance of E. coli, other functions of Ivy cannot be ruled out at this point. One possible function could be the regulation of autolysin ac­tivity, because some bacterial autolysins have muramidase activity, and because some E. coli autolysins such as the lytic transglycosylases Slt70, MltC and MltD are structurally related to type C lysozymes [48]. However, in view of the important role of autolysins in cellular growth and division, one would expect loss of this control func­tion by knock-out of the ivy gene to have a major effect on cell growth or morphology, and this was not the case (data not shown). On the other hand, if control of au­tolysin function is so important, the cell may dispose of redundant inhibitor systems, and this could explain the lack of a phenotype related to cell growth or division. We also demonstrated that the lysozyme resistance of an MG1655 mutant with enhanced resistance to lysozyme under high pressure [25] is not due to overproduction of Ivy, because an Ivy knock-out derivative of this strain did not become lysozyme sensitive. This illustrates that Ivy is not the only factor that confers lysozyme resistance in E. coli. Alterations in outer-membrane composition, for example, may affect the passage of lysozyme through this barrier, and thus lysozyme sensitivity. This was demonstrated earlier for Salmonella Typhimurium rough and deep-rough mutants under high pressure [26].

In conclusion, the finding that Ivy confers lysozyme resistance in E. coli may be of broad ecological importance because it highlights Ivy, and lysozyme inhibitors in general, as a totally novel type of factor that can contribute to the fitness of bacterial populations when they are chal­lenged with lysozymes produced by plants, animals or perhaps even bacteriophages. This finding will also be relevant in the development of applications that make use of the antibacterial properties of lysozyme.

Acknowledgements.

B. M. and A. M. have postdoctoral fellowships from the Research Fund K. U. Leuven, and D. D. and A. A., doctoral fellowships from the Flemish Institute for the Promotion of Scien­tific Technological Research (IWT). This research was conducted in the framework of research project GOA/03/10 financed by the Re­search Fund K. U. Leuven. The authors wish to thank C. Abergel and J. M. Claverie for the Ivy knock-out E. coli TE2680 strain, J. Robben from the Biomedical Research Institute (BIOMED) of the Limburgs Universitair Centrum for the mass spectrometry analysis and B. De Ketelaere (Laboratory for Agricultural Machinery and Processing, K.U.Leuven) for the statistical analysis.

 REFERENCES
 

1 Beintema J. J. and Terwisscha van Scheltinga A. C. (1996) Plant lysozymes. In: Lysozymes: Model Enzyme in Biochemistry and Biology, pp. 75–86. Joll6s P. (ed.), Birkhauser, Basel

2 Ellison R. T. and Giehl T. J. (1991) Killing of gram-negative bacteria by lactoferrin and lysozyme. J. Clin. Invest. 88: 1080–1091

3 Hancock R. E. W. and Scott M. G. (2000) The role of antimi­crobial peptides in animal defense. Proc. Natl. Acad. Sci. USA 97: 8856–8861

4 Ohno N. and Morrison D. C. (1989) Lipopolysaccharide inter­actions with lysozyme differentially affect lipopolysaccharide immunostimulatory activity. Eur. J. Biochem. 186: 629–636

5 Taylor P. W. (1983) Bactericidal and bacteriolytic activity of serum against gram-negative bacteria. Microbiol. Rev. 47: 46– 83

6 Rydman P. S. and Bamford D. H. (2002) Phage enzymes digest peptidoglycan to deliver DNA. ASM News 68: 330–335

7 Young R. (1992) Bacteriophage lysis: mechanism and regulation. Microbiol. Rev. 56: 430–481

8 H6ltje J.-V. (1995) From growth to autolysis: the murein hydro­lases of Escherichia coli. Arch. Microbiol. 164: 243–254

9 Shockman G. D. and H6ltje J.-V. (1994) Microbial peptidogly­can (murein) hydrolases. In: Bacterial Cell Wall: New Compre­hensive Biochemistry, vol. 27, pp. 131–160, Ghuysen J. M. and Hakenbeck R. (eds), Elsevier, Amsterdam

10 Losso J. N., Nakai S. and Charter E. A. (2000) Lysozyme. In: Natural Food Antimicrobial Systems, pp. 185–210, Naidu A. S. (ed.), CRC Press LLC, Boca Raton, Fla

11 Proctor V. A. and Cunningham F. E. (1988) The chemistry of lysozyme and its use as a food preservative and a pharmaceuti­cal. Crit. Rev. Food Sci. Nutr. 26: 359–395

12 Masschalck B. and Michiels C. M. (2003) Antimicrobial prop­erties of lysozyme in relation to foodborne vegetative bacteria. Crit. Rev. Microbiol. 29: 191–214

13 Hancock R. E. W. (1984) Alterations in outer membrane per­meability. Annu. Rev. Microbiol. 38: 237–294

14 Hancock R. E. W. (1997) Peptide antibiotics. Lancet. 349: 418– 422

15 Ibrahim H. R., Kato A. and Kobayashi K. (1991) Antimicrobial effects of lysozyme against gram-negative bacteria due to co­valent binding of palmitic acid. J. Agric. Food Chem. 39: 2077– 2082

16 Ibrahim H. R., Kobayashi K. and Kato A. (1993) Length of hy­drocarbon chain and antimicrobial action to gram-negative bac­teria of fatty acylated lysozyme. J. Agric. Food Chem. 41: 1164– 1168

17 Arima H., Ibrahim H. R., Kinoshita T. and Kato A. (1997) Bac­tericidal action of lysozymes attached with various sizes of hy­drophobic peptides to the C-terminal using genetic modifica­tion. FEBS Lett. 415: 114–118

18 Facon M. J. and Skura B. J. (1996) Antibacterial activity of lactoferricin, lysozyme and EDTA against Salmonella enteri­tidis. Int. Dairy J. 6: 303–313

19 Payne K. D., Oliver S. P. and Davidson P. M. (1994). Compari­son of EDTA and Apo-lactoferrin with lysozyme on the growth of foodborne pathogenic and spoilage bacteria. J. Food Prot. 57: 62–65

20 Vaara M. (1992) Agents that increase the permeability of the outer membrane. Microbiol. Rev. 56: 395–411

21 Garcia-Graells C., Masschalck B. and Michiels C. (1999) Inac­tivation of Escherichia coli in milk by high hydrostatic pressure treatment in combination with antimicrobial peptides. J. Food Prot. 62: 1248–1254

22 Hauben K., Wuytack E., Soontjens C. and Michiels C. (1996) High pressure transient sensitization of Escherichia coli to lysozyme and nisin by disruption of outer-membrane perme­ability. J. Food Prot. 59: 350–355

23 Masschalck B., Van Houdt R. and Michiels C. W. (2001) High pressure increases bacteriricidal activity and spectrum of lacto­ferrin, lactoferricin and nisin. Int. J. Food Microbiol. 64: 325– 332

24 Hauben K., Barlett D., Soontjes C., Cornelis K., Wuytack E. Y. and Michiels C. W. (1997) Escherichia coli mutants resistant to inactivation by high hydrostatic pressure. Appl. Environ. Mi­crobiol. 63: 945–955

25 Masschalck B., Garcia-Graells C., Van Haver E. and Michiels C. W. (2000). Inactivation of high pressure resistant Esch­erichia coli by lysozyme and nisin under high pressure. Inno­vat. Food Sci. New Emerg. Technol. 1: 39–47

26 Masschalck B., Deckers D. and Michiels C. W. (2003) Sensiti­zation of outer membrane mutants of Salmonella typhimurium and Pseudomonas aeruginosa to antimicrobial peptides under high pressure. J. Food Prot. 66: 1360–1367

27 Monchois V., Abergel C., Sturgis J., Jeudy S. and Claverie J. M. (2001) Escherichia coli ykfE ORFan gene encodes a potent in­hibitor of C-type lysozyme. J. Biol. Chem. 276: 18437–18441

28 Guzman L.-Z., Belin D., Carson M. J. and Beckwith J. (1995) Tight regulation, modulation, and high-level expression by vec­tors containing the arabinose _PBAD promoter. J. Bacteriol. 177: 4121–4130

29 Valdivia R. H. and Falkow S. (1996) Bacterial genetics by flow cytometry: rapid isolation of Salmonella typhimurium acid-in­ducible promoters by differential fluorescence induction. Mol. Microbiol. 22: 367–378

30 Datsenko K. A. and Wanner B. L. (2000) One-step inactivation of chromosomal genes in Escherichia coli K-12 using PCR products. Proc. Natl. Acad. Sci. USA 97: 6640–6645

31 Miller J. H. (1992) A Short Course in Bacterial Genetics, Cold
Spring Harbor Laboratory Press, Cold Spring Harbor, N. Y.

32 Weisner B. (1984) Lysozyme. In: Methods of Enzymatic Analysis, vol. 4, pp. 189–194, Bergmayer H. R. (ed.), Verlag Chemie, Weinheim

33 Leammli M. (1970) Cleavage of structural protein during the assembly of the head of bacteriophage T4. Nature 227: 680– 685

34 Heukeshoven J. and Dernick R. (1988) Improved silver staining procedure for fast staining in PhastSystem Development Unit. I. Staining of sodium dodecyl sulfate gels. Electrophoresis 9: 28–32

35 Shevchenko A., Wilm M., Vorm O. and Mann M. (1996) Mass spectrometric sequencing of proteins from silver stained poly­acrylamide gels. Anal. Chem. 68: 850–858

36 Kitamura Y. and Itoh T. (1987) Reaction volume of protonic ionization for buffering agents: prediction of pressure depen­dence of pH and pOH. J. Solut. Chem. 16: 715–725

37 Neter J., Kutner M. H., Nachtsheim C. J. and Wasserman W. (1996) Applied Linear Statistical Models, Irwin, Boston

38 Murao S., Kato M., Wang S.-L., Hoshino M. and Arai M. (1990) Isolation and characterization of a novel hen egg white lysozyme inhibitor from Bacillus subtilis I-139. Agric. Biol. Chem. 54: 1129–1136

39 Wang S.-L., Murao S. and Arai M. (1990) Some properties of hen egg white lysozyme inhibitor from Bacillus subtilis I-139. Agric. Biol. Chem. 54: 2447–2448

40 Wang S.-L., Shieh S.-T. and Pai C.-S. (1995) Production, pu­rification and characterization of two proteinaceous hen-egg­white lysozyme inhibitors from Pseudomonas aeruginosa M-1001. Proc. Natl. Sci. Council B Life Sci. 19: 166–175

41 Wang S.-L., Pai C.-S. and Sieh S.-T. (1997) Production, purifi­cation and characterization of the hen egg-white lysozyme inhibitor from Enterobacter cloacae M-1002. J. Chin. Chem. Soc. 44: 349–355 

42 Thompson A. B., Bohling T., Payvandi F. and Rennard S. I. (1990) Lower respiratory tract lactoferrin and lysozyme arise primarily in the airways and are elevated in association with chronic bronchitis. J. Lab. Clin. Med. 115: 148–158

43 Marthur N. B., Dwarkadas A. M., Sharma V. K., Saha K. and Jain N. (1990) Anti-infective factors in preterm human colostrum. Acta Paediatr. Scand. 79: 1039–1044

44 Ganz T. and Weiss J. (1997) Antimicrobial peptides of phago­cytes and epithelia. Semin. Hematol. 34: 343–354

45 Laible N. J. and Germaine G. R. (1985) Bactericidal activity of human lysozyme, muramidase-inactive lysozyme, and cationic polypeptides against Streptococcus sanguis and Streptococcus faecalis: inhibition by chitin oligosaccharides. Infect. Immun. 48: 720–728

46 Masschalck B., Deckers D. and Michiels C. W. (2002) Lytic and non-lytic mechanism of inactivation of gram-positive bacteria by lysozyme under atmospheric and high hydrostatic pressure. J. Food Prot. 65: 1916–1923

47 Ibrahim H. R., Higashiguchi S., Sugimoto Y. and Aoki T. (1997) Role of divalent cations in the novel bactericidal activity of the partially unfolded lysozyme. J. Agric. Food Chem. 45: 89–94

48 Dijkstra A. J. (1997) The Lytic Transglycosylase Family of Escherichia coli, Ph. D. thesis, University of Groningen, The Netherlands

(order Full Text from publisher)

© 2005 Transgalactic Ltd (manufacturer of Bioscreen C software) | Privacy Statement | P.O. Box 1393, 00101 Helsinki, Finland, phone: +358 9 85172920, fax: +358 9 8749481, e-mail: microbiology@bionewsonline.com
 

Last modified: May 25, 2005