HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

J. Biol. Chem., Vol. 277, Issue 49, 47636-47644, December 6, 2002

This Article Abstract Full Text (PDF) All Versions of this Article: 277/49/47636    most recent M205104200v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Koprivnjak, T. Articles by Weiss, J. P. Search for Related Content PubMed PubMed Citation Articles by Koprivnjak, T. Articles by Weiss, J. P. Social Bookmarking                      What's this?

Role of Charge Properties of Bacterial Envelope in Bactericidal Action of Human Group IIA Phospholipase A₂ against Staphylococcus aureus^*

Tomaz Koprivnjak^§, Andreas Peschel^¶, Michael H. Gelb, Ning S. Liang^**, and Jerrold P. Weiss^§§§

From the Departments of  Microbiology,  Internal Medicine, and the ^§ Inflammation Program, University of Iowa, Iowa City Veterans Affairs Medical Center, Iowa City, Iowa 52246, the ^¶ Mikrobielle Genetik, Universität Tübingen, Auf der Morgenstelle 28, 72076 Tübingen, Germany, the  Departments of Chemistry and Biochemistry, University of Washington, Seattle, Washington 98195-1700, and the ^** Department of Pharmacy, Guangxi Cancer Hospital/Institute, Nanning City, Guangxi Province 530021, Peoples Republic of China

Received for publication, May 23, 2002, and in revised form, September 12, 2002

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Mammalian Group IIA phospholipases A₂ (PLA₂) potently kill Staphylococcus aureus. Highly cationic properties of these PLA₂ are important for Ca²⁺-independent binding and cell wall penetration, prerequisites for Ca²⁺-dependent degradation of membrane phospholipids and bacterial killing. To further delineate charge properties of the bacterial envelope important in Group IIA PLA₂ action against S. aureus, we examined the effects of mutations that prevent specific modifications of cell wall (dltA) and cell membrane (mprF) polyanions. In comparison to the parent strain, isogenic dltA bacteria are ~30-100× more sensitive to PLA₂, whereas mprF bacteria are <3-fold more sensitive. Differences in PLA₂ sensitivity of intact bacteria reflect differences in cell wall, not cell membrane, properties since protoplasts from all three strains are equally sensitive to PLA₂. A diminished positive charge in PLA₂ reduces PLA₂ binding and antibacterial activity. In contrast, diminished cell wall negative charge by substitution of (lipo)teichoic acids with D-alanine reduces antibacterial activity of bound PLA₂, but not initial PLA₂ binding. Therefore, the potent antistaphylococcal activity of Group IIA PLA₂ depends on cationic properties of the enzyme that promote binding to the cell wall, and polyanionic properties of cell wall (lipo)teichoic acids that promote attack of membrane phospholipids by bound PLA₂.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

A wide variety of mammalian peptides and polypeptides has been described that can kill staphylococci and many other Gram-positive bacteria. Of these, possibly the most potent is the 14-kDa secretory group IIA phospholipase A₂ (PLA₂)¹ (1). Levels of this enzyme sufficient for significant antibacterial action are found at a variety of sites including tears and seminal and many other inflammatory fluids, as well as within granules of certain leukocytes and platelets (2-7). The Group IIA PLA₂ can act independently and also in concert with other host defense systems (1). It has been shown that the bactericidal activity of inflammatory fluids against Staphylococcus aureus is due mostly to mobilized Group IIA PLA₂ (5, 8) and that killing of encapsulated S. aureus by whole rabbit peritoneal inflammatory exudates is also dependent on PLA₂.² In addition, transgenic mice with overexpressed PLA₂ show much greater resistance to S. aureus than their non-transgenic littermates, implying a protective role of Group IIA PLA₂ against S. aureus infections in vivo (9, 10).

The ability of Group IIA PLA₂ to attack S. aureus and other Gram-positive bacteria reflects the ability of the enzyme to bind to and penetrate the cell wall to gain access to phospholipids (PL) in the cell membrane. At least four steps are involved: binding of the enzyme to the bacterial cell surface, penetration of the enzyme through peptidoglycan layers, degradation of PL in the cell membrane, and activation of bacterial autolysins (11). The phospholipolytic activity of PLA₂ and, hence, ultimate bacterial killing requires calcium as a cofactor (8), but initial PLA₂ binding to the cell surface does not.³

Initial binding of PLA₂ to the cell surface of S. aureus involves electrostatic interactions³ between PLA₂ and the bacterial cell surface. Among the more than 100 structurally related low molecular weight (~14 kDa) PLA₂ that have been characterized to date, the mammalian Group IIA PLA₂ are unique in their very high net positive charge ranging from +12 to +17. This very high net basicity is essential for the enzyme's potent bactericidal activity toward Gram-positive bacteria, principally by promoting initial interactions and penetration of the cell wall (12-14). ³ In contrast, highly cationic properties of PLA₂ are not essential for calcium-dependent catalytic activity (12) nor for the ability of structurally related 14-kDa PLA₂ to degrade PL in cell wall-depleted bacterial protoplasts (15, 16).³ Specific bacterial binding sites of PLA₂ are not known, but probably involve cell envelope anionic moieties. Major cell envelope-associated polyanions in S. aureus include the cell wall lipoteichoic and teichoic acids (LTA and TA, Ref. 17) and cell membrane PL (phosphatidylglycerol, PG and cardiolipin, CL; Ref. 17). The net charge on these molecules may be regulated by substitution of repeating glycerol (alditol)-phosphate moieties with D-alanine in LTA and TA (18, 19) or with L-lysine in PG (20). These substitutions depend on the dlt operon and mprF gene, respectively (19-21). The susceptibility of S. aureus to many small cationic antibacterial compounds is similarly increased in dltA or mprF mutants, indicating that both cell wall and cell membrane polyanions are important determinants of bacterial sensitivity to these antibacterial compounds (19, 20). Because the cationic properties of PLA₂ seem important in the interactions of the enzyme with the bacterial cell wall but not the cell membrane, we hypothesized that D-alanylation of cell wall LTA and TA would have a greater effect on bacterial susceptibility to PLA₂ than mprF-dependent modification of the cell membrane PG. The results presented confirm this scenario. We have made use of site-specific mutagenesis of the human Group IIA PLA₂ to further demonstrate that the higher susceptibility of the dltA strain does not depend on either the net charge of the PLA₂ or higher initial binding of the enzyme to the bacterial surface of dltA bacteria. Instead, the absence of D-alanine modification of LTA and TA probably results in pleitropic modifications of the staphylococcal cell wall that promote Ca²⁺-dependent attack of bacterial PL by bound PLA₂.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Materials-- [Ala^8,13,18]Magainin II amide, lysostaphin (3000 units/mg of protein), DNase I (2500 Kunitz units/mg of protein), and polyethylenimine (PEI) were purchased from Sigma Chemical Co. Human--defensin-3 (HBD-3) was a generous gift from Dr. Paul McCray (University of Iowa, Department of Pediatrics, Iowa City, IA). [1-¹⁴C]Oleic acid was purchased from PerkinElmer Life Sciences. IODO-GEN was obtained from Pierce Chemical Co., and HP-TLC plates from Merck. PhastGels were purchased from Amersham Biosciences, GF/C glass microfiber filters from Whatman International Ltd (Maidstone, England), and bovine serum albumin (BSA) from Roche Molecular Biochemicals. RPMI 1640 was obtained from Invitrogen and Hanks balanced salt solution with (HBSS⁺) or without (HBSS) Ca²⁺ and Mg²⁺ were obtained from Cellgro Mediatech Co. (Herndon, VA).

Bacterial Strains and Growth Conditions-- Strains of S. aureus used were the parent SA113 (22), isogenic mprF, dltA (AG1) mutants and their corresponding complemented strains (mprF pRB mprF and dltApRB dltA, respectively; Refs. 19 and 20). Bacteria were grown overnight at 37 °C in BM broth (1% tryptone, 0.5% yeast extract, 0.5% NaCl, 0.1% K₂HPO₄, 0.1% glucose), washed, and resuspended in fresh medium with a starting OD₅₅₀ of 0.05 and subcultured for 2-2.5 h until mid-log phase.

Recombinant Human PLA₂-- Recombinant wild type and mutant human secretory Group IIA PLA₂, R92E, K38E/K116E (16), and D49S³ were expressed in Escherichia coli and purified as described previously (12). Purity of PLA₂ was confirmed by SDS-PAGE and by reversed-phase HPLC (12). The enzymatic activity of [D49S]PLA₂ against autoclaved E. coli was ~0.001% of wt PLA₂.³

Assay of Bacterial Viability-- The effect of various antimicrobial proteins and peptides on bacterial viability was determined by measuring the ability of treated bacteria to form colonies on BM agar. Typical incubation mixtures contained 10⁶ or 10⁷ bacteria/ml in RPMI 1640 supplemented with 10 mM Hepes (pH 7.4), 1% (w/v) BSA, and 1 mM CaCl₂ ± indicated protein/peptide. In assays with [Ala^8,13,18]Magainin II amide and HBD-3, CaCl₂ was omitted, since these peptides are more sensitive to inhibition by added divalent cations than is PLA₂. Incubations were carried out at 37 °C for up to 2 h. At each time point, aliquots of bacterial suspensions were serially diluted in sterile physiological saline and plated in 5 ml of molten (50 °C) BM agar. Bacterial viability was measured by counting bacterial colonies (i.e. colony-forming units (CFU)) after 18-24 h incubation at 37 °C.

Radiolabeling of S. aureus Lipids during Bacterial Growth-- Bacterial PL were radiolableled during growth in subculture as previously described (11). Briefly, bacteria were subcultured at 37 °C to mid-logarithmic phase in BM supplemented with 1 µCi/ml of [1-¹⁴C]oleic acid and 0.01% BSA, washed, resuspended in half the volume of fresh BM medium without [1-¹⁴C]oleic acid, and incubated for another 20 min at 37 °C. BSA was then added to the medium to a final concentration of 0.5% (w/v), and bacteria were washed to remove remaining free oleic acid. Washed bacteria were resuspended to desired concentrations in incubation medium and used promptly.

Assay of Bacterial Phospholipid Degradation-- Samples consisted of 10⁷ bacteria/ml in RPMI 1640 supplemented with 10 mM Hepes (pH 7.4), 1% (w/v) BSA, and 1 mM CaCl₂ or in HBSS supplemented as indicated. PL degradation products generated during PLA₂ treatment were quantitatively recovered in the extracellular medium as complexes with BSA (11). Therefore accumulation of radioactive material in extracellular supernatants after sedimentation (14,000 rpm for 4 min) of bacterial suspensions was used to routinely monitor PL degradation. To confirm that released ¹⁴C-labeled material corresponded to PL degradation, lipids of S. aureus present in whole cell suspensions and extracellular supernatants, with and without PLA₂ incubation, were extracted using the method of Bligh and Dyer (23). Extracted lipids were dried under N₂ flow, dissolved in chloroform/methanol (2:1, v/v), and resolved by thin layer chromatography (TLC) in a chloroform/methanol/water/acetic acid (65:25:4:1, v/v/v) solvent system. Lipids were identified by comparison to migration of lipid standards and quantified by phosphorimage analysis using PhosphorImager and ImageQuant software (Amersham Biosciences, Molecular Dynamics Division). Intact bacterial PL and PL breakdown products (i.e. lyso-PL and free fatty acids) were quantitatively recovered in the bacterial pellet and extracellular medium, respectively. To maximize detection of lysyl-PG, extracts were collected and analyzed by TLC within 3 h of the end of the incubation. Storage of extracts for longer times before recovery of labeled lipids in the chloroform phase and TLC led to diminished recovery of lysyl-PG, but had no effect on overall recovery of intact PL and PL breakdown products. However, lyso-PG is incompletely recovered in the chloroform phase of lipid extracts because of its increased polarity.

Radioiodination of PLA₂-- PLA₂ were radiolabeled with ¹²⁵I in tubes precoated with IODO-GEN as described previously (12, 24). Radiolabeled PLA₂ was separated from free iodide by chromatography on Sephadex G25 (Amersham Biosciences) equilibrated with 0.2 M sodium acetate buffer, pH 4.0. Recovered radiolabeled material was more than 95% trichloroacetic acid-precipitable.

Assay of PLA₂ Binding to S. aureus-- Binding assays were performed with 10⁷ bacteria/ml in HBSS supplemented with 10 mM Hepes (pH 7.4) and 0.1% (w/v) BSA. To determine effects of envelope-associated Ca²⁺ and subsequent Ca²⁺-dependent envelope damage by PLA₂ on PLA₂ binding, parallel assays were carried out ± 2 mM MgCl₂/2 mM EGTA. Radiolabeled 0.5 ng of ¹²⁵I-PLA₂ (wild type or mutant PLA₂) was mixed with 5 ng of unlabeled wild type or D49S PLA₂ (1:10) before addition to the incubation mixture to ensure exposure of bacteria to similar destructive effects of PLA₂ when Ca²⁺ was present, irrespective of the source of ¹²⁵I-PLA₂. Incubations were carried out in a total volume of 250 µl at 37 °C for 30 min. Bound ¹²⁵I-PLA₂ was collected on Whatman glass microfiber GF/C filters that had been presoaked in 0.5% PEI/20 mM Tris-HCl, pH 7.4, for 15 min. Filters were washed five times with 20 mM Tris-HCl, pH 7.4. Radioactivity associated with filters was measured in a gamma counter (Cobra II^TM, Packard Biosciences Co.).

Production and Assay of S. aureus Protoplasts-- S. aureus was grown as described above until mid-log phase in medium containing [¹⁴C]oleate. Metabolically labeled bacteria were harvested and resuspended at a final concentration of 4 × 10⁹ bacteria/ml in Tris-buffered saline osmotically stabilized with 30% (w/v) raffinose. Lysostaphin and DNase I were added to a final concentration of 250 µg/ml and 500 µg/ml, respectively, and bacteria were incubated at a final concentration of 2 × 10⁹ bacteria/ml, shaking slowly at 37 °C for 1 h. Protoplasts were collected by centrifugation at 4,500 rpm for 15 min at room temperature. The supernatant was removed, and protoplasts were gently resuspended in RPMI 1640 containing 1% BSA, 10 mM Hepes (pH 7.4), 1 mM CaCl₂, and 30% raffinose in RPMI 1640. Greater than 80% of [¹⁴C]oleate-labeled bacterial PL of intact bacteria were present in the recovered protoplasts. The virtual absence of intact/viable bacteria in the recovered protoplast samples was confirmed by Gram staining and assay of CFU without raffinose.

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Effect of dltA- and mprF-dependent Modifications of Cell Envelope Polyanions on Sensitivity of S. aureus to Human Group IIA PLA₂-- It has been previously shown that dltA and mprF mutants of S. aureus are each ~10-30-fold more susceptible to killing by small cationic antimicrobial peptides (CAP) than the parent dltA⁺/mprF⁺ S. aureus strain (19, 20). When comparing effects of the Group IIA PLA₂ against wild type (wt) and the two mutant dltA and mprF strains, we obtained different results. Whereas dltA bacteria were ~30-100-fold more sensitive (Fig. 1, A, B, D, and I), mprF bacteria were 3-fold more susceptible than wt S. aureus to PLA₂ killing (Fig. 1, A, C, D, and I). In contrast, under the same test conditions, both mutant strains were ~5× more sensitive to the small CAP [Ala^8,13,18]Magainin II amide and HBD-3 (Fig. 1, G and H). Higher susceptibility of the mutant S. aureus strains to PLA₂ was reversed by complementation with plasmid-bearing wild type dltA or mprF genes (Fig. 1, E and F). These findings demonstrate that dltA-dependent modifications of wall LTA and TA are much more important in resistance of S. aureus to PLA₂ than mprF-dependent modification of cell membrane PG.

View larger version (38K): [in this window] [in a new window]   Fig. 1.   Sensitivity of S. aureus strains to killing by human Group IIA PLA2, [Ala8,13,18]Magainin II amide, and human--defensin-3. Incubations contained either 106 bacteria/ml (G and H) or 107 bacteria/ml (A-F and I) with or without PLA2 (A-F and I), [Ala8,13,18]Magainin II amide (G), or human--defensin-3 (HBD-3) (H). All incubation mixtures contained RPMI 1640 supplemented with 10 mM Hepes (pH 7.4). PLA2 assays were also supplemented with 1% BSA and 1 mM CaCl2 whereas assays with [Ala8,13,18]Magainin II amide and HBD-3 contained 0.1% BSA and no additional Ca2+. After various times at 37 °C as indicated, bacterial viability was measured as CFU in BM agar. A-C, E, and F show effect of increasing concentration of PLA2: broken line represents no added PLA2; solid lines correspond to PLA2-treated samples. The increasing size of symbols corresponds to increasing PLA2 concentrations (30, 300, 1000 ng/ml). D shows the dose-dependent effects of PLA2 toward the five different strains during a 60-min incubation; the different symbols correspond to those used in A-C, E, and F. G-H show results after 120 min of incubation. The results shown represent the mean of three or more experiments ± S.E.

Differences in Killing of wt, dltA, and mprF S. aureus by PLA₂ Parallel Differences in Phospholipid Degradation-- Killing of S. aureus by PLA₂ depends on rapid degradation of membrane PL due to its catalytic activity and on activation of bacterial autolysins secondary to loss of membrane PL (25). To test whether differences in killing by PLA₂ of the three S. aureus strains paralleled differences in PL degradation, bacterial lipids were prelabeled during growth with [¹⁴C]oleic acid, and extracellular accumulation of radioactive lipid degradation products during PLA₂ treatment was measured. Dose-dependent accumulation of extracellular radioactive products during PLA₂ treatment was greatest in dltA but similar in mprF and wt strains, just as was observed in the killing experiments (compare Figs. 1 and 2). Release of [¹⁴C]lipids in untreated samples (Fig. 2, dashed lines) was also greatest in the dltA strain compared with the wt and mprF strains, suggesting greater instability of cell envelope lipids in the dltA S. aureus strain even in the absence of PLA₂ treatment. Increased sensitivity to PLA₂ was reversed by complementation of the mutant strains.

View larger version (37K): [in this window] [in a new window]   Fig. 2.   Accumulation of radioactive lipids in extracellular medium during PLA2 treatment of S. aureus strains. [14C]Oleate-labeled S. aureus (107 bacteria/ml) harvested from mid-logarithmic phase were incubated in RPMI 1640 supplemented with 1 mM CaCl2, 10 mM Hepes (pH 7.4), and 1% BSA in the presence or absence of PLA2 at 37 °C. At indicated times, samples were removed to measure accumulation of 14C-labeled lipids in extracellular medium as described under "Experimental Procedures." A-C, E, and F show effects of increasing concentrations of PLA2 (increasing sizes of symbols represent increasing concentration of PLA2; 30, 300, 1000 ng/ml). D represents the dose-dependent effect of PLA2 toward the five different strains during 30 min of incubation where differences were the greatest. The results represent the mean of three or more experiments ± S.E.

To verify that release of ¹⁴C-labeled lipids during PLA₂ treatment corresponds to bacterial PL breakdown radiolabeled bacterial lipids in whole suspensions and extracellular supernatants of PLA₂-treated bacteria were extracted and analyzed by TLC (Fig. 3). No significant differences were observed between the profiles of labeled lipids extracted from the untreated wt and dltA bacteria, whereas the labeled lipid profile of mprF bacteria revealed, as expected, the absence of lysyl-phosphatidylglycerol (lysyl-PG) (Fig. 3). The predominant species in all strains was PG (Fig. 3). After treatment with 30 ng/ml PLA₂, extracts of recovered supernatants contained mainly [¹⁴C]lyso-PG and to a lesser extent ¹⁴C-labeled free fatty acids. At this dose, PLA₂ caused almost complete loss of intact prelabeled PL, phosphatidic acid (PA), PG, and CL in the dltA strain during a 60-min incubation. In contrast, under the same conditions only a limited portion of the intact prelabeled PL species was degraded in mprF and wt bacteria (Fig. 3, whole cell suspensions). Therefore, PLA₂-dependent PL degradation paralleled differences in sensitivity to killing by PLA₂: dltA mprF wt. Major [¹⁴C]lipids in supernatants upon PLA₂ treatment were lyso-PL, i.e. lyso-PG. PLA₂-independent release of [¹⁴C]lipids was also greater in dltA compared with mprF and wt bacteria, but [¹⁴C]lipids in these supernatants were mainly intact lipids and free fatty acids (Fig. 3).

View larger version (95K): [in this window] [in a new window]   Fig. 3.   TLC analysis of total 14C-labeled lipids in S. aureus and released [14C]lipids in extracellular medium during PLA2 treatment of bacteria. Whole cell suspensions or extracellular supernatants recovered from [14C]oleate-labeled S. aureus (107 bacteria/ml) in RPMI 1640 supplemented with 10 mM Hepes (pH 7.4), 1 mM CaCl2, and 1% BSA ± 30 ng/ml PLA2 were extracted and analyzed by TLC as described under "Experimental Procedures." Major labeled lipids present before (time, 0 min) and after (time, 60 min) incubation ± PLA2 are indicated as determined by standards. Approximately the same amount of [14C]lipids (cpm) were applied per lane. Whole cell suspensions were extracted the same day to maximize recovery of lysyl-PG.

Cell Wall-depleted Membrane Protoplasts from wt, dltA, and mprF S. aureus Are Equally Sensitive to PLA₂-- Differences in PL degradation by PLA₂ of intact wt, dltA, and mprF S. aureus could reflect either greater access of the enzyme to the membrane of more sensitive (i.e. dltA mprF, wt) bacteria or greater sensitivity of PL within the membrane of these more sensitive bacteria. To distinguish between these two possibilities we prepared bacterial protoplasts from wt, dltA, and mprF bacteria and tested their susceptibility to PLA₂ hydrolytic activity as described under "Experimental Procedures." Recovery of bacterial membranes in the form of protoplasts was ~80%, as deduced from recovery of ¹⁴C-labeled material. Contamination with intact bacteria, as assessed by plating protoplasts on BM agar without 30% raffinose, was 0.001%. The radiolabeled lipid pattern of protoplasts prior to PLA₂ treatment resembled that of intact bacteria (compare Figs. 3 and 4, first lanes in each set). In contrast to intact bacteria, protoplasts of all three strains were equally sensitive to PLA₂-dependent PL degradation (Fig. 4). This indicates that neither dltA-dependent modification of cell wall TA and LTA nor mprF-modification of cell membrane PG affects the sensitivity of membrane PL to PLA₂, when the cell wall is removed. Comparison of the degradative activity of PLA₂ toward intact bacteria and protoplasts (Fig. 5) shows that the resistance conferred by the cell wall in wt and mprF bacteria is absent in the dltA strain.

View larger version (73K): [in this window] [in a new window]   Fig. 4.   Sensitivity of cell wall-depleted membrane protoplasts from wt, dltA, and mprF S. aureus to PLA2 degradation. Membrane protoplasts derived from 107 bacteria were incubated with increasing concentrations of PLA2 for 60 min at 37 °C in RPMI 1640, supplemented with 10 mM Hepes (pH 7.4), 1% BSA, 1 mM CaCl2, and osmotically stabilized with 30% raffinose. Lipids before (0') and after (60') incubation in the presence or absence of PLA2 were analyzed by TLC (upper panels). Densitometry analyses are shown in the lower panels. The amount of PLA2 used for the TLC analyses that are shown was 10 ng/ml. Note that appearance of lyso-PL+FFA is less than disappearance of PL from protoplasts because of incomplete recovery of lyso-PG in the chloroform phase of lipid extracts (see "Experimental Procedures").

View larger version (14K): [in this window] [in a new window]   Fig. 5.   Comparison of PLA2 activity toward protoplasts versus intact S. aureus from wt, dltA and mprF strains. Effects of increasing concentrations of PLA2 toward [14C]oleate-labeled bacteria (107/ml) and protoplasts (107 equivalents/ml) where measured as described in the legends to Figs. 2 and 4. Incubations were for 60 min at 37 °C. PL degradation in the protoplasts was monitored by disappearance of PL observed by TLC whereas PL degradation of bacteria was measured either as accumulation of 14C-labeled lipids in extracellular medium or disappearance of PL observed by TLC. All data are expressed as percent of total 14C-labeled lipids. Open symbols represent PL degradation in intact bacteria and closed symbols represent degradation of PL in protoplasts at different PLA2 concentrations.

Increased Sensitivity of dltA Bacteria Does Not Depend on Net Charge of PLA₂-- Previous studies indicated that dltA S. aureus are more sensitive to a wide range of cationic antibacterial compounds (19). To test if differences in sensitivity of wild type and dltA S. aureus are dependent on the net positive charge of the antibacterial compound, we compared the bactericidal potency of the wild type PLA₂ and two mutant PLA₂, R92E and K38E/K116E (16), against both S. aureus strains. As previously shown in other dltA⁺ strains of S. aureus³ (15), the bactericidal potency of R92E and K38E/K116E PLA₂ was much less (respectively, 10-fold and 100-fold reduced) than that of the wild type PLA₂ (Fig. 6, left panel). Nevertheless, the dltA mutant strain was ~30 times more susceptible to killing by each of the PLA₂. These results demonstrate that the absence of dltA-dependent cell wall modifications renders S. aureus more susceptible to killing, not only by wt PLA₂, but also by mutant enzymes with reduced net positive surface charge. Hence, within this range of net charge of the PLA₂ (+15  +11; (Lys+Arg)  (Glu+Asp)), effects of dltA-dependent modifications of the bacterial cell wall on bacterial sensitivity to PLA₂ are independent of the net charge of the PLA₂.

View larger version (17K): [in this window] [in a new window]   Fig. 6.   Sensitivity of wt and dltA bacteria to wt, R92E, and K38E/K116E PLA2. S. aureus (106/ml) in RPMI 1640, supplemented with 10 mM Hepes (pH 7.4), 1% BSA, and 1 mM CaCl2 were incubated with various doses of either wt PLA2, R92E, or K38E/K116E. After 120 min of incubation at 37 °C, bacterial viability was measured as CFU. The CFU of experimental samples is expressed as percent of the initial bacterial inoculum. The results shown represent the mean ± S.E. of three or more independent experiments.

Effect of PLA₂ Net Charge and dlt-dependent Modifications on PLA₂ Binding to the Bacterial Surface-- Because the absence of D-alanylation of LTA and TA is believed to increase the net negative charge of the bacterial surface (18, 19), the increased sensitivity of the dltA S. aureus to PLA₂ could simply reflect greater binding of PLA₂ to the dltA mutant strain. We therefore compared binding of ¹²⁵I-PLA₂ to the two strains. Initial PLA₂ binding was measured in Hepes-buffered HBSS supplemented with 2 mM MgCl₂/2 mM EGTA (Mg²⁺·EGTA) to chelate ambient and cell envelope-associated Ca²⁺ and thereby preclude Ca²⁺-dependent envelope disruption by added PLA₂ (Fig. 7A). Under these conditions, PLA₂ binding to S. aureus declined with reduced net charge of the PLA₂ (wt > R92E > K38E/K116E), but there was no significant difference in binding of wt or mutant PLA₂ to the parent and dltA strains. Similar results were obtained using 10⁷ or 10⁸ bacteria (not shown) except that PLA₂ binding was higher when more bacteria were present. In contrast, when envelope-associated Ca²⁺ was retained by carrying out incubations in HBSS without EGTA, PLA₂ binding was much greater to the dltA bacteria (Fig. 7B). This increase in PLA₂ binding when Ca²⁺ is present was seen with wt but not with a catalytically inactive mutant D49S PLA₂ (Fig. 7C) indicating that increased PLA₂ binding is secondary to Ca²⁺-dependent envelope damage (which is greater in the dltA strain; Fig. 2). These results demonstrate that Ca²⁺-independent initial binding depends on cationic properties of the PLA₂ (wt  R92E > K38E/K116E) but is not significantly different than the wild type and dltA S. aureus strains.

View larger version (21K): [in this window] [in a new window]   Fig. 7.   Effect of PLA2 net charge and dltA-dependent modifications on PLA2 binding to the bacterial surface in the presence or absence of Ca2+. 107 bacteria/ml (wt, dltA) were incubated in HBSS supplemented with 10 mM Hepes (pH 7.4), 0.1% BSA, with (A and C) and without (B and C) 2 mM MgCl2 and 2 mM EGTA (Mg2+·EGTA). 125I-labeled PLA2 (wt and mutants) and unlabeled PLA2 (wt, A and B; D49S, C) were added (1:10) as described under "Experimental Procedures," and incubation was carried out at 37 °C for 30 min. Bound 125I-labeled PLA2 was collected by membrane filtration and measured in a gamma-counter. The results are presented as percent of added 125I-PLA2 bound to the bacteria and represent the mean of 2-8 experiments ± S.E. Note that in B, only binding of wt PLA2 to wt and dltA S. aureus is shown.

Bound PLA₂ Is More Active against dltA S. aureus-- The fact that initial PLA₂ binding is not different from wt and dltA S. aureus implies that the activity of bound PLA₂ must be greater toward the dltA bacteria. To test this hypothesis more directly, we took advantage of the findings described above, namely that in the presence of Mg²⁺·EGTA, similar amounts of enzyme bound to the two strains. After an initial 30-min incubation of [¹⁴C]oleate-prelabeled bacteria with PLA₂ in the presence of Mg²⁺·EGTA, the bacteria were washed and resuspended in HBSS⁺ (i.e. HBSS supplemented with Ca²⁺ and Mg²⁺). As expected, in the presence of Mg²⁺·EGTA, PLA₂ was inactive, and there was little or no PLA₂-dependent accumulation of extracellular radioactive degradation products from either strain (compare left bars of Fig. 8, A and B). However, after washing the bacteria to remove unbound PLA₂ and subsequent resuspension and incubation in HBSS⁺ (containing Mg²⁺/Ca²⁺), PL degradation and killing (data not shown) was much greater in the mutant strain than in the parent S. aureus demonstrating much greater Ca²⁺-dependent activity of prebound PLA₂ toward dltA S. aureus (Fig. 8A, center). Under these experimental conditions, effects of prebound PLA₂ following Ca²⁺ addition nearly matched the activity of the enzyme against the two strains when incubations were carried out in medium containing Ca²⁺ throughout (Fig. 8A, right bars). In conclusion, bound PLA₂ displays greater Ca²⁺-dependent activity toward dltA S. aureus. This difference largely accounts for the greater sensitivity of dltA bacteria for this cationic antibacterial enzyme.

View larger version (27K): [in this window] [in a new window]   Fig. 8.   Bound PLA2 is more active against dltA S. aureus. S. aureus strains (107/ml) prelabeled with [14C]oleate were incubated at 37 °C with (A) or without (B) PLA2 (10 ng/ml), for 120 min in HBSS supplemented with 2 mM MgCl2 and 2 mM EGTA (Mg2+·EGTA) (left bars), in HBSS+ (right bars) or first for 30 min in HBSS supplemented with Mg2+·EGTA, washed to remove unbound PLA2 and incubated an additional 90 min in HBSS+ (center bars). All incubation mixtures also contained 10 mM Hepes (pH 7.4) and 0.1% BSA. At the end of the total incubation time (120 min), accumulated labeled lipids in the supernatant were measured as described before and expressed as percent of total labeled lipids present initially. The results shown represent the mean of three or more experiments ± S.E.

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

In the present study, we have shown, that in contrast to the wt and mprF strains, dltA S. aureus is 30-100-fold more sensitive to killing by human Group IIA PLA₂. Higher susceptibility of dltA S. aureus is independent of the net charge of the PLA₂ and is not due to greater initial binding of the enzyme to the bacterial surface, but rather is due to greater Ca²⁺-dependent activity of bound PLA₂.

It has been previously shown (19, 20) that dltA and mprF S. aureus mutants are each 10-30× more susceptible to CAP such as [Ala^8,13,18]Magainin II amide and human defensins, suggesting the importance of both cell wall modifications of LTA/TA and cell membrane modifications of PG in bacterial resistance to these cationic compounds. We have reproduced these findings although under our experimental conditions the dltA and mprF mutants are each only ~5× more sensitive than wt S. aureus to [Ala^8,13,18]Magainin II amide and HBD-3 (Fig. 1, G and H). In contrast, disruption of the dltA locus increases bacterial sensitivity to Group IIA PLA₂ nearly 100-fold whereas disruption of the mprF locus has much less effect (only ~3×; Fig. 1, D and I). Overall, these findings indicate that dltA-dependent modification of bacterial cell wall LTA and TA has a much greater effect on sensitivity of S. aureus to the antibacterial action of PLA₂ than do mprF-dependent modifications of the cell membrane PG. The contrasting effects of dltA- and mprF-dependent modifications of cell envelope polyanions on the action of PLA₂ and that of non-catalytic cationic antibacterial peptides suggest differences in the determinants of action of these antibacterial compounds. Such differences may have important biologic as well as mechanistic implications especially at anatomic sites (e.g. mucosal secretions, inflammatory exudates; 1, 2, 8, 27) where non-catalytic peptides and PLA₂ are jointly present and could act in synergy (5).⁴

The properties of Group IIA PLA₂ important in antibacterial action differ in many ways from those of CAP. The PLA₂ has enzymatic activity that is essential for its antibacterial activity and probably explains its much greater antibacterial potency against S. aureus and other Gram-positive bacteria as compared with CAP (Fig. 1, G-I and Refs. 5 and 8). In contrast to many CAP, the PLA₂ is fully active in incubation media that contain physiological extracellular salt concentrations, including divalent cations, (8, 26, 27). The reduced effects of dltA and mprF mutations on [Ala^8,13,18]Magainin II amide and HBD-3 activity in our experiments might in fact be due to the higher salt concentrations present in our incubation media. Higher salt concentration could decrease activity of salt-sensitive cationic peptides and function as counter ions to shield polyanions of the bacterial cell envelope, thereby reducing initial binding of cationic peptides to the bacterial cell wall.

The mechanism of antibacterial action of CAP and PLA₂ is also different. For both, the initial step is interaction between cationic groups in the peptide/polypeptide and anionic moieties in the bacterial cell wall, followed by penetration through the cell wall to reach the bacterial membrane. Subsequent interactions at the bacterial membrane, however, likely differ. Killing by CAP is generally believed to require membrane insertion of the peptide (28-31) whereas PL degradation and killing by PLA₂ can be mediated by protein bound to the extracytoplasmic surface of the membrane (32-34). Membrane insertion of CAP may require very extensive electrostatic interactions between the peptide and negatively charged membrane lipids to trigger an amphiphilic transition of the peptide and membrane insertion (for review see Ref. 35). In contrast, PLA₂ membrane interactions may be satisfied by more limited electrostatic interactions and not be affected by even marked variations in either enzyme or substrate/membrane interface net charge. The closely similar activity of natural and mutant PLA₂ charge variants against cell wall-depleted membranes of Gram-positive bacteria (Fig. 4 and Ref. 15) and the similar sensitivity to PLA₂ of membrane protoplasts derived from wt, dltA, and mprF S. aureus (Fig. 4) support this view. Once bound at the membrane interface, closely related PLA₂ also show limited substrate (head group) specificity (36). We observed no difference in degradation of PG and lysyl-PG by PLA₂ in wt and dltA bacteria/protoplasts (Figs. 3 and 4 and Ref. 25), and differences in membrane PG content between wt and mprF strains may be too limited to have more global effects on PLA₂ action. As the sensitivity of membrane protoplasts to PLA₂ is the same, whether or not lysyl-PG is present, even the modest effect of mprF disruption on bacterial PLA₂ sensitivity may not be due to changes in membrane properties but other, as yet unidentified, mprF-dependent bacterial alterations. The slightly greater resistance conferred by overexpression of mprF (Fig. 1F) is not associated with changes in membrane content of lysyl-PG. Lysyl-PG represents 16.7 and 17.2% of total ¹⁴C-membrane PL in mprF⁺ and plasmid-overexpressing strains, respectively (data not shown), implying that effects on PLA₂ (CAP?) sensitivity could be caused by other mprF-dependent bacterial alterations.

Previous studies have indicated that increased activity of CAP toward dltA S. aureus reflect increased binding of the cationic peptides to the cell wall LTA and TA not masked by covalently substituted D-alanine (19). In contrast, our findings indicate that increased PLA₂ activity toward these bacteria is not caused by increased binding but rather by increased Ca²⁺-dependent activity of the bound PLA₂. Whereas binding of PLA₂ was nearly the same to wt and dltA bacteria under calcium-free conditions (Fig. 7A), the antibacterial activity of the bound PLA_2, subsequently triggered by addition of Ca²⁺, was much greater in the mutant strain (Fig. 8A) demonstrating directly the greater activity of PLA₂ bound to dltA bacteria. PLA₂ binding to the mutant strain was much greater when binding was measured in the presence of the envelope associated Ca²⁺ (Fig. 7, B and C) but without appreciably increasing antibacterial activity (Fig. 8A) indicating that, under these conditions, increased binding did not significantly affect PLA₂ activity. In model membranes such as PL vesicles, the generation in situ of PL breakdown products can increase the avidity of PLA₂-membrane interactions (37). Therefore, the increased binding of PLA₂ to the mutant strain we observed when calcium was present, might simply reflect differences induced, directly or indirectly, by greater membrane damage within the envelope of the dltA bacteria. In support of this view are findings presented in Fig. 7C showing that increased (Ca²⁺-dependent) binding is only observed with wt and not with a catalytically inactive mutant (D49S) of PLA₂ despite similar Ca²⁺-independent binding of wt and D49S PLA₂. Thus, increased binding of PLA₂ to the mutant strain when Ca²⁺ is present is much more likely a secondary consequence of increased activity and not itself a determinant of greater activity.

We were able to distinguish the effects of changes of bacterial envelope structure on initial PLA₂ binding and activity of bound enzyme, because these events can be completely separated by experimental manipulation of the presence or absence of Ca²⁺ (this study).³ It is not known if similar separation of initial CAP binding and cytotoxicity can be experimentally achieved. Therefore it is not yet possible to discern whether increased CAP binding to dltA S. aureus previously reported might also be secondary to greater susceptibility of the mutant bacteria to CAP-induced envelope damage or truly reflect higher initial CAP binding.

Effects of disruption of dltA-dependent LTA and TA modifications on initial binding of CAP or PLA₂ are also likely to depend on the presence of electrolytes in the incubation medium, especially Mg²⁺ and Ca²⁺, which are known to bind with relatively high affinity to unsubstituted repeating glycerol phosphate moieties (38, 39). For this reason, we supplemented the incubation medium with Mg²⁺ when Ca²⁺ was depleted by EGTA. In contrast, earlier studies used divalent cation-poor media creating conditions more favorable for CAP binding to unsubstituted LTA and TA in the dltA bacteria.

We do not, as yet, understand how dltA-dependent modifications affect activity of bound PLA₂. It has been shown by Peschel's group that dltA-dependent modifications neither affect overall LTA and TA content, the amount of N-acetyl glucosamine in TA nor overall surface protein content (19, 40). However, it is possible that less substitution of the polyanionic LTA and TA with D-alanine, while unimportant for initial PLA₂ attachment to the cell surface, permits greater penetration of the bound enzyme to the cell membrane. LTA and TA form a negatively charged network that connects the bacterial membrane and cell wall (17). Depending on concentrations of ambient counter ions, this network could be more densely anionic in the dltA strain and could facilitate penetration of PLA₂ through anionic pores of the cell wall by "electrostatic steering" (41). In the absence of surrounding cell wall layers, such steering may be unnecessary rendering membrane protoplasts from wt and dltA bacteria equally sensitive to added PLA₂ despite differences in remaining membrane-associated LTA. The similar effect of dltA disruption on each of the PLA₂ charge variants studied (Fig. 5), suggests that effects not directly related to electrostatic interactions between PLA₂ and LTA/TA may be more important. These could include changes in packing of LTA and TA chains, binding of cationic autolysins and/or levels of cell wall-bound Ca²⁺ (see Fig. 9).

View larger version (48K): [in this window] [in a new window]   Fig. 9.   Possible effects of D-alanylation of LTA and TA on sensitivity of S. aureus to Group IIA PLA2. Alanylation (A) of LTA and TA may be important in maintaining barrier properties of the cell wall to macromolecules such as PLA2. The absence of alanylation may allow easier penetration of the PLA2 to the cell membrane either by steric effects and/or increased electrostatic binding of autolysins (H, hydrolases). Highly cationic PLA2 may displace cell wall-associated autolysins leading to localized cell wall degradation and easier penetration of the PLA2. Absence of alanylation of LTA and TA in dltA bacteria may also provide a denser network of negative charges facilitating movement of bound PLA2 and/or increased binding of divalent cations (++), which could be displaced upon binding of PLA2 and be locally available for Ca2+-dependent antibacterial activity of bound PLA2. Blue, basic; red, acidic; K, Lysine.

We believe that our studies add significant new insight concerning the mechanism of PLA₂ antibacterial action and of cationic peptides, more generally. It should be noted that the seminal studies of Beers et al. (15) have previously shown the importance of PLA₂ net positive charge in permitting enzyme penetration of the cell wall for attack of bacterial membrane PL of intact bacteria. However, this study did not reveal more precisely how enzyme positive charge facilitated access of PLA₂ to the bacterial cytoplasmic membrane. Our studies show a correlation of enzyme positive charge and antibacterial potency with Ca²⁺-independent binding and describe an experimental approach that permits unambiguous assessment of initial enzyme binding to the bacterial cell wall by precluding subsequent Ca²⁺-dependent envelope alterations. The use of this method has also permitted us to recognize that the effect of modification of cell wall polyanions on PLA₂ antibacterial activity is, unexpectedly, not on PLA₂ binding but rather on the activity of PLA₂ once bound to the bacterial envelope. The contrasting effects of mutational alterations of PLA₂ and bacterial surface charge properties has revealed the likely importance of both initial enzyme-cell wall electrostatic interactions, regulated by enzyme charge properties, and subsequent movement and actions of bound enzyme, regulated by cell wall charge properties, in the antibacterial action of PLA₂.

Whatever the mechanism of greater PLA₂ activity against dltA S. aureus, it is apparent that this modification could play a substantial role in regulation of bacterial sensitivity to this potent host defense protein (42). Effects of dltA-dependent cell wall modifications are not limited to modulating bacterial sensitivity to antimicrobial peptides and polypeptides. Biofilm formation, adhesion, acid sensitivity, protein folding, and autolytic activity are also affected (42-48), implying pleitropic effects of this single cell wall modification on cell envelope structure and function. Of particular interest is recent evidence of an apparent novel two-component system involved in regulation of the dlt operon (49, 50). This finding strongly suggests that alanylation may be a regulated modification of the cell wall LTA and TA. If so, it is tempting to speculate that environmental cations, including host defense cationic peptides and polypeptides, could serve as environmental cues to regulate these cell wall modifications and thereby enhance bacterial resistance and persistence in potentially adverse environments such as the host.

	ACKNOWLEDGEMENTS

We thank Polonca Prohinar and Dr. Theresa Gioannini for careful and critical review of the article.

	FOOTNOTES

^* This work was supported by United States Public Health Service Grant AI-18571 (to J. W.) and National Institutes of Health Grant HL36235 (to M. H. G.).The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

^§§ To whom correspondence should be addressed: Dept. of Internal Medicine, University of Iowa, 200 Hawkins Dr., Iowa City, IA 52246. Tel.: 319-384-8622; Fax: 319-356-4600; E-mail: jerrold-weiss@uiowa.edu.

Published, JBC Papers in Press, September 30, 2002, DOI 10.1074/jbc.M205104200

² A. Foreman-Wykert, J. Lee, P. Elsbach, and J. P. Weiss, manuscript in preparation.

³ N. S. Liang, R. S. Koduri, M. H. Gelb, and J. P. Weiss, manuscript in preparation.

⁴ K. Zarember, P. Elsbach, and J. P. Weiss, manuscript in preparation.

	ABBREVIATIONS

The abbreviations used are: PLA₂, phospholipase A₂; BSA, bovine serum albumin; CAP, cationic antimicrobial peptides; CFU, colony-forming units; CL, cardiolipin; DAG, diacylglycerol; HBD-3, human--defensin-3; HBSS ⁺/, Hanks' balanced salt solution: +/ Ca²⁺ and Mg²⁺; HPLC, high pressure liquid chromatography; LTA, lipoteichoic acid; PA, phosphatidic acid; PEI, polyethylenimine; PG, phosphatidylglycerol; PL, phospholipids; TA, teichoic acid; wt, wild type.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES