Analytical Biochemistry 439 (2013) 73–79
Contents lists available at SciVerse ScienceDirect
Analytical Biochemistry journal homepage: www.elsevier.com/locate/yabio
Assay for peptidoglycan O-acetyltransferase: A potential new antibacterial target Patrick J. Moynihan, Anthony J. Clarke ⇑ Department of Molecular and Cellular Biology, University of Guelph, Guelph, Ontario, Canada N1G 2W1
a r t i c l e
i n f o
Article history: Received 12 February 2013 Accepted 16 April 2013 Available online 7 May 2013 Keywords: Peptidoglycan O-acetylation O-acetyltransferase Transacetylase Muramic acid
a b s t r a c t The O-acetylation of peptidoglycan occurs at the C-6 hydroxyl group of muramoyl residues in many human pathogens, both gram positive and gram negative, such as Staphylococcus aureus and species of Campylobacter, Helicobacter, Neisseria, and Bacillus, including Bacillus anthracis. The process is a maturation event being catalyzed either by integral membrane O-acetylpeptidoglycan transferase (Oat) of gram-positive bacteria or by a two-component peptidoglycan O-acetyltransferase system (PatA/PatB) in gram-negative cells. Here, we describe the development of the first in vitro assay for any peptidoglycan O-acetyltransferase using PatB from Neisseria gonorrhoeae as the model enzyme. This assay is based on the use of chromogenic p-nitrophenyl acetate as the donor substrate and chitooligosaccharides as model acceptor substrates in place of peptidoglycan. The identity of the O-acetylated chitooligosaccharides was confirmed by matrix-assisted laser desorption/ionization time-of-flight mass spectrometry. Rates of transacetylations were determined spectrophotometrically by monitoring p-nitrophenol release after accounting for both spontaneous and enzyme-catalyzed hydrolysis of the acetate donor. Conditions were established for use of the assay in microtiter plate format, and its applicability was demonstrated by determining the first Michaelis–Menten kinetic parameters for PatB. The assay is readily amenable for application in the high-throughput screening for potential inhibitors of peptidoglycan O-acetyltransferases that may prove to be leads for novel classes of antibiotics. Ó 2013 Elsevier Inc. All rights reserved.
The peptidoglycan (PG)1 sacculus completely surrounds the cytoplasmic membrane of nearly all bacterial cells to maintain its integrity (the exceptions are the mycoplasmas). As such, PG represents the major structural element essential to bacteria. PG is a heteropolymer of alternating N-acetylglucosaminyl (GlcNAc) and Nacetylmuramoyl (MurNAc) residues linked b-(1–4) and short ‘‘stem’’ peptides that are attached to the lactyl moiety of each MurNAc residue (Fig. 1). The existence of the peptides permits cross-linking between neighboring glycan strands to provide the continuous threedimensional PG sacculus. Given its importance to bacterial cell viability, PG is the target of the lytic enzymes associated with the innate immune systems of host cells such as lysozymes (muramidases). To protect from this lysis, many bacteria decorate their PG with simple
⇑ Corresponding author. Fax: +1 519 837 1802. E-mail address:
[email protected] (A.J. Clarke). Abbreviations used: PG, peptidoglycan; GlcNAc, N-acetyl-D-glucosamine; MurNAc, N-acetyl-D-muramic acid; Oat, O-acetylpeptidoglycan transferase; CoA, coenzyme A; Pat, peptidoglycan O-acetyltransferase; SDS, sodium dodecyl sulfate; IPTG, isopropyl b-D-1-thiogalactopyranoside; SPE, solid phase extraction; NTA, nitrilotriacetic acid; pNP-Ac, p-nitrophenyl acetate; DP, degree of polymerization; DTNB, 5,50 -dithiobis-(2nitrobenzoic acid); HPLC, high-performance liquid chromatography; TFA, trifluoroacetic acid; MALDI–TOF, matrix-assisted laser desorption/ionization time-of-flight; MS, mass spectrometry. 1
0003-2697/$ - see front matter Ó 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.ab.2013.04.022
aglycon moieties, particularly acetate (recently reviewed in Refs. [1,2]). The O-acetylation of PG occurs at the C-6 hydroxyl group of muramoyl residues (Fig. 1). This modification is performed by many human pathogens, both gram positive and gram negative, such as Staphylococcus aureus and species of Campylobacter, Helicobacter, Neisseria, and Bacillus, including Bacillus anthracis [1–3]. The extent of O-acetylation varies with species and strain, as well as culture condition, and ranges between 20% and 70% (relative to MurNAc residues). These levels are significant because the activity of lysozyme is inhibited in a concentration-dependent manner [4– 8] through steric hindrance that precludes productive binding of Oacetylated PG. With B. anthracis, spore PG is heavily O-acetylated [9] in addition to the PG of vegetative cells, as is the PG of VBNC (viable but non-culturable) cells of Enterococcus faecalis [10]. Presumably, this O-acetylation helps to protect these different cell forms from degradation and thereby contributes to the persistence of the potential pathogens. In addition to protecting cells from lysis, O-acetylation leads to persistence of high-molecular-weight fragments of PG circulating within mammalian hosts, thereby causing serious pathobiological effects such as complement activation, pyrogenicity, somnogenesis, and arthrogenicity (reviewed in Refs. [1,2]). From the physio-
74
Assay for peptidoglycan O-acetyltransferase / P.J. Moynihan, A.J. Clarke / Anal. Biochem. 439 (2013) 73–79
CH3 C=O
O
O
O H3C-CH C=O L-Ala
CH2O
CH2OH O
CH2OH O
NH C=O CH3
D-Glu
OH
O
O
O NH H3C-CH C=O C=O L-Ala CH3 D-Glu
NH C=O CH3
CH2
CH2OH O
O
OH
O
O NH H3C-CH C=O C=O L-Ala CH 3
m-DAP
m-DAP
D-Ala
D-Ala
D-Ala
GlcNAc
O-Ac-MurNAc
NH C=O CH3
D-Glu
m-DAP
MurNAc
O O
GlcNAc
1,6-anhydro-MurNAc
Fig.1. Model structure of PG. The O-acetylation of PG in gram-negative bacteria occurs specifically at the C-6 position of MurNAc residues, whereas glycan chains terminate with 1,6-anhydromuramoyl residues due to the action of lytic transglycosylases. The lytic transglycosylases act like lysozyme to specifically cleave PG between MurNAc and GlcNAc residues (denoted by solid arrow), whereas the presence of O-acetyl groups totally precludes their activity (open arrow) because a free C-6 hydroxyl group is required to generate their 1,6-anhydromuramoyl product.
logical perspective of the producing bacterium, PG O-acetylation controls the function of autolysins [11], endogenous enzymes required for the biosynthesis and turnover of PG, insertion of appendages and secretion systems, and cell division/septation (reviewed in Refs. [12,13]). This applies particularly to the major class of autolysins involved with these processes, the lytic transglycosylases. Like lysozyme, these enzymes cleave PG between MurNAc and GlcNAc residues, but they are not hydrolases; rather, they catalyze the bond cleavage with the concomitant formation of 1,6anhydromuramoyl residues (Fig. 1) [14]. Consequently, free and unmodified C-6 hydroxyl groups on MurNAc residues are a strict requirement for lytic transglycosylases. Because the rampant activity of these enzymes would lead to cell lysis, the addition and removal of the O-acetylation provides important control of their activity at the substrate level. As such, the enzymes responsible for the O-acetylation of PG have been proposed to serve as new antibacterial targets [1,12]. PG O-acetylation occurs as a maturation event, taking place outside of gram-positive bacteria and within the periplasm of gramnegative bacteria. Two distinct enzymatic systems have been identified for the respective bacterial types (Fig. 2). With gram-positive cells, an integral membrane, O-acetylpeptidoglycan transferase (OatA or OatB), is postulated to first translocate acetate from cyto-
P
P
P
P
CH3CO O
OatA
PatA
PatB
plasmic pools of acetyl-CoA (coenzyme A) through the cytoplasmic membrane and then transfer it to acceptor sites on PG (recently reviewed in Ref. [11]). An analogous process has been discovered in gram-negative bacteria but requiring two separate proteins: an integral membrane protein serving as the acetyl translocator, named peptidoglycan O-acetyltransferase A (PatA), and a periplasmic PatB that functions as an acetyltransferase [15]. Interestingly, PatA and PatB are present in some species of the gram-positive genus Bacillus, including B. anthracis, in addition to their Oat enzymes [1,9]. PatA remains uncharacterized, but we have demonstrated experimentally that PatB of Neisseria gonorrhoeae is localized to the periplasm and that it requires an acetate translocating protein, PatA or a homolog, to function as an O-acetyltransferase in vivo [15]. More recent studies have confirmed that PatA and PatB are required for O-acetylation of the PG in Neisseria meningitides [16] and Helicobacter pylori [17]. However, as with the Oat proteins, the enzymatic properties of PatB have not been determined because an in vitro kinetic assay for any PG O-acetyltransferase has not been developed despite their initial discovery more than 7 years ago [18]. This is perhaps understandable given the technical challenges associated with the production and then recombination in vitro of membrane proteins acting on an insoluble substrate; PG remains insoluble even in boiling sodium dodecyl sulfate (SDS), and this in fact forms the basis of its isolation from all other cellular material. In the current study, we present the development of the first in vitro assay for PG O-acetyltransferase using N. gonorrhoeae PatB. This was accomplished by our discovery that PatB is able to accommodate and use simple activated acetate donors and catalyze the transfer of acetate to chitooligosaccharides, homopolymers of GlcNAc that serve as a simple mimic of PG.
Materials and methods Chemicals and biochemicals cytoplasm CoA-S-C-CH3 O CoASH
CoA-S-C-CH3 O CoASH
Fig.2. Proposed pathway for the O-acetylation of PG by OatA and PatA/PatB. The membrane-bound module of OatA of gram-positive bacteria, or PatA of gramnegative bacteria, translocates acetate from cytoplasmic pools of acetyl-CoA to the outer leaflet of the cytoplasmic membrane for its transfer to PG by the exterior module of OatA or PatB, respectively.
Acrylamide and glycerol were obtained from Fisher Scientific (Nepean, ON, Canada), whereas isopropyl b-D-1-thiogalactopyranoside (IPTG) was purchased from Roche Diagnostics (Laval, QC, Canada). Chitooligosaccharides were products of Toronto Research Chemicals (Toronto, ON, Canada) or Carbosynth (Berkshire, UK). All growth media were obtained from Difco Laboratories (Detroit, MI, USA). Qiagen (Valencia, CA, USA) supplied Ni2+–NTA (nitrilotriacetic acid) agarose, Source Q was purchased from GE Healthcare
Assay for peptidoglycan O-acetyltransferase / P.J. Moynihan, A.J. Clarke / Anal. Biochem. 439 (2013) 73–79
(Piscataway, NJ, USA), graphitized carbon and graphitized carbon solid phase extraction columns (Carbograph SPE) were products of Grace Canada (Ajax, ON, Canada), Hypercarb PGC columns were supplied by Thermo Electron (Rockford, IL, USA), and Rezex ROA– Organic Acid H+ columns were products of Phenomenex (Torrance, CA, USA). The K-ACETRM acetic acid analysis kit was purchased from Megazyme (Wicklow, Ireland), whereas mouse anti-His6 antibody was obtained from Santa Cruz Biotechnology (Santa Cruz, CA, USA). Unless otherwise stated, all other chemicals and reagents were supplied by Sigma–Aldrich Canada (Oakville, ON, Canada). Production and purification of PatB Due to low protein yields with prior constructs [15], an N-terminally truncated variant of PatB (lacking its 26-amino-acid, periplasmic-localizing signal sequence) was cloned into the Champion pET-SUMO expression system encoding an N-terminal SUMO tag (details of the cloning will be described elsewhere). Escherichia coli BL21⁄ was transformed with this construct, named pACPM33, for overproduction of PatB–SUMO fusion protein. Cells were grown in 1 L of Super Broth (32 g of tryptone–peptone, 20 g of yeast extract, and 5 g of NaCl per liter) at 37 °C to an optical density (OD) of 0.6. Expression of sumo-patB was induced with a final concentration of 1 mM IPTG for a period of 3 h. Cells were harvested by centrifugation (6000g, 15 min, 4 °C) and immediately resuspended in 30 ml of Lysis Buffer (20 mM sodium phosphate buffer [pH 8.0] containing 500 mM NaCl and 10 mM imidazole) and subjected to lysis by four successive passages through an EmulsiFlex C-3 homogenizer (AVESTIN, Ottawa, ON, Canada) at 15,000 psi. The lysate was clarified by centrifugation (6000g, 15 min, 4 °C) and then applied to an Ni2+–NTA gravity flow column with a bed volume of 500 ll. The resin was washed with Lysis Buffer, and bound proteins were eluted with Lysis Buffer containing 250 mM imidazole. Eluted proteins were dialyzed exhaustively at 4 °C first against 20 mM sodium phosphate buffer (pH 8.0) containing 100 mM NaCl and 50 mM L-arginine and then against the same buffer but lacking the NaCl and arginine. Further purification of PatB was achieved by anion exchange chromatography on Source Q. Samples were applied to resin, previously equilibrated in the dialysis buffer, at a flow rate of 1 ml/min, and proteins were recovered by application of a linear gradient of 0 to 1.0 M NaCl over 50 min. SUMO–PatB eluted in approximately 100 mM NaCl. Residual low-molecularweight contaminants were removed by ultrafiltration with four washes of 500 mM NaCl in 20 mM sodium phosphate buffer (pH 8.0) using an Amicon centrifugal filtration device (30-kDa cutoff), and the purified protein was maintained at 4 °C and used immediately.
75
drawing 10 ll of the 2.0-ml reaction mixture through a pipette tip packed with graphitized carbon. Both the acceptor sugars and the acetyl donor are retained on the carbon, whereas the eluent can be analyzed for free acetate content. Acetate analysis Analysis of acetate was performed either by a coupled acetate kinase/phosphotransacetylase enzyme assay using the K-ACETRM kit or by high-performance liquid chromatography (HPLC)-based organic acid analysis. For HPLC, samples were loaded onto a Rezex ROA–Organic Acid H+ column at 60 °C using a Beckman System Gold HPLC instrument. Elution was performed isocratically with 5 mM H2SO4 at a flow rate of 0.6 ml/min, and acetic acid was detected at 210 nm. With both assays, the quantity of acetate observed in the bulk solvent at a given time point was subtracted from the quantity of acetyl donor consumed to determine the amount of acetate transferred to the sugar acceptor. Analysis of O-acetylation For the direct analysis of O-acetylation, sugars were separated from the acetyl donors by adsorption chromatography using 4 ml of graphitized carbon SPE columns (Carbograph SPE). Prior to the application of sample, columns were washed with 3 column volumes of 80% acetonitrile in 0.1% trifluoroacetic acid (TFA) and then equilibrated with 3 volumes of water. Samples (100 ll) were applied to the column, and unbound material was washed out with 6 volumes of water. Sugars were eluted with 1 ml of 50% acetonitrile in 0.1% TFA and dried under vacuum in a SpeedVac (Thermo Electron). Ester-linked acetate was removed from the sugars by saponification in 100 ll of 500 mM NaOH for 1 h at 25 °C. Purification and quantification of O-acetylated oligosaccharides During the course of a typical enzyme assay (as described above), 100-ll samples of reaction mixtures were removed at 5min intervals and quenched with the addition of 30 ll of cold 8% perchloric acid. Analysis of O-acetylated oligosaccharides was performed by adsorption chromatography HPLC on porous graphitic carbon using a Hypercarb PGC column (100 4.6 mm, 3 lm) at 80 °C. Samples (50 ll) of quenched reactions were injected at 1 ml/min onto the column previously equilibrated in Milli-Q water, and O-acetylated products were separated from substrate using a linear gradient from 0% to 50% acetonitrile over 50 min. Sugars were detected spectrophotometrically at 210 nm.
Enzymatic assay of PatB MS analysis Samples of PatB (1.0 or 3.0 lM) were incubated at 37 °C with varying concentrations (0.05–8 mM) of different potential acceptors and 0.5 to 6.0 mM of acetyl donor in either 50 mM sodium phosphate buffer (pH 6.5) or 50 mM SPG buffer (succinic acid/sodium dihydrogen phosphate/glycine, 2:7:7, 5% glycerol, pH 6.5). Acetyl donors included p-nitrophenyl acetate (pNP-Ac, dissolved in ethanol, 5% final volume) and acetyl-CoA, whereas acceptors included GlcNAc, MurNAc, GlcNAc–MurNAc dipeptide, and chitooligosaccharides (degree of polymerization [DPs] 2–6). The progress of reactions was monitored spectrophotometrically at 410 nm for the release of p-nitrophenol and for the production of CoA (by its free thiol) following the addition of 0.5 mM 5,50 -dithiobis-(2-nitrobenzoic acid) (DTNB) dissolved in ethanol (final volume of ethanol in reaction mixtures: 5%). The quantity of acetate released from the acetyl donor to the bulk solvent during transacetylase reactions with acceptor substrates (other than water) was monitored by
Matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI–TOF MS) was used for the routine analysis of amino sugars employing a Bruker Reflex III mass spectrometer. Prior to analysis, samples were purified and desalted on Carbograph SPE columns as described above for the analysis of O-acetylation. 5-Chloro-2-mercaptobenzothiazole was previously demonstrated to be an effective matrix for the analysis of amino sugars derived from PG [15], and it was also found to be suitable for chitooligosaccharides. Recrystallized matrix was suspended in acetonitrile to generate a saturated solution, and an equal volume of it was added to samples previously spotted onto plates. MALDI– TOF MS was conducted in the reflectron and positive mode using a 337-nm nitrogen laser set to 109- to 121-lJ output. The resultant spectra were examined using the mMass open source MS tool (http://www.mmass.org).
76
Assay for peptidoglycan O-acetyltransferase / P.J. Moynihan, A.J. Clarke / Anal. Biochem. 439 (2013) 73–79
Other analytical methods
Rates of hydrolysis of acetyl donors
SDS–PAGE (polyacrylamide gel electrophoresis) on 12% acrylamide gels was conducted by the method of Laemmli [19] with Coomassie Brilliant Blue staining and Western immunoblot analysis as described previously [15]. Protein concentrations were determined using the bicinchoninic acid-based assay according to the manufacturer’s instructions (Pierce Biotechnology, Rockford, IL, USA).
The rate of acetyl-CoA hydrolysis catalyzed by PatB was measured by monitoring the production of free CoASH spectrophotometrically at 420 nm using 0.5 mM of the thiol reagent DTNB. Initially, these reactions were conducted using SPG buffer, but it was found that the enzyme was more active in simple sodium phosphate buffer, and so all following assays were performed in 50 mM sodium phosphate. The specific activity was calculated to be 9.3 nmol min1 mg protein1 for reaction with 2 mM acetylCoA at pH 6.5 and 37 °C. Using acetyl-CoA concentrations ranging from 0.5 to 6.0 mM, the Michaelis–Menten parameters of KM and kcat for this hydrolysis were determined to be 0.99 mM and 0.011 s1, respectively, providing an overall efficiency (kcat/KM) of 10.6 M1 s1 (Table 1). However, this relatively high rate of (wasteful) hydrolysis in combination with the high cost of the commercial product prompted us to test for alternative acetyl donors. Recognizing earlier that PatB functions as a weak esterase when assayed with pNP-Ac as substrate [15], we investigated the potential of using it as an acetyl donor for transacetylation reactions. After accounting for spontaneous loss, a specific activity for hydrolysis of 730 nmol min1 mg protein1 was determined with 2 mM pNP-Ac at pH 6.5 and 37 °C. Under these same conditions, the KM and kcat for this hydrolysis catalyzed by PatB were 0.96 mM and 0.76 s1, respectively. Thus, the overall catalytic efficiency of 797 M1 s1 for this reaction was 75 times greater than that for acetyl-CoA hydrolysis, suggesting that pNP-Ac may serve as a better co-substrate for in vitro transacetylation reactions.
Results and discussion Qualitative analysis PatB was first identified as an O-acetyltransferase using an in vivo assay that monitored the addition of O-acetyl groups to the PG sacculus of E. coli transformants harboring its encoding gene patB [15]. This was possible because E. coli lacks the oap gene cluster coding for the acetylating and deacetylating enzymes [18]; thus, its PG does not normally get modified in this fashion. PatB functions in the periplasm, and the source of translocated acetate was shown to be provided by WecH, a protein proposed to be involved in the O-acetylation of the enterobacterial common antigen, an exo-polysaccharide produced by E. coli. Attempts to refine the assay by using cell-free extracts of the walls (possessing both membrane and PG fractions) supplemented with acetyl-CoA were met with limited success [15]. As expected, developing an in vitro assay involving both integral and peripheral membrane proteins in combination with a totally insoluble substrate was extremely challenging. Nonetheless, the preliminary work did suggest that PatB may be able to use the soluble acetate donor. An ideal assay would involve both the acetate donor and acceptor in soluble form and so our preliminary experiments tested the ability of PatB to transfer acetate from acetyl-CoA to commercially available mono- and disaccharides related to PG. Thus, 1.0 lM PatB in 50 mM SPG buffer (pH 6.0) and 0.5 mM acetyl-CoA was incubated individually with GlcNAc, chitobiose (di-GlcNAc), MurNAc, and GlcNAc–MurNAc dipeptide at 25 °C for 30 min. Qualitative analysis of the reaction mixtures by MALDI–TOF MS indicated that neither of the monosaccharides or disaccharides was acetylated. However, control experiments lacking any acceptor revealed that PatB catalyzes the hydrolysis of acetyl-CoA. This finding was not too surprising given that the enzyme was previously observed to function as a weak esterase when assayed with pNP-Ac as substrate [15]. Despite this lack of initial success, we tested the ability of PatB to use oligomeric chitooligosaccharides with DPs 3 to 6 as potential acceptors. Again, reaction products were analyzed qualitatively by MALDI–TOF MS, which revealed the production of acetylated products and that PatB appeared to be increasingly more active with increasing DP. Moreover, the enzyme catalyzed multiple acetylations of the individual chitooligosaccharides. Thus, treatment of chitohexaose, for example, led to the formation of the mono-, di-, tri-, and tetra-acetylated products (Fig. 3). Such multi-acetylations would complicate the kinetic analysis of the enzyme reaction recognizing the gradual production of different substrates for the enzyme (i.e., initial production of mono-acetylated species, which becomes a second substrate along with the unacetylated starting material, etc.). This potential complication for kinetic analyses was compounded by the fact that the larger commercially produced chitooligosaccharides were contaminated with shorter forms, all of which served as substrates for the multi-acetylations (Fig. 3). Both of these issues of multiple acetylations and substrate contamination were minimized with chitotriose, and so all further reactions involved chitotriose as acceptor for assay development.
Transacetylation with pNP-Ac as acetyl donor Because the kcat for the hydrolysis pNP-Ac was two orders of magnitude greater than that for acetyl-CoA, whereas its KM was not significantly different, use of the former as an acetyl donor was investigated using chitotriose as the acceptor. Qualitative analyses of reaction products were determined by MALDI–TOF MS, which indicated the major product to be mono-O-acetylchitotriose (data not shown). That O-acetylation that had been performed was confirmed by the loss of the reaction product following mild base hydrolysis; ester-linked, but not amide-linked, acetate is susceptible to this saponification treatment. Under the same conditions, the inability of PatB to use the smaller chitobiose as an acceptor for the transacetylation was confirmed. Determination of transacetylation rates: analysis of sugars Various methods were developed to quantify the amount of product formed both directly and indirectly. For direct determination, different methods of HPLC were tested, including high-pH anion exchange chromatography, reverse-phase chromatography, and adsorption chromatography, with varying success. The most effective of these was adsorption chromatography on porous graphitic carbon. Thus, complete resolution of starting materials from reaction products was achieved by chromatography using a 100 4.6-mm Hypercarb PGC column (3 lm) operating at 80 °C. The presence of the amide and ester linkages permitted the detection of the substrates and products, respectively, by absorbance at 210 nm. Individual fractions were identified by MS either in-line or after their collection. As seen in Fig. 4, complete resolution of substrate from products was achieved. Quantification of reaction products by this method, however, was complicated by (i) the apparent presence and resolution of anomers, (ii) the contamination of the individual chitooligosaccharides provided by the manufacturer, (iii) the lack of appropriate standards to generate standard curves, and (iv) the relatively high concentrations of substrate required for detection of products.
77
Assay for peptidoglycan O-acetyltransferase / P.J. Moynihan, A.J. Clarke / Anal. Biochem. 439 (2013) 73–79
GlcNAc6
Relative Intensity
1260.1130
GlcNAc6 + 2OAc 1344.1899
GlcNAc6 + 1OAc 1302.1518
GlcNAc6 + 4OAc 1428.2210
1386.2233
GlcNAc7 + 1OAc 1505.3627
GlcNAc6 + 3OAc GlcNAc7
GlcNAc7 + 2OAc
1463.3287
1547.3570
a b c 1250
1300
1350
1400
m/z
1450
1500
1550
Fig.3. MALDI–TOF MS analysis of transacetylation reaction products produced by PatB. Enzyme (1.0 lM) in 50 mM SPG buffer (pH 6.0) was incubated with 0.5 mM acetylCoA and chitohexaose (G6) at 25 °C. After 30 min of incubation, samples were withdrawn and applied to a Bruker Reflex III mass spectrometer using 5-chloro-2mercaptobenzothiazole as the matrix: (a) complete reaction mixture following incubation; (b) saponified reaction products following incubation in 100 mM ammonium hydroxide (to release ester-linked acetate); (c) negative control, G6, and acetyl-CoA incubated in the absence of PatB. The solid bar denotes 5000 absolute intensity units, and all of the ions detected were the sodium adducts.
Table 1 Kinetic parameters of chitotriose O-acetylation catalyzed by PatB. Substrate
Co-substrate
KM (mM)
kcat (s1)
kcat/KM (M1 s1)
AcetylCoA pNP-Ac Chitotriose
H2O
0.99 ± 0.08
0.011 ± 0.0003
10.6 ± 0.90
H2O 4 mM pNPAc
0.96 ± 0.10 2.63 ± 0.22
0.76 ± 0.03 0.74 ± 0.024
797 ± 88 280 ± 25
Note: Reactions of 1.0 lM PatB were performed in 50 mM sodium phosphate buffer (pH 6.5) at 37 °C.
(both substrate and products). The latter were then recovered free of salts by elution with acetonitrile. Following their concentration, the amount of O-acetylation was determined by first saponification to specifically release any ester-linked acetate followed by its quantitative analysis. Two methods of quantification were tested: a chromogenic enzyme-based acetic analysis kit and direct analysis by HPLC using an organic acid column. Whereas technically these approaches did provide rates of individual reactions, they were very laborious, involving multiple steps that led to unacceptable margins of error. To compound this issue, variability in the packing of the disposable graphitic columns used led to occasional carryover of acetate donor. Hence, further attempts to quantify the sugar-based reaction products were abandoned.
G3 Determination of transacetylation rates: analysis of co-product formation
0.2 au A 210nm
G4 O-AcG3 20 min 15 10 5 0.5 0
/
/
35
40
45 Time (min)
50
Fig.4. Separation of reaction products produced by PatB. The enzyme (1 lM final concentration) in 50 mM sodium phosphate buffer (pH 6.0) (200 ll final volume) was incubated with 1 mM chitotriose (G3) and 4 mM acetyl-CoA at 37 °C for the times shown. Samples (50 ll) of the reaction volumes were applied to a 100 4.6mm Hypercarb PGC column previously equilibrated in Milli-Q water, and elution of products was achieved at 1.0 ml min1 using a linear gradient of 0–50% acetonitrile over 50 min. G4 denotes the chitotetraose contamination present in the commercial preparation of chitotriose.
In a final attempt to determine reaction rates of the transacetylation reactions by quantifying the sugar products produced, time course samples were removed from reaction mixtures and subjected to adsorption chromatography on graphitized carbon solid-phase extraction columns. This treatment successfully separated acetyl donors and any free acetic acid from the sugars
Despite the amount of both spontaneous and enzyme-catalyzed hydrolysis of the acetyl donor co-substrate known to occur, we explored the possibility of determining accurate rates of reaction by monitoring the release of the co-product produced during the transacetylations, p-nitrophenol. The great advantage of this approach is that this co-product is chromogenic and, thus, can be quantified reliably by spectrophotometry in real time during the course of reactions without the need for any subsequent sample manipulation. Thus, PatB (1 lM final concentration) was incubated with 5 mM pNP-Ac (i.e., 5 KM value) and 1.0 mM chitotriose in a total volume of 250 ll of 50 mM sodium phosphate buffer (pH 6.0), and the amount of p-nitrophenol released was monitored with time at 410 nm. To control for the spontaneous hydrolysis of pNP-Ac, separate reactions that involved incubation of it alone and together with the chitotriose co-substrate were performed, whereas controls of its enzyme-catalyzed hydrolysis involved enzyme reaction mixtures lacking added chitotriose. As the representative data in Fig. 5 show, the rates of spontaneous pNP-Ac hydrolysis were unaffected by the presence of chitotriose. On the other hand, the rates of PatB-catalyzed p-nitrophenol production were higher when incubated in the presence of chitotriose compared with control reactions lacking the added sugars. Thus, it appeared that the transacetylation activity of PatB could
78
Assay for peptidoglycan O-acetyltransferase / P.J. Moynihan, A.J. Clarke / Anal. Biochem. 439 (2013) 73–79
A
p-NP (nmol)
800
600
400
200 0 0
5
10
15
lation rates is indeed to determine and account for donor acetate hydrolysis spectrophotometrically, as presented in Fig. 5A. To test and demonstrate the utility of this spectrophotometric assay, the specific activity and Michaelis–Menten parameters of PatB toward chitotriose were determined. Thus, rates of transacetylation (in triplicate) were determined for the incubation of 1 lM PatB (final concentration) in 50 mM sodium phosphate buffer (pH 6.5) at 37 °C with 4 mM pNP-Ac and 0.2 to 8.0 mM chitotriose. The plot of initial velocity as a function of chitotriose concentration (Fig. 5B) indicated that the reaction followed typical Michaelis– Menten kinetics using the fixed excess (i.e., >4 KM) of the acetate donor pNP-Ac. Nonlinear regression analysis of these data gave kcat and KM values of 0.74 s1 and 2.63 mM, respectively, providing a kcat/KM value of 280 M1 s1 (Table 1).
Time (min) 10
Concluding remarks
B
6
1.0
1 / vo (min.mmol -1)
vo (nmol.min-1)
8
4
0.5
2 0
0.0 -1
0
2
4
0 1 2 3 4 1 / [GlcNAc3] (mM)-1
6
8
10
[GlcNAc3] (mM) Fig.5. Kinetic analysis of PatB transacetylation activity on chitotriose. (A) Time course of p-nitrophenol release from 4 mM pNP-Ac alone (N) and together with 4 mM chitotriose (), 1.0 lM PatB (j), and 1.0 lM PatB and 4 mM chitotriose (d). Reactions were incubated at 37 °C in 50 mM sodium phosphate buffer (pH 6.5), and rates of p-nitrophenol release were measured spectrophotometrically at 410 nm. (B) Determination of Michaelis–Menten parameters of PatB-catalyzed O-acetylation of chitotriose. PatB (1 lM final concentration) in 50 mM sodium phosphate buffer (pH 6.5) was incubated at 37 °C with 4 mM pNP-Ac and 0.2–8.0 mM chitotriose, and initial velocities were determined from the first 10 min of reaction after accounting for pNP-Ac hydrolysis. Inset: double-reciprocal plot of initial velocities as a function of substrate concentration (for presentation purposes only; Michaelis–Menten parameters were determined from the primary plot by nonlinear regression analysis).
be determined from the difference of the rates of these two enzyme-catalyzed reactions because the latter control reaction would also account for spontaneous hydrolysis. However, it was possible that the presence of the sugar-based co-substrate in the complete reactions might either increase or decrease the rate of PatB-catalyzed hydrolysis of pNP-Ac, thereby either overestimating or underestimating, respectively, the calculated rate of transacetylations. If this were to occur, the relative rates of pnitrophenol release and acetic acid production would not be the same because the latter is the product only of hydrolysis. To test for this, samples from complete reactions were withdrawn and immediately adsorbed to activated charcoal to both quench any further reaction and remove substrates and p-nitrophenol from solution. The amount of acetic acid present was then assayed for by HPLC-based organic acid analysis. This analysis indicated that the rate of acetic acid release in the enzyme reactions incubated in the presence of the chitotriose was the same, within error, as that incubated with only the co-substrate pNP-Ac, and both matched the rate of p-nitrophenol release of the latter enzyme control as quantified by absorbance measurements. This provided assurance that the appropriate control for determining transacety-
We have described the first in vitro assay to measure the activity of a peptidoglyan O-acetyltransferase. Furthermore, in addition to PatB, preliminary studies showed that it can be used to assay the activity of OatA, the PG O-acetyltransferase produced by gram-positive bacteria. This assay was developed based on our earlier observation that PatB has weak esterase activity against both Oacetylated PG and typical chromogenic esterase substrates such as pNP-Ac [20]. Thus, this situation is analogous to the finding that authentic O-acetylpeptidoglycan esterase Ape1a from N. gonorrhoeae is also capable of performing reverse transacetylation reactions, albeit only when conducted in the presence of organic solvents [21]. With PatB, its natural acetate donor is proposed to be the integral membrane protein PatA, which is thought to translocate acetate from cytoplasmic pools of acetyl-CoA to the periplasm [22]. Indirect evidence in support of this proposal has been obtained [20], but without a biochemical assay direct evidence is still lacking. Moreover, and unfortunately, the threedimensional structure of any PG O-acetyltransferase has not been solved, thereby precluding any mechanistic insight. Presumably, however, PatB has a binding site to accommodate both PatA and its bound acetate, and so it is not unreasonable to expect that the simple chromogenic acetate donor substrates such as pNP-Ac would serve as substitutes in the transacetylation reaction. Likewise, it is reasonable to expect the binding cleft of PatB to accommodate the structurally more simple chitooligosaccharides as substrates in place of PG given their chemical and structural similarity. Indeed, this would be analogous to the situation with many lysozymes, enzymes that also act on PG as their natural substrate, which can also function as weak chitinases [23]. The availability of a facile assay to probe the kinetic behavior of PatB (and OatA) and their inhibition will greatly advance studies on the characterization of these enzymes as potential new antibiotic targets. Moreover, the assay would be amenable with little effort for application in a high-throughput screen to identify new inhibitors and potential leads to novel antibiotics. In this regard, a recent study demonstrated that an inhibitor of O-acetylpeptidoglycan esterase, discovered by a high-throughput screen, can function as an antibacterial [24], indicating the sensitivity of bacteria that O-acetylate their PG to its alteration. Given the role of PG Oacetylation in control of the autolytic activity of the lytic transglycosylases [1,11], it would be expected that blocking this reaction would be even more deleterious to cells. Acknowledgments We thank Dyanne Brewer and Armen Charchoglyan of the Mass Spectrometry Facility (Advanced Analysis Centre, University of Guelph) for their expert assistance with mass spectrometric analy-
Assay for peptidoglycan O-acetyltransferase / P.J. Moynihan, A.J. Clarke / Anal. Biochem. 439 (2013) 73–79
ses. This study was supported by a team operating Grant (TGC114045) to A.J.C. from the Canadian Institutes for Health Research and an Ontario Graduate Scholarship to P.J.M. from the Province of Ontario.
References [1] J.M. Pfeffer, P.J. Moynihan, C.A. Clarke, A.J. Clarke, Control of lytic transglycosylase activity within bacterial cell walls, in: C.W. Reid, S.M. Twine, A.N. Reid (Eds.), Bacterial Glycomics, Caister Academic Press, Norfolk, UK, 2012, pp. 55–68. [2] W. Vollmer, Structural variation in the glycan strands of bacterial peptidoglycan, FEMS Microbiol. Rev. 32 (2008) 287–306. [3] A.J. Clarke, C. Dupont, O-acetylated peptidoglycan: its occurrence, pathobiological significance, and biosynthesis, Can. J. Microbiol. 38 (1992) 85–91. [4] C. Dupont, A.J. Clarke, Dependence of lysozyme-catalysed solubilization of Proteus mirabilis peptidoglycan on the extent of O-acetylation, Eur. J. Biochem. 195 (1991) 763–769. [5] S.G. Swim, M.A. Gfell, C.E. Wilde III, R.S. Rosenthal, Strain distribution in extents of lysozyme resistance and O-acetylation of gonocccal peptidoglycan determined by high-performance liquid chromatography, Infect. Immun. 42 (1983) 446–452. [6] R.S. Rosenthal, W.J. Folkening, D.R. Miller, S.C. Swim, Resistance of Oacetylated gonococcal peptidoglycan to human peptidoglycan-degrading enzymes, Infect. Immun. 40 (1983) 826–829. [7] R.S. Rosenthal, J.K. Blundell, H.R. Perkins, Strain-related differences in lysozyme sensitivity and extent of O-acetylation of gonococcal peptidoglycan, Infect. Immun. 37 (1982) 826–829. [8] J.K. Blundell, G.J. Smith, H.R. Perkins, The peptidoglycan of Neisseria gonorrhoeae: O-acetyl groups and lysozyme sensitivity, FEMS Microbiol. Lett. 9 (1980) 259–261. [9] M.-H. Laaberki, J. Pfeffer, A.J. Clarke, J. Dworkin, O-acetylation of peptidoglycan is required for proper cell separation and S-layer anchoring in Bacillus anthracis, J. Biol. Chem. 286 (2011) 5278–5288. [10] J.M. Pfeffer, H. Strating, J.T. Weadge, A.J. Clarke, Peptidoglcyan O-acetylation and autolysin profile of Enterococcus faecalis in the viable but non-culturable state, J. Bacteriol. 188 (2006) 902–908.
79
[11] P.J. Moynihan, A.J. Clarke, O-acetylated peptidoglycan: controlling the activity of bacterial autolysins and lytic enzymes of innate immune systems, Int. J. Biochem. Cell Biol. 43 (2011) 1655–1659. [12] E.M. Scheurwater, C.W. Reid, A.J. Clarke, Lytic transglycosylases: bacterial space-making autolysins, Int. J. Biochem. Cell Biol. 40 (2008) 586–591. [13] G. Koraimann, Lytic transglycosylases in macromolecular transport systems of gram-negative bacteria, Cell. Mol. Life Sci. 60 (2003) 2371–2388. [14] J.-V. Höltje, D. Mirelman, N. Sharon, U. Schwarz, Novel type of murein transglycosylase in Escherichia coli, J. Bacteriol. 124 (1975) 1067–1076. [15] P.J. Moynihan, A.J. Clarke, O-acetylation of peptidoglycan in gram-negative bacteria: identification and characterization of peptidoglycan Oacetyltransferase in Neisseria gonorrhoeae, J. Biol. Chem. 285 (2010) 13264– 13273. [16] F.J. Veyrier, A.H. Williams, S. Mesnage, C. Schmitt, M.-K. Taha1, I.G. Boneca, DeO-acetylation of peptidoglycan regulates glycan chain extension and affects in vivo survival of Neisseria meningitides, Mol. Microbiol. 87 (2013) 1100–1112. [17] G. Wang, L. Lo, L.S. Forsberg, R.J. Maier, Helicobacter pylori peptidoglycan modifications confer lysozyme resistance and contribute to survival in the host, mBio 3 (2012) e00409–e00412. [18] J.T. Weadge, J.M. Pfeffer, A.J. Clarke, Identification of a new family of enzymes with potential O-acetylpeptidoglycan esterase activity in both gram-positive and gram-negative bacteria, BMC Microbiol. 5 (2005) 49. [19] U.K. Laemmli, Cleavage of structural proteins during the assembly of the head of bacteriophage T4, Nature 227 (1970) 680–685. [20] J.T. Weadge, A.J. Clarke, Identification and characterization of Oacetylpeptidoglycan esterase: a novel enzyme discovered in Neisseria gonorrhoeae, Biochemistry 45 (2006) 839–851. [21] J.T. Weadge, A.J. Clarke, Transacetylations of carbohydrates in organic solvent catalysed by O-acetylpeptidoglycan esterase from Neisseria gonorrhoeae, Biocatal. Biotransform. 26 (2008) 68–75. [22] A.J. Clarke, H. Strating, N.T. Blackburn, Pathways for the O-acetylation of bacterial cell wall polysaccharides, in: R.J. Doyle (Ed.), Glycomicrobiology, Plenum, New York, 2000, pp. 187–223. [23] T. Yoshimoto, S. Hayashida, M. Tobiishi, K. Kado, D. Tsuru, Purification of several bacteriolytic enzymes by affinity chromatography on lysozyme–lysate of Micrococcus lysodeikticus cell wall coupled with Sepharose, J. Biochem. 78 (1975) 253–259. [24] J.M. Pfeffer, A.J. Clarke, Identification of first-known inhibitors of Oacetylpeptidoglycan esterase: a potential new antibacterial target, ChemBioChem 13 (2012) 722–731.
