TY - JOUR
AU - Blanot,, Didier
AB - Abstract A standard assay for the MurG enzyme using a lipid I analogue [MurNAc(Nɛ-dansylpentapeptide)-pyrophosphoryl (R, S)-α-dihydroheptaprenol] and radioactive UDP-N-acetylglucosamine was set up. A high concentration (35%) of dimethylsulfoxide was necessary for maximal activity. Separation and quantitation were accomplished by reverse-phase high performance liquid chromatography (HPLC) in isocratic conditions and on-line radioactivity detection, thereby providing a rapid and accurate assay. The kinetic parameters of the MurG reaction were determined; the reaction was shown to also catalyse the reverse reaction at a measurable rate. A lipid I analogue containing dihydroundecaprenol as the prenyl chain turned out to be a poor MurG substrate, presumably owing to aggregation. MurG, Lipid I, Lipid II, Peptidoglycan, Dimethylsulfoxide 1 Introduction In the biosynthesis of bacterial cell wall peptidoglycan [1], the MurG enzyme catalyses the transfer of an N-acetylglucosaminyl residue from UDP-GlcNAc to MurNAc(pentapeptide)-pyrophosphoryl undecaprenol (lipid I), to yield GlcNAc-MurNAc(pentapeptide)-pyrophosphoryl undecaprenol (lipid II). MurG has been shown to be associated to the cytoplasmic membrane [2]; its purification [3,4] and the determination of its structure [5] have recently been reported. However, some types of experiments (e.g., kinetic studies) require a quantitative and reproducible functional assay, which is itself dependent on the availability of lipid I or analogues thereof. Assays in which lipid I is brought by bacterial membranes [6–10] are useful for, e.g., enzyme purification or inhibitor screening, but not suitable for accurate experiments since the amount of lipid I cannot be controlled. On the other hand, assays that utilise synthetic analogues of lipid I are totally suitable provided these analogues are fairly good substrates for MurG. Recently, three synthetic MurG substrates have been reported: dansyl derivative 1[11], biotinyl derivative 2[12] and compound 3[13]. Substrate 2 allowed the elaboration of an assay based on biotin capture by avidin-derivatised resin [4]; substrate 3 was used in a coupled assay in which the formation of the UDP product was detected by the association of pyruvate kinase and lactate dehydrogenase [13]. In this communication, we report on the utilisation of our dansylated substrate 1[11] in a MurG assay based on HPLC separation; the inclusion of a high concentration (35%) of dimethylsulfoxide (DMSO) in the assay mixture increased the enzymatic activity and improved the reproducibility of the results. 2 Materials and methods 2.1 General procedures Thin layer chromatography was performed on pre-coated plates of silica gel 60 (Merck, Darmstadt, Germany) in chloroform/methanol/water 30:25:4 (v/v); compounds were located under ultraviolet (UV) light or after reaction with iodine vapours. The detection of the dansyl derivatives upon HPLC was carried out by monitoring their fluorescence emission at 514 nm (excitation at 340 nm) with an FL3000 fluorometer (Thermo Finnigan, Les Ulis, France). Amino acid analyses of acid hydrolysates (6 M HCl, 95°C, 16 h) were performed with a Hitachi L8800 analyser (ScienceTec, Les Ulis, France); amino acid ratios were expressed as means±S.D. of three analyses. Matrix-assisted laser desorption/ionisation-time of flight (MALDI-TOF) mass spectra were recorded on a PerSeptive Voyager-DE STR instrument (Applied Biosystems, Foster City, CA, USA) in linear mode with delayed extraction. For the preparation of the samples, the procedure of Therisod et al. [14] was followed with slight modifications. A solution of compound at 10 pmol µl−1 in methanol/isopropanol 1:1 (v/v) was desalted with a few beads of Dowex 50W-X8 (H+). 0.5 µl of a 10 mg ml−1 solution of 2,5-dihydroxybenzoic acid in 0.5 M citric acid was deposited on the plate, followed by 0.5 µl of compound solution. After evaporation of the solvents, desorption and ionisation were obtained by pulses from a 337 nm nitrogen laser. Spectra were recorded in the negative-ion mode at an acceleration voltage of −20 kV. UDP-MurNAc and UDP-MurNAc-pentapeptide were used as external calibrants. 2.2 Compounds UDP-[U-14C]GlcNAc (9.85–11.11 GBq/mmol−1) was purchased from Amersham Pharmacia Biotech (Saclay, France). α-Dihydroundecaprenyl phosphate [structure: (R, S)ωt3c6s] was provided by the Institute of Biochemistry and Biophysics of the Polish Academy of Sciences (Warsaw, Poland). C-terminal His-tagged MurG from Escherichia coli was purified according to Crouvoisier et al. [3]. 2.3 MurNAc(Nɛ-dansylpentapeptide)-pyrophosphoryl (R, S)-α-dihydroheptaprenol 1 This compound was synthesised and purified as previously described [11] except that 1-equivalent portions of activated α-dihydroheptaprenyl phosphate were added at 0, 46 and 72 h. The reaction was stopped at 98 h. Yield: 27%. Stock solutions in CHCl3/MeOH 2:1 (v/v) were kept at −20°C, and appropriate aliquots were pipetted and evaporated prior to enzymatic assays. 2.4 GlcNAc-MurNAc(Nɛ-dansylpentapeptide)-pyrophosphoryl (R, S)-α-dihydroheptaprenol 4 To UDP-GlcNAc (250 nmol) and 1 (83 nmol), dissolved in 0.2 M Tris–HCl, pH 7.5, 10 mM MgCl2, 35% (v/v) DMSO, MurG (28 µg) was added (final volume: 3.1 ml). After 5 h at 37°C, the reaction was stopped by boiling for 3 min. The product was purified by HPLC on a Vydac 218TP column (250×4.6 mm; eluent A, 20 mM ammonium acetate in water/methanol/isopropanol 3:1:1; eluent B, 20 mM ammonium acetate in methanol/isopropanol 1:1; gradient: 0% A for 5 min, 0–100% B for 20 min, 100% B for 15 min, at 0.6 ml min−1); retention time: 35 min. Yield: 60 nmol (72%). Rf 0.38. Amino acid analysis: Mur, 0.92±0.01; Glu, 1.00; Ala, 3.03±0.08; A2pm, 0.22±0.01; GlcN, 0.98±0.02. [M-H]−: calculated 1882.1, found 1882.7. Compound 4 was also synthesised from UDP-[14C]GlcNAc (9.4 nmol, 92.6 kBq) and 1 (2.77 nmol); yield: 1.37 nmol, 13.5 kBq (50% from 1). 2.5 MurNAc(Nɛ-dansylpentapeptide)-pyrophosphoryl (R, S)-α-dihydroundecaprenol 5 This compound was synthesised as described for 1; 1-phospho-MurNAc-pentapeptide(Nɛ-dansyl) [11] (3 µmol) and α-dihydroundecaprenyl phosphate (3×3 µmol) were engaged. It was purified by HPLC on a Vydac 218TP column (250×4.6 mm; eluent A, 20 mM ammonium acetate in water/methanol/isopropanol 4:3:3; eluent B, 20 mM ammonium acetate in methanol/isopropanol 1:1; gradient: 0% A for 5 min, 0–100% B for 30 min, 100% B for 15 min, at 0.6 ml min−1); retention time: 46 min. Yield: 107 nmol (3.6%). Rf 0.50. Amino acid analysis: Mur, 0.89±0.01; Glu, 1.00; Ala, 2.87±0.03; A2pm, 0.07±0.01. [M-H]−: calculated 1951.4, found 1951.9. 2.6 Standard MurG assay Reaction mixtures contained, in a final volume of 25 µl, 0.2 M Tris–HCl, pH 7.5, 10 mM MgCl2, 16 µM UDP-[14C]GlcNAc (1.7 kBq), 16 µM 1, 35% (v/v) DMSO, and enzyme. The reaction was initiated by the addition of enzyme. After 30 min at 37°C, it was stopped by boiling for 3 min. The mixtures were lyophilised and taken up in 100 µl of the HPLC mobile phase. Radioactive substrate and product were separated on a Vydac 214TP column (150×4.6 mm) in water/methanol/isopropanol (9:8:8, v/v) containing 20 mM ammonium acetate. The flow rate was 0.6 ml min−1. Detection was performed with a radioactive flow detector (model LB506-C1, Berthold France, La Garenne-Colombes, France) using the Quicksafe Flow 2 scintillator (Zinsser Analytic, Maidenhead, UK) at 0.6 ml min−1. Quantification was carried out with the Winflow software (Berthold). The retention times of UDP-GlcNAc and lipid II analogue were 3.7 and 7.6 min, respectively. 2.7 Determination of the MurG kinetic constants The MurG activity was assayed as described above with various concentrations of one substrate (10–100 µM for UDP-GlcNAc, 1.6–16 µM for 1) while maintaining the other at a fixed value (20 µM for 1, 25 µM for UDP-GlcNAc). Data were fitted to the equation v=VA/(K+A) using the MDFitt software developed by M. Desmadril (UMR 8619 CNRS, Orsay, France). 2.8 Reverse MurG reaction The experiments were carried out in conditions similar to those of the forward reaction except that the substrates were UDP (16 µM or 1.6 mM) and radiolabelled 4 (16 µM, 1.7 kBq). The amount of radioactive 4 decreased whereas a radioactive product appeared at 3.7 min. The newly formed compound was further characterised as UDP-GlcNAc in another HPLC system, a Nucleosil 100 C18 5 µm column (150×4.6 mm) in 50 mM triethylammonium formate, pH 4.55, at 0.6 ml min−1 (retention time: 24.5 min, identical with that of authentic UDP-[14C]GlcNAc; in this system, 4 was not eluted from the column). 3 Results In a previous work [11], we synthesised 1 (Fig. 1) as a lipid I analogue. Compound 1 differed from natural lipid I by (i) the replacement of the undecaprenyl group (C55) by the shorter α-dihydroheptaprenyl one (C35); (ii) the absence of double bond in the α-terminal isoprene residue; (iii) the dansylation of the Nɛ-amino group of the meso-diaminopimelyl residue in the peptide. We showed that 1 was a substrate for MurG: indeed, the enzyme was able to catalyse the transfer of the [14C]GlcNAc moiety from UDP-[14C]GlcNAc to 1 to yield lipid II analogue 4 (Fig. 1). The assay conditions were similar to those of a membrane-based assay [10], but membranes were omitted. The reaction mixture contained 0.1% (w/v) deoxycholate to increase the solubility of the lipid. Radioactive substrate (UDP-GlcNAc) and product (4) were separated by paper chromatography and quantitated with a radioactivity scanner. In the present work, we improved this enzymatic assay by (i) defining the optimal conditions for the reaction; (ii) using HPLC as a separation method. Large deviations among duplicate or triplicate assays had led us to suspect insufficient solubility of 1; this was confirmed by the turbid aspect of the reaction mixtures. We therefore replaced deoxycholate by an organic solvent. Methanol, trifluoroethanol and DMSO were tested. Whereas the two former inhibited the enzyme even at low concentration, the latter was compatible with the activity (Table 1). Very interestingly, the enzymatic activity was greatly enhanced in the presence of DMSO, the optimal proportion being ∼35% (v/v). At 35% DMSO, the activity was 120-fold that obtained without DMSO, and 10-fold that obtained in our previous conditions (without DMSO and with 0.1% deoxycholate). 1 Open in new tabDownload slide Formulae of the lipid analogues synthesised or mentioned in this study. 1 Open in new tabDownload slide Formulae of the lipid analogues synthesised or mentioned in this study. 1 Optimisation of the parameters of the MurG assay Parameter Range studied Optimum found Value selected DMSO content (%, v/v) 0–60 30–40 35 pH 7–9 7.5 7.5 Tris concentration (M) 0.05–0.3 0.2–0.3 0.2 MgCl2 concentration (mM) 1–40 5–20 10 Parameter Range studied Optimum found Value selected DMSO content (%, v/v) 0–60 30–40 35 pH 7–9 7.5 7.5 Tris concentration (M) 0.05–0.3 0.2–0.3 0.2 MgCl2 concentration (mM) 1–40 5–20 10 Open in new tab 1 Optimisation of the parameters of the MurG assay Parameter Range studied Optimum found Value selected DMSO content (%, v/v) 0–60 30–40 35 pH 7–9 7.5 7.5 Tris concentration (M) 0.05–0.3 0.2–0.3 0.2 MgCl2 concentration (mM) 1–40 5–20 10 Parameter Range studied Optimum found Value selected DMSO content (%, v/v) 0–60 30–40 35 pH 7–9 7.5 7.5 Tris concentration (M) 0.05–0.3 0.2–0.3 0.2 MgCl2 concentration (mM) 1–40 5–20 10 Open in new tab The beneficial effect of DMSO being established, we optimised the other parameters of the assay, namely the pH value and the concentrations of Tris and MgCl2 (Table 1). According to the results, the following conditions were selected: 0.2 M Tris–HCl buffer, pH 7.5, 10 mM MgCl2. The next step of the study was the search for HPLC conditions allowing the rapid separation of radioactive 4 from residual UDP-[14C]GlcNAc. This was accomplished with a C4 column and isocratic elution with a water/methanol/isopropanol mixture. The duration of the run was 17 min. Using this assay, the kinetic parameters of the MurG reaction were determined. The Km values for UDP-GlcNAc and 1 were 150±20 µM and 2.8±1.0 µM, respectively. It is important to mention that the Km for 1 is an apparent Km since, owing to the huge difference between the two constants, it was determined at subsaturating concentration (25 µM) of UDP-GlcNAc, the labelled substrate. The kcat of the reaction was 56±5 min−1. Since 1 contains a dansyl group, it was logical to attempt the separation of fluorescent 1 and 4 in order to set up a nonradioactive assay. However, the two compounds co-eluted whatever the column and the solvent conditions used. It is obvious that the addition of the GlcNAc residue to 1 is not sufficient to modify its polarity, which is essentially influenced by the contribution of the bulky, hydrophobic prenyl and dansyl groups. The MurG reaction was used to prepare 4, the lipid II analogue, from 1. Since 1 and 4 could not be separated by HPLC, it was essential to operate in conditions providing a total consumption of 1. These conditions were fulfilled by using an excess of UDP-GlcNAc over 1 and a large amount of enzyme. Using radioactive 4, we showed that MurG catalyses the reverse reaction, namely the production of radioactive UDP-GlcNAc: at 16 µM of both UDP and 4, the rate was 57-fold lower than that of the forward reaction at 16 µM of both UDP-GlcNAc and 1. When the UDP:4 molar ratio was raised to 100 (i.e. at 1.6 mM UDP), the rate of the reverse reaction was increased 36-fold, thus becoming significant. The natural substrate of the reaction bearing a C55 prenyl chain, we undertook the synthesis of 5, a lipid I analogue possessing the α-dihydroundecaprenyl moiety. Although the conditions were identical with those of the synthesis of 1, the yield was much lower (3.6 vs. 27%). Surprisingly, this compound behaved as a very poor substrate for MurG: in the conditions of the standard assay, the rate with 5 was 19-fold lower than with 1. The increase of the DMSO content (47%) and/or the addition of phosphatidylglycerol (100 µg ml−1) only resulted in slight improvement (ca. 1.5-fold). 4 Discussion Immediately preceding the polymerisation reactions, the MurG enzyme catalyses the last step of synthesis of the monomer unit of peptidoglycan. It has recently focussed interest as a potential target for new antibacterial compounds [4,15,16]. As mentioned in Section 1, most currently available assays rest upon the membrane content of lipid I, which is not known with accuracy; moreover, some of these assays involve other enzymatic steps (MraY, transglycosylases, transpeptidases [6–9]). It is therefore of the utmost interest to set up specific MurG assays, such as the one described here, relying upon substrates whose concentration is easily quantitable. The lipid I analogue 1 used can be prepared from natural UDP-MurNAc-pentapeptide by semisynthesis [11]. For reasons of solubility, it is important that the assay mixture contains a high DMSO concentration (35%). The beneficial effect of this solvent has already been reported for membrane-based assays at lower concentrations (4–10% [6–9]), but these assays involve several simultaneous reactions and not MurG alone. The isocratic HPLC separation of UDP-GlcNAc and 4 as well as the use of an automatic injector and of a radioactive flow detector render the assay easy and rapid. Walker and co-workers [4] described a MurG assay based on lipid I analogue 2, which possesses citronellol (C10) as the prenyl chain (Fig. 1). This compound, which was water-soluble, did not necessitate the presence of an organic solvent in the assay mixture. However, we feel that the conditions of our assay, which uses a hydrophobic substrate, are closer to the in vivo MurG functioning conditions. Interestingly, the apparent Km for 1 (2.8 µM) is 13-fold lower than the Km for 2 (37 µM); the kcat value determined in our conditions (56 min−1) is higher than Walker's kcat (16 min−1). Conversely, the Km for the hydrophilic substrate UDP-GlcNAc is higher in our case (150 vs. 58 µM). These variations of the kinetic parameters may reflect differences (hydrophobicity of the lipid substrate, polarity of the medium) in the assay conditions. In some extreme cases, these differences can lead to discrepancies: for instance, Silva et al. pointed out that 1-phospho-MurNAc-pentapeptide inhibits MurG in Walker's assay, but does not in a membrane-based assay; they concluded that hydrophilic 1-phospho-MurNAc-pentapeptide cannot compete for the active site of membrane-bound MurG [16]. In the present work, we tried to evaluate the effect of the length of the prenyl chain by synthesising 5, which is closer to natural lipid I since it contains a C55 prenol. However, we were confronted to a poor yield of synthesis and, mainly, to a low reactivity as a MurG substrate. This is in agreement with the observations of Walker and colleagues, who ascertained the low in vitro reactivity of lipid I and attributed it to its propensity to aggregate [17]. Lately, this group of investigators tested a series of analogues of lipid I with different prenyl chains (C10 to C55). In the presence of 15% methanol, the best substrate found was nerylneryl (C20) analogue 3, i.e. a compound much more soluble in water than lipid I. Triton X-100 (0.2%) was found to stimulate the activity by a factor of 10 [13]. In our standard assay, we found a 3-fold activation by 0.2% Triton X-100. However, methanol (15–40%, with or without Triton X-100) almost totally inhibited MurG when replacing DMSO. We have no satisfactory explanation for this discrepancy with Walker's results. Using MurG as a catalyst, we synthesised 4, the lipid II analogue originating from 1. Radioactive 4 is of potential interest for the study of the next step of peptidoglycan biosynthesis, namely the transglycosylation reaction. In the present work, we showed, using radioactive 4, that the reverse MurG reaction can take place in the standard conditions, although at a low rate. The reversibility of MraY, the enzyme catalysing the preceding step in the biosynthetic pathway, was observed 30 years ago; in that case, the reverse reaction was favoured in vitro [18]. To the best of our knowledge, the present results constitute the first report of the reversibility of the MurG reaction. Acknowledgements This work was supported by grants from the Centre National de la Recherche Scientifique (UMR 8619) and the Ministère de l'Education Nationale, de la Recherche et de la Technologie (Biotechnologies). References [1] van Heijenoort J. ( 2001 ) Nat. Prod. Rep. 18 , 503 – 519 . Crossref Search ADS PubMed [2] Bupp K. van Heijenoort J. ( 1993 ) J. Bacteriol. 175 , 1841 – 1843 . Crossref Search ADS PubMed [3] Crouvoisier M. Mengin-Lecreulx D. van Heijenoort J. ( 1999 ) FEBS Lett. 449 , 289 – 292 . Crossref Search ADS PubMed [4] Ha S. Chang E. Lo M.C. Men H. Park P. Ge M. Walker S. ( 1999 ) J. Am. Chem. Soc. 121 , 8415 – 8426 . Crossref Search ADS [5] Ha S. Walker D. Shi Y. Walker S. ( 2000 ) Protein Sci. 9 , 1045 – 1052 . Crossref Search ADS PubMed [6] Barbosa M.D.F.S. Ross H.O. Hillman M.C. Meade R.P. Kurilla M.G. Pompliano D.L. ( 2002 ) Anal. Biochem. 306 , 17 – 22 . Crossref Search ADS PubMed [7] Brandstrom A.A. Midha S. Goldman R.C. ( 2000 ) FEMS Microbiol. Lett. 191 , 187 – 190 . Crossref Search ADS PubMed [8] Brandstrom A.A. Midha S. Longley C.B. Han K. Baizman E.R. Axelrod H.R. ( 2000 ) Anal. Biochem. 280 , 315 – 319 . Crossref Search ADS PubMed [9] Chandrakala B. Elias B.C. Mehra U. Umapathy N.S. Dwarakanath P. Balganesh T.S. de Sousa S.M. ( 2001 ) Antimicrob. Agents Chemother. 45 , 768 – 775 . Crossref Search ADS PubMed [10] Mengin-Lecreulx D. Texier L. Rousseau M. van Heijenoort J. ( 1991 ) J. Bacteriol. 173 , 4625 – 4636 . Crossref Search ADS PubMed [11] Auger G. Crouvoisier M. Caroff M. van Heijenoort J. Blanot D. ( 1997 ) Lett. Pept. Sci. 4 , 371 – 376 . [12] Men H. Park P. Ge M. Walker S. ( 1998 ) J. Am. Chem. Soc. 120 , 2484 – 2485 . Crossref Search ADS [13] Chen L. Men H. Ha S. Ye X.Y. Brunner L. Hu Y. Walker S. ( 2002 ) Biochemistry. 41 , 6824 – 6833 . Crossref Search ADS PubMed [14] Therisod H. Labas V. Caroff M. ( 2001 ) Anal. Chem. 73 , 3804 – 3807 . Crossref Search ADS PubMed [15] Cudic P. Behenna D.C. Yu M.K. Kruger R.G. Szewczuk L.M. McCafferty D.G. ( 2001 ) Bioorg. Med. Chem. Lett. 11 , 3107 – 3110 . Crossref Search ADS PubMed [16] Silva D.J. Bowe C.L. Brandstrom A.A. Baizman E.R. Sofia M.J. ( 2000 ) Bioorg. Med. Chem. Lett. 10 , 2811 – 2813 . Crossref Search ADS PubMed [17] Ye X.Y. Lo M.C. Brunner L. Walker D. Kahne D. Walker S. ( 2001 ) J. Am. Chem. Soc. 123 , 3155 – 3156 . Crossref Search ADS PubMed [18] Neuhaus F.C. ( 1971 ) Acc. Chem. Res. 4 , 297 – 303 . Crossref Search ADS © 2003 Federation of European Microbiological Societies
TI - A MurG assay which utilises a synthetic analogue of lipid I
JF - FEMS Microbiology Letters
DO - 10.1016/S0378-1097(02)01203-X
DA - 2003-02-01
UR - https://www.deepdyve.com/lp/oxford-university-press/a-murg-assay-which-utilises-a-synthetic-analogue-of-lipid-i-VCdOXTZ6w9
SP - 115
EP - 119
VL - 219
IS - 1
DP - DeepDyve
ER -