Inhibitors of Glucosamine-6-Phosphate Synthase - Europe PMC

Vol. 35, No. 1

ANTIMICROBIAL AGENTS AND CHEMOTHERAPY, Jan. 1991, p. 36-43

0066-4804/91/010036-08$02.00/0 Copyright t 1991, American Society for Microbiology

Mechanism of Action of Anticandidal Dipeptides Containing Inhibitors of Glucosamine-6-Phosphate Synthase SZAWOMIR

MILEWSKI,'* RYSZARD ANDRUSZKIEWICZ,l LESZEK KASPRZAK,1 JAN MAZERSKI,1

EDWARD BOROWSKI' Department of Pharmaceutical Technology and Biochemistry, Technical University of Gdan'sk, 11112 Majakowskiego Street, 80-952 Gdan'sk, Poland,' and Department of Cellular BiologyFIORENZO MIGNINI,2

AND

Section of Microbiology, University of Camerino, 62032 Camerino, Italy2 Received 4 June 1990/Accepted 15 October 1990

The mechanism of anticandidal action of novel synthetic dipeptides containing N3-(4-methoxyfumaroyl)-L2,3-diaminopropanoic acid (FMDP) residues was shown to be consistent with the "warhead delivery" concept. FMDP dipeptides were shown to be transported into Candida albicans cells by the di-tripeptide permease and subsequently hydrolyzed by intracellular peptidases, especially aminopeptidase. The anticandidal activity of the particular FMDP dipeptide was influenced by the rate of its transport and, to a lower extent, by the intracellular cleavage rate. A high transport rate accompanied by a high cleavage rate resulted in the high anticandidal activity of L-norvalyl-FMDP. The strong growth-inhibitory effect of this compound was the consequence of inhibition of the enzyme glucosanine-6-phosphate synthase by the released FMDP. The action of L-norvalyl-FMDP on exponentially growing C. albicans cells resulted in a sharp decrease of incorporation of 14C label from [14C]glucose into chitin, mannoprotein, and glucan. This effect, as well as the growth-inhibitory effect, was fully reversed by exogenous N-acetyl-D-glucosamine. Glucosamine-6-phosphate synthase was proved to be the only essential target for FMDP dipeptides. Scanning electron microscopy of C. albicans cells treated with L-norvalyl-FMDP revealed highly distorted, wrinkled, and collapsed forms. Cells formed long, bulbous chains, and partial lysis occurred.

constitute a relatively large group of compounds, we have decided to perform some experiments in order to establish the factors determining their biological activity.

The concept of the rational utilization of peptide transport systems present in microorganisms for the delivery of toxic, impermeable amino acids has recently been described by many authors (22, 32, 34). The idea is based upon a relatively low selectivity of microbial peptide permeases. Ames et al. (1) and Fickel and Gilvarg (15) were the first to formulate the concept of "illicit" or "portage" transport. Thus far, many examples of antimicrobial agents whose mechanism of action follows this general pattern have been described. They include some antifungal antibiotics: tetaine (bacilysin) (11, 18), chlorotetaine (34), and compound A 19009 (Lindenbein) (30). Moreover, several synthetic antimicrobial agents have been designed which follow the "warhead delivery" concept: alaphosphin and related antibacterial phosphonopeptides (5), m-fluorophenylalanyl peptides (19), pyrimidinepeptide conjugates (20, 39), L-arginyl-X-L-phenylalanine (26), and epoxypeptides (10). Recently, we have described the synthesis and some biological properties of antifungal dipeptides containing a

MATERIALS AND METHODS

Microorganism and growth conditions. Candida albicans ATCC 26278 was used throughout this work. Cells were maintained on Sabouraud agar slants and transferred monthly. Cells were propagated on YNB modified medium containing 1.7 g of YNB without amino acids and ammonium sulfate (Difco, Detroit, Mich.), 0.4 g of sodium glutamate, 10 mg of L-histidine, 20 mg of DL-methionine, 20 mg of DLtryptophan, and 10 g of D-glucose in 1,000 ml of distilled water. Synthesis of 4-ABDA. To synthesize 4-azidobenzoyldialanine (4-ABDA), L-Ala-L-Ala (320 mg, 2 mmol) was treated with the N-hydroxysuccinimide ester of 4-azidobenzoic acid (520 mg, 2 mmol) (16) in dioxane-water (2:1) containing sodium bicarbonate (420 mg, 5 mmol). After 12 h at room temperature, the solvent was partially evaporated (up to 4 ml). The mixture was chilled to 4°C and acidified to pH 2 with concentrated HCl. The resulting precipitate was filtered off and recrystalized from water. Finally, 0.15 g of the white solid was obtained (yield, 23%; mp, 170 to 176°C). The substance was homogeneous by thin-layer chromatography (TLC; solvent systems A, B, and C, specified below). The UV spectrum of the product dissolved in 50 mM KH2PO1 (pH 4.5) showed a maximum at 271 nm (e = 2.1 x 104 Mcm-'). The infrared spectrum included a strong azide band at 2,130 cm-1. Other chemicals. FMDP dipeptides, FMDP, N"-AcFMDP, and FMDP-OMe were synthesized as described previously (2, 3). 2,3-Dihydroxybenzoyl-L-alanyl-L-threonine (DBAT) was synthesized by the published procedure

glutamine analog. N3-(4-methoxyfumaroyl)-L-2,3-diaminopropanoic acid (FMDP) (3). This amino acid was previously shown to be a strong inactivator of the enzyme L-glutamine: D-fructose-6-phosphate amidotransferase (GlcN-6-P synthase; EC 2.6.1.16 [2, 9, 28]). One of the FMDP dipeptides, L-norvalyl-FMDP (Nva-FMDP), was also shown to be an effective antifungal agent in the mouse model of systemic candidosis (29). Although the general idea of portage transport is well known, there are very few data on the rational design of synthetic anticandidal agents of this type. In our opinion, this is because of the poor diversity of known structures, making generalization impossible. Since FMDP dipeptides *

Corresponding author. 36

VOL. 35, 1991

(31). Lyticase and chitinase were from Sigma, St. Louis, Mo. [U-"'Cl-D-glucose solution (0.22 mCi/ml) was from UVVVR, Prague, Czechoslovakia. Other chemicals were of the finest grade commercially available. All amino acid residues are in the L-configuration unless otherwise stated. Susceptibility testing. The MICs of FMDP dipeptides were determined in twofold dilution series in YNB modified medium inoculated with 5 x 104 CFU/ml from the overnight culture on Sabouraud medium. Test tubes were incubated at 37°C for 24 h. The MIC was defined as the lowest concentration of a given FMDP peptide that prevented visible growth after 24 h of incubation. Cells from the overnight culture on Sabouraud medium were harvested, washed with sterile saline, and suspended in YNB modified medium (105 CFU/ml). The cell suspension was incubated at 37°C with gentle shaking. After 2 h of preincubation, the cell density was adjusted to 0.100 by diluting with fresh medium. FMDP peptides and/or any other compounds were added, and the incubation was continued. The growth rate was monitored turbidimetrically at 660 nm. For the determination of fungicidal effect, 1-ml samples were withdrawn every 3 h. Cells were harvested, washed twice with saline, and suspended in 1 ml of saline. Samples were diluted 100-fold, and 0.1-ml portions were plated on Sabouraud dextrose agar. Plates were incubated at 37°C for 24 h, and colonies were counted. Transport studies. Conditions for determination of the peptide transport rate were described previously (27). Photoaffinity inhibition of peptide transport was conducted exactly as described by Sarthou et al. (36). Preparation of crude extract. Cells from the overnight culture on Sabouraud medium were harvested by centrifugation at 4°C and washed twice with cold water. Subsequently, cells were transferred to a chilled mortar, mixed withi Alcoa A-305 (2 g of Alcoa per 1 g [wet weight] of cells), and allowed to freeze at - 18°C for 2 h. Frozen cells were disrupted by grinding until the mixture had become sticky. Then, proteins were extracted twice with small portions of 50 mM potassium phosphate buffer, pH 6.5 (containing 1 mM EDTA if GlcN-6-P synthase activity was to be assayed). The broken-cell suspension was centrifuged (27,000 x g, 15 min, 4°C). The supernatant (cell extract) was treated with 1% protamine sulfate solution added dropwise with vigorous shaking (1 ml of protamine sulfate solution per 75 mg of protein in the cell extract). The suspension was centrifuged, and the supernatant was discarded. The precipitate was washed with potassium phosphate buffer and mixed with a small volume of 0.1 M PP1 buffer, pH 6.5. The suspension was stirred for 30 min and subsequently centrifuged (27,000 x g, 15 min, 4°C). The precipitate was discarded, and the supernatant was saved as a PPi extract. The GlcN-6-P synthase activity was assayed by the previously described procedure (18). Determination of cleavage rate. The incubation mixtures, consisting of 10 ml of 200 ,uM peptide solution in 50 mM potassium phosphate buffer, pH 6.5, and 2 ml of diluted crude or PP1 extract (final protein concentration, 0.1 to 0.2 mg/ml), were incubated at 37°C. At 5-min intervals, 2-ml samples were withdrawn, heated at 90°C, and centrifuged, and the concentration of amino acids in the supernatant was determined by the Cd-ninhydrin procedure (12). Incorporation of ["4C]glucose into cell wall polysaccharides. Cells from the exponential phase of growth were suspended in 0.1 M Tris hydrochloride buffer, pH 7.2, containing 0.5 M MgSO4 and 5 mM glucose. Each flask contained a known number of cells (about 108/ml), 5.5 ,uCi of ["4C]glucose, and

MECHANISM OF ACTION OF FMDP DIPEPTIDES

37

antifungal agents at appropriate concentrations, in 40 ml of medium. Incubation was carried out at 37°C for 3 h. At hourly intervals, 10-ml samples of cell suspensions were withdrawn. Cells were centrifuged, and the pellet was washed twice with 20-ml portions of unlabeled medium and once with Tris hydrochloride buffer. The distribution of 14C label into the various polysaccharide fractions was studied by a slightly modified method of Gopal et al. (17). Pellets were extracted with 1 M NaOH at 90°C for 1 h. Mannoproteins were precipitated with Fehling solution (0°C, 18 h) (17) and then dissolved in 1 ml of 0.1 M potassium phosphate buffer, pH 6.0. Alkali-insoluble fractions were suspended in 2-ml portions of 0.1 M potassium phosphate buffer, pH 6.0, and divided into two parts. One part was treated with 1 mg of lyticase, and the other was treated with 1 mg of chitinase. Enzymatic hydrolyses were carried out at 37°C for 12 h. Afterwards, the insoluble residues were removed by centrifugation, and the supernatants were analyzed for radioactivity. Preparation of cells for scanning electron microscopy. Cells were centrifuged, washed three times with saline, and then fixed in a 3% solution of glutaraldehyde in 0.1 M phosphate buffer, pH 7.4 for 2 h at 4°C. After rapid washing with the buffer solution, the samples were dehydrated in ascending concentrations of ethanol up to absolute ethanol, followed by acetone. Finally, a drop of the cell suspension in acetone was deposited on a slide, dried in the air, shadowed in SEM Coating Unit PS 3, and observed in a Stereo Scan 200 Cambridge microscope at 20 kV. Other methods. Protein concentration was measured by the method of Lowry et al., with modifications as reported by Layne (21), using bovine serum albumin as a standard. TLC was performed on silica gel-coated aluminum sheets. Chromatograms were developed in three solvent systems: A, n-butanol-acetic acid-water (4: 1: 1); B, ethyl acetateethanol-water (5:1:0.75); and C, chloroform-ethanol (2: 1). Amino acids were located by ninhydrin staining, and FMDPcontaining compounds were located by quenching of UV light. RESULTS Factors influencing the anticandidal activity of FMDP dipeptides. The MICs of 21 FMDP dipeptides against C. albicans ATCC 26278 are summarized in Table 1. The table also contains data for initial velocities of transport (V,) and intracellular hydrolysis (Vc) for each dipeptide. The data were subjected to statistical analysis in order to determine the correlation coefficients for the relationships -log MIC = f (V,) and -log MIC = f (Vc), respectively. The equations of the regression lines were found to be as follows: -log MIC = 0.92 (±0.33) V, - 1.60 (±0.47), n = 18, SD = 0.42, r = 0.83; -log MIC = 0.07 (±0.03) Vt - 1.28 (+0.45), n = 18, SD = 0.47, r = 0.77. In both cases statistically significant correlation was found. On the other hand, a higher slope coefficient obtained for the activity-transport relationship (0.92 versus 0.075) indicated that the velocity of transport of a particular FMDP dipeptide was a more significant determinant of its anticandidal activity than the velocity of intracellular cleavage. Transport of FMDP dipeptides. TLC analysis of the spent medium, i.e., aliquots obtained during transport rate determinations (data not shown), revealed that FMDP peptides were rapidly and continuously taken up by C. albicans cells and neither the constitutive amino acids nor any other ninhydrin-positive degradation products were effluxed. The

38

MILEWSKI ET AL.

ANTIMICROB. AGENTS CHEMOTHER.

TABLE 1. Anticandidal activity and initial rates of transport and intracellular cleavage of FMDP dipeptides Peptide Peptide

rate Transport (nmol/min/mg)

(TnranOs5molmirn/rmg)

Intracellular

cleavage rate (nmollmin/mg)

MIC (g/ml)

1.0 1.3 2.4 2.0 1.8 2.0 1.8 2.1 0.8 0.4 3.9 0.6 0.7 0.2 1.1 1.3 1.2 1.1 1.4 1.0 0.2

0.5 12.5 17.5 25.3 9.1 17.8 15.4 9.2 14.2 6.7 0.5 14.2 0.5 0.8 10.2 15.5 16.7 23.2 5.0 7.5 0.5

50 3.12 0.78 0.20 0.78 0.20 0.78 0.78 3.12 12.5 6.25 >200 >200 100 1.56 3.12 3.12 1.56 3.12 3.12 12.50

TABLE 3. Photoinactivation of peptide transport by 4-ABDA Peptide

UV irradiation

Ala-Ala Gly-FMDP Ala-FMDP Abua-FMDP Nva-FMDP Val-FMDP

Nlea-FMDP Leu-FMDP Met-FMDP Phe-FMDP Tyr-FMDP Lys-FMDP D-Ala-FMDP D-Ala-D-FMDP FMDP-Gly FMDP-Ala FMDP-Nva FMDP-Val FMDP-Leu FMDP-Met FMDP-Phe FMDP-Tyr a Abu, L-a-Aminobutyric acid; Nle, L-norleucine.

kinetic parameters of transport of a number of FMDP dipeptides and some reference peptides were determined and are summarized in Table 2. The kinetic values were obtained from the use of the integrated rate equation and Lineweaver-Burk analysis of initial rate data. The maximal velocity (Vma,) values for FMDP dipeptide transport were similar to those of the reference dipeptides. On the other hand, the Km values, reflecting affinity for the transport system, were found to be considerably higher. This resulted in low values of uptake efficiency (VmaxlKm). In order to identify the peptide permease responsible for the transport of FMDP dipeptides, experiments on selective inhibition by the photoaffinity label 4-ABDA were conducted (Table 3). The uptake of 4-ABDA into C. albicans cells (no UV irradiation) was antagonized by tri- and tetrapeptides but not by dipeptides. Therefore, 4-ABDA was shown to be transported preferentially by the oligopeptide permease. The partial inactivation of this permease by 4-ABDA (UV irradiation) had only a slight effect on Nva-FMDP and Ala-Ala uptake, whereas the uptake of tri- and tetrapeptides was strongly reduced. TABLE 2. Kinetic constants for peptide uptakea Peptide

Km (,LM)

Ala-Ala Met-Leu Phe-Leu Lys-Leu Nva-FMDP

22.2 23.2 25.0 15.0 130.0 160.0 62.2 220.0

Lys-FMDP Phe-FMDP FMDP-Leu

V,,a (nmol/ min/mg) 10.0 6.7 2.0 8.5 3.7 6.7 1.1 6.7

Ala-Ala-Ala +

+

Nva-FMDP + a

In the presence of 0.5 mM 4-ABDA.

Transport of Nva-FMDP was affected by some metabolic inhibitors, including NaN3 and N-ethylmaleimide (Table 4). In contrast, dicyclohexylocarbodiimide, sodium arsenite, and phenylmethylsulfonyl fluoride (PMSF) had little, if any, effect on this process. The transport rate was reduced when glucose was omitted from the incubation medium. Intracellular cleavage of FMDP dipeptides. Incubation of Nva-FMDP (100 ,uM) with the cell extract of C. albicans ATCC 26278 (1 mg of protein per ml) at 37°C and pH 6.5 resulted in very fast hydrolysis of the peptide bond. The TLC analysis of deproteinized samples, withdrawn from the incubation mixtures, revealed that ninhydrin-positive spots corresponding to Nva and FMDP appeared as quickly as 30 to 45 s (data not shown). The constituent amino acids were the only products of degradation. Cleavage was inhibited by some enzyme inhibitors and accelerated by some divalent cations (Table 5). Strong inhibition was observed for Zn2+ ions, N-ethylmaleimide, PMSF, and DBAT. On the other hand, Co2+ and Mn2+ cations behaved as activators of hydrolysis of Nva-FMDP. GlcN-6-P synthase present in cell and PP1 extracts obtained from C. albicans cells was strongly inhibited by FMDP with a 50% inhibitory concentration (IC50) of 4 ,uM. FMDP derivatives, i.e., FMDP dipeptides, Na-Ac-FMDP, and FMDP-OMe, showed generally lower inhibitory potency (Table 6). The relative enzyme-inhibitory ability of X-FMDP dipeptides was about 10 to 20 times higher for GlcN-6-P synthase from cell extract than for partially purified enzyme. In contrast, for FMDP-X dipeptides, this difference was considerably lower. It should be noted that the general peptidase activity of the PPi extract, as measured by the Cd-ninhydrin method, was 21 times lower than that of the cell extract. N-acetylation of FMDP resulted in complete loss of enzyme-inhibitory activity, whereas esterification of the a-carboxyl caused only a partial decrease. The latter should not be attributed to possible esterase activity of the cell and PPi extracts, since a similar result was obtained in the presence

efficiency Uptake (Vmax/Km) 0.45 0.29 0.08 0.57 0.03 0.04 0.02 0.03

a Values are the means of three independent determinations differing by no

more than 10%o.

4 8 20 60 25 75 2 7

+

Ala-Ala-Ala-Ala

% Inhibition of initial transport rate by 4-ABDAa

TABLE 4. Effect of inhibitors on uptake of Nva-FMDP Inhibitor tested

NaN3 N-Ethylmaleimide

Sodium arsenite Dicyclohexylcarbodiimide

Concn

(mM) 0.1 10.0 1.0 1.0 1.0

% Inhibition of NvaFMDP uptake

100 100 65 10 5

VOL. 35, 1991

MECHANISM OF ACTION OF FMDP DIPEPTIDES

39

TABLE 5. Effect of metabolic inhibitors and activators on the cleavage of Nva-FMDP by intracellular peptidases in cell extract Compound tested

None (control) Zn2+ Co2+ Mn2+

N-Ethylmaleimide PMSF DBAT

(% of control)

Concn (mM)

100 56 148

1 1 1 10 1 10 1 3 0.3

191