Chen LL et al / Acta Pharmacol Sin 2004 Jul; 25 (7): 907-914

Type I methionine aminopeptidase from Saccharomyces cerevisiae is a potential target for antifungal drug screening¹

Ling-ling CHEN, Jia LI, Jing-ya LI, Qun-li LUO, Wei-feng MAO, Qiang SHEN, Fa-jun NAN³, Qi-zhuang YE^2,3

National Center for Drug Screening, Institute of Materia Medica, Shanghai Institutes for Biological Sciences, Chinese Academy of Sciences, Shanghai 201203, China;
²The High Throughput Screening Laboratory, University of Kansas, Lawrence, Kansas 66047, USA

¹ Project supported by the National Natural Science Foundation of China, No 30271528 and 39725032 and by the NIH COBRE award 1 P20 RR15563 and matching support from the State of Kansas.

³ Correspondence to Prof Qi-zhuang YE. Phn 1-785-330-4330. Fax 1-785-330-4332. E-mail qye@ku.edu

Prof Fa-jun NAN, Phn 86-21-5080-1313, ext 231. Fax 86-21-5080-0721. E-mail fjnan@mail.shcnc.ac.cn

Received 2003-09-24 Accepted 2004-02-06

KEY WORDS methionine aminopeptidase; high-throughput screening; antifungal agent

ABSTRACT

AIM: To screen antifungal drug candidates using in vitro and in vivo assays based on type I methionine aminopeptidase from Saccharomyces cerevisiae (ScMetAP1). METHODS: A colorimetric assay suitable for high throughput screening (HTS) using recombinant ScMetAP1 protein expressed in Escherichia coli was established for antifungal lead discovery. A series of pyridine-2-carboxylic acid derivatives were characterized and a chemical library of 12 800 pure organic compounds was screened with the in vitro ScMetAP1 assay. Active compounds from the in vitro assay were further evaluated by a growth inhibition assay on yeast strain with deletion of ScMetAP1 gene map1 in comparison with the wild-type yeast strain and the yeast strain with deletion of type II enzyme (ScMetAP2) gene map2. RESULTS: Active ScMetAP1 inhibitors were identified from HTS. Some of the pyridine-2-carboxylic acid derivatives (compound 2 and 3) had selective inhibition of the growth of map2 deletion yeast and weak inhibition on wild-type yeast growth, while no inhibition on map1 deletion yeast. CONCLUSION: ScMetAP1 is a novel potential target for developing antifungal drugs. The in vitro and in vivo ScMetAP1 assays can serve as tools in discovering antifungal drug candidates.

INTRODUCTION

Methionine aminopeptidase (MetAP), existing in both prokaryotic and eukaryotic cells, plays an important role in removing the first methionine from nascent polypeptides^[1,2]. Up to now, two major isoforms of MetAP (type I and type II MetAPs) have been identified^[3]. Eubacteria has only type I MetAP, and archaea has only type II MetAP, while eukaryotic cells contain both types of MetAPs^[1]. MetAPs are organized in a similar `pita-bread' conformation as characterized by several available X-ray structures, and they are highly conservative in catalytic domain with respect to five amino acid residues coordinating two cobalt ions^[4-6]. MetAPs have been considered as Co²⁺-enzymes since MetAPs can be reproducibly activated by Co²⁺ in vitro, although recent studies showed other divalent metal ions such as Zn²⁺, Fe²⁺, Mn²⁺, and Ni²⁺ could also activate MetAPs^[7-12].

MetAP has important physiological function. Deletion of the unique MetAP gene from Escherichia coli^[13] or Salmonella typhimurium^[14] is lethal, demonstrating the essential role of MetAP in bacteria. In Saccharomyces cerevisiae, disrupting either type I or type II MetAP renders a slow growth phenotype, and removal of both is nonviable^[15]. Therefore, MetAPs are potential targets for developing antibacterial and antifungal drugs^[16]. The natural product fumagillin selectively inhibited type II MetAPs and killed yeast which lacking type I enzyme ScMetAP1, and it did not affect the growth of wild-type and ScMetAP2 deletion strain^[17], suggesting dual inhibitors for both type I and type II enzymes is required for antifungal application. However, in vivo experiments showed that ScMetAP1 in yeast played a bigger role in N-terminal methionine removal and in methionine metabolism than ScMetAP2^[18,19], indicating the development of ScMetAP1 inhibitors is more important for therapeutic application.

Although the therapeutic potentials have been recognized, there are few inhibitors against MetAPs reported in literatures. Fumagillin and its analogues are specific inhibitors for type II MetAPs with covalent modification of the enzyme^[17,20]. The substrate-like inhibitor AHHpA-Ala-Leu-Val-Phe-OMe inhibited the E coli type I enzyme EcMetAP1 with a reported IC₅₀ at 5 µmol/L^[21], and we recently discovered pyridine-2-carboxylic acid derivatives which showed potent inhibition against type I enzymes EcMetAP1 and ScMetAP1^[22]. In this study, we prepared apo-ScMetAP1 and used it in high throughput screening (HTS). The ScMetAP1 inhibitors were further tested on yeast growth with wild-type, ScMetAP1 lacking or ScMetAP2 lacking yeast strains. The information presented here would help in designing functional inhibitors against MetAP1 that were effective in vivo for therapeutic applications.

MATERIALS AND METHODS

Materials and instruments Expression vector pGEX-KG was a generous gift from Prof Kun-liang GUAN at University of Michigan (USA)^[23]. Yeast strain W303-1A (MATa ade2-1 can1-100 ura3-1 leu2-3, 112 trp-1 his3-11, 15) contains wild-type ScMetAP1 and ScMetAP2. Null ScMetAP1 (Dmap1) and null ScMetAP2 (Dmap2) strains are isogenic haploid derivatives of W303-1A containing an additional map1::HIS3 or map2::URA3 gene disruption, respectively^[24,25]. These series of yeast strains were generous gifts from Prof Yie-hwa CHANG at St Louis University.

Chromogenic thiopeptolide substrate Met-S-Gly-Phe^[26], inhibitors (3R)-amino-(2S)-hydroxyheptanoic acid (AHHpA)^[22] and pyridine-2-carboxylic acid derivatives were synthesized in this laboratory. The 12 800 compounds were all small organic compounds with unrelated and diverse chemical structures and purchased from SPECS/Bio SPECS (Rijswijk, The Netherlands) with purity higher than 90 %.

Continuous kinetic monitoring of enzyme activity was performed on SpectraMax 340 microplate reader (Molecular Devices, Sunnyvale, CA) for UV absorption and controlled by the Softmax software. Liquid handling for random screening was carried out on the ORCA automation system with a Biomek 2000 liquid handling workstation from Beckman Coulter (Fullerton, CA) and with the Hydro 96 semi-automated 96-channel pipettor from Robbins (Sunnyvale, CA). Protein purification was carried out on an Akta FPLC system with UV and conductivity monitors from Amersham Biosciences (Piscataway, NJ).

Preparation of ScMetAP1 Full-length ScMetAP1 gene was obtained by direct PCR amplification from the total genomic DNAs of the S cerevisiae strain YPH499 with the forward primer (GGAATTCTAAT GAGCACTGCAACTACAAC) with EcoR I site (italic) and the reverse primer (GTCGACTTACTATTTAATTC-TCTGTCTTGG) with Sal I site (italic). The PCR fragment was then cloned into the EcoR I and Sal I sites of pGEX-KG vector for expressing glutathione S-transferase fusion ScMetAP1 protein (GST-ScMetAP1) in strain BL21(DE3)pLysS under the control of a Tac promoter.

Protein expression and purification were carried out as described in literature with slight modifications^[27,28]. E coli cells bearing plasmid were cultured in 500 mL of LB medium in the presence of ampicillin (100 mg/L) with shaking at 37 ºC. When OD₆₀₀ reached 0.6-0.8, the expression of GST-ScMetAP1 was induced by adding isopropyl thil-b-D-galactoside (IPTG) to 0.25 mmol/L and allowed to continue shaking at 22 ºC for another 14 h. Cells (0.81 g cell paste) from the 500 mL culture were resuspended in 50 mL PBS (NaCl 140 mmol/L, KCl 2.7 mmol/L, Na₂HPO₄ 10 mmol/L, KH₂PO₄ 1.8 mmol/L) with 0.1 % Triton X-100 (Bio-Rad) and sonicated on ice for 2 min. After centrifugation at 16 000×g for 15 min twice, the filtered supernatant was loaded to a 5 mL Glutathione Sepharose 4B column (Pharmacia) that previously equilibrated with 50 mL PBS. The loaded column was washed with 200 mL PBS and the bound active protein was eluted with glutathione 10 mmol/L, Tris 50 mmol/L, pH 8.0. The elution was then exchanged to Tris 50 mmol/L, pH 8.0, NaCl 150 mmol/L, CaCl₂ 2.5 mmol/L buffer for thrombin cleavage^[23]. After incubated with thrombin (0.3 kU/L, Sigma) at 4 ºC for 16-20 h, the mixture was loaded to 5 mL Glutathione Sepharose 4B column again to remove free GST. To make an apo-enzyme, ScMetAP1 were incubated with EDTA 5 mmol/L overnight, and exchanged into HEPES 50 mmol/L, pH 7.5, KCl 150 mmol/L, which has been treated with Chelex resin (Bio-Rad) to remove adventurous metal ions. Apo-ScMetAP1 (3.78 mg) showed a single band on 12 % SDS-PAGE gel by Coomassie Brilliant Blue staining.

Protein concentration determination and metal analysis Protein concentration was determined by the Bradford method with BSA as the standard. Metal analysis of the apo-ScMetAP1 solution was carried out by inductively coupled plasma-atomic emission spectrometry (ICP-AES) (performed by Shanghai Institute of Measurement and Testing Technology).

ScMetAP1 activity assays The ScMetAP1 activity was detected by a colorimetric assay with the thiopeptolide Met-S-Gly-Phe as substrate as described previously^[26] with modifications for reading on microplates. The assay was usually performed at room temperature on a 96-well clear polystyrene microplate. In metal activation experiments, the assay mixture in each well containing MOPS 50 mmol/L, pH 7.0, DTNB 1 mmol/L, Met-S-Gly-Phe 100 µmol/L, apo-ScMetAP1 0.84 µmol/L and various amounts of divalent metal ions (CoCl₂, MnCl₂, or ZnCl₂). The hydrolysis of Met-S-Gly-Phe was monitored continuously by change of UV absorbance at 412 nm, and the initial rate of the hydrolysis was determined using the early linear region of the enzymatic reaction curve. The rate for background hydrolysis of the thiopeptolide was subtracted.

ScMetAP1 inhibitors screening A total of 12 800 pure organic compounds were used for HTS against Co²⁺-ScMetAP1 with the aforementioned colorimetric assay. The test compounds were solubilized in Me₂SO at 1 g/L, and then distributed 2 µL on the daughter plates. The Me₂SO (2 µL) as the test vehicle was distributed in eight blank cells. The enzymatic assay was carried out on 96-well clear polystyrene plates in a 100 µL volume in each well and immediately before the assay, apo-ScMetAP1 2.1 µmol/L was mixed with CoCl₂ 250 µmol/L. An assay mixture (78 µL) containing MOPS 50 mmol/L, pH 7.0, DTNB 1 mmol/L, Met-S-Gly-Phe 100 µmol/L, was added to each well with Robbins liquid-handling system, and then added 20 µL of the Co²⁺ activated ScMetAP1 (final concentrations: ScMetAP1 0.42 µmol/L and CoCl₂ 50 µmol/L). The hydrolysis of Met-S-Gly-Phe was monitored on SpectraMax 340 microplate reader at 412 nm for 1 min at room temperature.

Kinetic characterization of metal-substituted ScMetAP1s and their inhibitors Metal-substituted ScMetAP1s were characterized by determining their kinetic parameters k_cat, K_m, and k_cat/K_m values using the chromogenic substrate Met-S-Gly-Phe. A 100 µL assay mixture containing MOPS 50 mmol/L, pH 7.0, DTNB 1 mmol/L, apo-ScMetAP1 0.42 µmol/L, CoCl₂ 50 µmol/L, MnCl₂ 50 µmol/L or ZnCl₂ 30 µmol/L and Met-S-Gly-Phe in two fold dilutions up to 5 mmol/L was used to obtain the k_cat, K_m, and k_cat/K_m values. The k_cat and K_m values of the metal-substituted ScMetAP1s were derived from a non-linear regression fitting of the curve in the plot of the initial rates vs the substrate concentrations, using the Michaelis-Menton equation: v=V_max·[S]/(K _m+[S]), where v is the initial rate, V_max is the maximum rate, and [S] is the substrate concentration.

The inhibitory activities of compounds were characterized by IC₅₀ determinations at multiple inhibitor concentrations. The assay mixture in each well of a 96-well microplate containing MOPS 50 mmol/L, pH 7.0, DTNB 1 mmol/L, Met-S-Gly-Phe 100 µmol/L, and apo-ScMetAP1 0.42 µmol/L with CoCl₂ 50 µmol/L plus compound at one of the several concentrations around its estimated IC₅₀ value. The hydrolysis of Met-S-Gly-Phe was monitored continuously by change of absorbance at 412 nm, and IC₅₀ was calculated from the non-linear curve fitting of percent inhibition (% inhibition) vs inhibitor concentration [I] by using the following equation: % Inhibition=100/{1+(IC₅₀/[I])^k }, where k is the Hill coefficient.

Yeast growth inhibition assay Compounds dissolved in Me₂SO were spotted onto sterile filter disks with diameter 7 mm and then placed on YPD (1 % yeast extract, 2 % peptone, and 2 % glucose, 1 % agar) containing logarithmically growing yeast strains. After incubated at 30 ºC for 48 h, the diameters that yeast can not grow were measured, indicating the inhibitory effects of relevant compounds. Solvent Me₂SO was used as negative control, while amphotericine B (25 mg/L) and fumagillin (10 µmol/L) were both used as positive controls, in which amphotericine B showed inhibitory activity in all the three types of yeast strains (wild, Dmap1 and Dmap2) and fumagillin selectively inhibited Dmap1 yeast strain which only contains ScMetAP2.

RESULTS

Preparation of ScMetAP1 Soluble fusion protein GST-ScMetAP1 was highly expressed in E coli BL21(DE3)pLysS cells, purified with affinity chromatography , and cleaved by thrombin. After flowing through affinity chromatography to remove free GST, the ScMetAP1 exhibited a single band with an apparent molecular weight (MW) about 44 kDa on the SDS-PAGE gel (Fig 1). The concentration of the purified ScMetAP1 protein was determined, and the yield of ScMetAP1 was 3-4 mg from 500 mL LB culture. When the protein was purified without adding divalent metal ions and removing adventurous metals in the buffer, the purified protein showed a basal activity. This basal activity was reduced to an undetectable level by EDTA treatment. Metal analysis of the apo-ScMetAP1 solution by ICP-AES demonstrated that each of metal ions Co, Mn, Ni, and Fe was <0.001 ppm (about 0.02 µmol/L), while the concentration of Zn was 0.007 ppm (about 0.1 µmol/L) for the 0.84 µmol/L protein solution which we used in the activation assay.

Fig 1. Expression and purification of ScMetAP1. Lane 1, protein molecular mass marker; Lane 2, pre-induction control of total cell lysate; Lane 3, post-induction whole cell lysate; Lane 4, soluble fraction of cell lysate; Lane 5, after GST-affinity chromatography; Lane 6, after thrombin cleavage and remove free GST.

Activations of ScMetAP1 by divalent metals Using the colorimetric activity assay, different divalent metal ions were added to test the stimulation of the EDTA-treated apo-ScMetAP1 and to find the appropriate metals and the concentrations for ScMetAP1 in vitro for high throughput screening. Along with the increasing amount of CoCl₂, ZnCl₂, or MnCl₂, the activity of apo-ScMetAP1 increased accordingly (Fig 2). However, CoCl₂ started to show inhibition on ScMetAP1 in a concentration above 62.5 µmol/L, ZnCl₂ started to show inhibition above 31.25 µmol/L, indicating that Zn²⁺ activation of apo-ScMetAP1 was in a relatively narrow range compared with Co²⁺. While Mn²⁺ had a broad concentration to active ScMetAP1 and did not show the maximal activation in the range we examined (1.95 to 250 µmol/L). Other divalent metal ions tested, such as Ni²⁺, Mg²⁺, Ca²⁺, Cd²⁺ or Cu²⁺, failed to active the enzyme in the range examined. So Co²⁺, Zn²⁺, or Mn²⁺ substituted apo-ScMetAP1s were further determined their efficiency for cleavage of the thiopeptolide substrate.

Fig 2. Activation of apo-ScMetAP1 by divalent metals Co²⁺ (circles), Mn²⁺ (squares) and Zn²⁺ (triangles). The enzyme activity was monitored by the hydrolysis of Met-S-Gly-Phe in a continuous colorimetric assay with apo-ScMetAP1 0.84 µmol/L.

Determination of kinetic properties of metal substituted apo-ScMetAP1 The kinetic constants for the metal-substituted enzymes were determined by hydrolyzing Met-S-Gly-Phe with the most favorable metals for ScMetAP1. Co²⁺-ScMetAP1 was an efficient enzyme in cleaving the thiopeptolide substrate with k_cat/K_m, 2894 L·mol ^-1·s^-1, Zn²⁺-ScMetAP1 showed a catalytically competent enzyme with about 60 % of Co²⁺-ScMetAP1 efficiency (k_cat/K_m, 1838 L·mol ^-1·s^-1) and Mn²⁺-ScMetAP1 was less effective in hydrolyze the thiopeptolide and showed a catalytic activity with only about 30 % of Co²⁺-ScMetAP1 activity (k_cat/K_m, 985 L· mol ^-1·s^-1) (Tab 1). The reduction of catalytic efficiency for Zn²⁺ or Mn²⁺ substituted enzymes mainly resulted from their K_m values significantly increased. These results, consistent with the previous observations that ScMetAP1 could be activated by Co²⁺, Mn²⁺, or Zn²⁺ with an HPLC assay and octapeptide^[7], showed that ScMetAP1 can be reproducibly activated by Co²⁺ in vitro and the Co²⁺-ScMetAP1 enzyme showed the most efficient catalytic activity by using Met-S-Gly-Phe as substrate. The Co²⁺ activated ScMetAP1 and the thiopeptolide assay could be used for high throughput screening.

Tab 1. Kinetic constants for the metal-substituted ScMetAP1s.

ScMetAP1s	Km (mmol/L)	kcat (s-1)	kcat/Km
			(L·mol -1·s-1)
Co2+-ScMetAP1	1.24±0.11	3.58±0.17	2895
Zn2+-ScMetAP1	6.75±0.38	12.8±0.56	1838
Mn2+-ScMetAP1	16.0±1.48	15.8±1.18	984

Assays with Met-S-Gly-Phe were carried out at room temperature with apo-ScMetAP1 0.42 µmol/L and CoCl₂ 50 µmol/L, ZnCl₂ 30 µmol/L or MnCl₂ 50 µmol/L.

High throughput screening for Co²⁺-ScMetAP1 inhibitors Compounds were solved in Me₂SO for high throughput screening, and the presence of 2 % Me₂SO in the HTS assay would not significantly affect the Co²⁺-ScMetAP1 activity with more than 80 % of the enzyme activity remained. In addition, Me₂SO solvent was used as negative control in this assay. The substrate-like bastatin-based inhibitor AHHpA-Ala-Leu-Val-Phe-OMe, with IC₅₀ around 5 µmol/L, is one of the few specific inhibitors toward EcMetAP1 as reported^[2,25] and part of the structure, the AHHpA (Fig 3, compound 1) moiety itself, was an inhibitor for EcMetAP1^[12]. We reported here that AHHpA (1) also had a moderate inhibitory activity against Co²⁺-ScMetAP1 with IC₅₀ 16.54±1.69 µmol/L by using the thiopeptolide assay and was used as positive control in this HTS assay.

Fig 3. Structures of ScMetAP1 inhibitors.

A random selected 12 800 organic compound library was tested by using the purified Co²⁺-ScMetAP1. All the compounds were initially screened at the concentrations of 20 µg/mL. A total of 83 compounds with larger than 60 % inhibition were obtained from the screening, and their inhibitory activities on ScMetAP1 were further confirmed by determining the IC₅₀ values at multiple concentrations.

Two compounds with pyridine-2-carboxylic acid moiety reported earlier^[22] showed good inhibitory activity. Compound 2 (Fig 3), which was the lead compound for structure modifications and inhibited Co²⁺-ScMetAP1 with IC₅₀ value at 7.10±0.36 mmol/L, and compound 3 (Fig 3), derived from compound 2, was one of the best Co²⁺-ScMetAP1 inhibitors with IC₅₀ at 0.46±0.11 µmol/L (Fig 4). Inhibitors with new structural types were also identified from the screening, such as compounds 4 and 5 with a 1,2,4-triazole moiety (Fig 3) with IC₅₀ values at 0.66±0.05 µmol/L and 0.99±0.04 µmol/L, respectively. Similar compounds with the 1,2,4-triazole moiety has been reported as HsMetAP2 inhibitor^[29].

Fig 4. Inhibiton of ScMetAP1 in the colorimetric assay with Met-S-Gly-Phe as the substrate in the presence of Co²⁺. Inhibitors are 1 (open circles), 2 (filled circles), 3 (filled squares), 4 (filled triangles) and 5 (open triangles). X-axis: Inhibition concentration (µmol/L), Y-axis: Relative activity (%). The colorimetric assay include apo-ScMetAP1 0.42 µmol/L; Met-S-Gly-Phe 100 µmol/L; CoCl₂ 50 µmol/L.

In vivo growth inhibition assay Some of ScMetAP1 inhibitors were further tested by an in vivo assay on yeast growth inhibition. In addition to wild-type yeast strain with two intact MetAPs (type I and type II)^[14], mutant null MetAP1(Dmap1) and null MetAP2 (Dmap2) yeast strain which only contain ScMetAP2 or ScMetAP1^[23,24] were also surveyed in this assay for further elucidating the sensitivity and selectivity of these compounds on ScMetAP1. Amphotericine B used as the positive inhibitor can inhibit all the three yeast strains with a similar inhibitory activity, and fumagillin markedly inhibited the growth of Dmap1 strain as expected due to its selective inhibition on ScMetAP2 (Tab 2).

Tab 2. Compounds in vivo inhibitory effects on wild-type, Dmap1 and Dmap2 yeast strains.

Compound	Diameter (mm) of yeast not growth
	wt	Dmap1	Dmap2
1	7.56±0.37	7	7
2	9.16±1.00	7	9.38±0.85
3	12.41±1.70	7	14.72±2.15
4	7	7.57±0.76	7
5	7.56±0.37	7	7
Amphertericine B	13.35±1.37	13.08±1.29	12.69 ± 1.26
Fumagillin	7	37.34±4.16	7
Me2SO	7	7	7

Sterile filter disks (the diameter is 7mm) impregnated with compounds 1 to 5 (10 mmol/L), fumagillin 10 µmol/L, or amphotericine B 25 µg/mL, or Me₂SO control were placed on wild-type strain W303-1A, a ScMetAP1 deletion strain, map1::HIS3, or a ScMetAP2 deletion strain, map2::URA3 of S.cerevisiae. The diameters that fungal growth inhibited by compounds were measured after 48-h incubation.

The inhibitors (compounds 1 to 5) for Co²⁺-ScMetAP1 behaved differently for the inhibitory effects on the three yeast strains. Compound 2 and compound 3 both inhibited the growth of Dmap2 yeast strain, which depended on the functional ScMetAP1 for viability. They also had weak inhibitory activities against wild-type yeast, in which the ScMetAP1 was selectively inhibited (Tab 2 and Fig 5). While Dmap1 yeast strain, which only has ScMetAP2 to maintain its growth, was resistant to these two ScMetAP1 inhibitors. These data indicated that pyridine-2-carboxylic acid derivatives showed antifungal activity by selectively inhibiting ScMetAP1 in vivo, although in high concentrations. On the contrary, the in vivo inhibitory activities of AHHpA (1), compound 4 and 5 on wild-type, Dmap2 as well as Dmap1 yeast strains were almost undetectable, even though they inhibited ScMetAP1 in vitro. It needs further studies to determine the differences between in vitro and in vivo inhibitory activities of these compounds.

Fig 5. In vivo inhibition of ScMetAP1 activity. Sterile filter disks impregnated with 10 mL compound 2, 3 (10 mmol/L) or Me₂SO (anti-clockwise from the top) were placed on wild-type strain W303 1A (left), a ScMetAP2 deletion strain, map2::URA3 (middle), or a ScMetAP1 deletion strain, map1::HIS3 S cerevisiae (right).

DISCUSSION

ScMetAP1 was expressed and purified from E coli with affinity chromagraphy and its activity was easily monitored by thiopeptolide substrate Met-S-Gly-Phe, which provided a convenient assay for further divalent metal activation investigations and large-scale random screening for inhibitor discovery. The results showed that divalent metals such as Co²⁺, Zn²⁺, and Mn²⁺ activated apo-ScMetAP1 by thiopeptolide assay was consistent with the results by using octapeptide MSSHRWDW as substrate^[7], which suggested that thiopeptolide substrate was an efficient and reliable substrate for monitoring ScMetAP1 activity. However, we could not observe the ScMetAP1 activity with Ni²⁺ under the assay, perhaps because the occupation by different metals may cause subtle change at the active site to deduce the change of specificity for substrate^[12].

Although recently studies showed that several divalent metal ions, such as Co²⁺, Zn²⁺, Fe²⁺, Mn²⁺, and Ni²⁺ could activate MetAPs^[7-12], metal substitution assay in this study indicated that Co²⁺ was the most favorite metal for ScMetAP1 activity with Met-S-Gly-Phe as substrate in vitro, so we selected Co²⁺-ScMetAP1 for the HTS assay. Through large scale screening, two groups of compounds with distinctive basal structures have been identified for Co²⁺-ScMetAP1 inhibitors. The one was compound 2 and 3 which contained pyridine-2-carboxylic acid structure with IC₅₀ value around 7.10 µmol/L and 0.46 µmol/L, respectively. And the other was compound 4 and 5 which derived from a basal 1,2,4-triazoles structure with IC₅₀ value less than 1 µmol/L.

Three types of yeast stains, named wild-type, ScMetAP1 and ScMetAP2 deletion strains, were all available to test the in vivo selectivity and sensitivity of these inhibitors described above. The pyridine-2-carboxylic acid derivatives inhibited the growth of wild-type and ScMetAP2 deletion strains with selective inhibition on ScMetAP1 in vivo, while no inhibitory effect on ScMetAP1 deletion strain were performed in this case. The inhibition of wild-type yeast growth perhaps due to ScMetAP1 plays a more significant physiological function than ScMetAP2 in vivo^[18,19], which at the same time indicated that ScMetAP1 alone could be a potential target for antifungal drug discovery.

However, pyridine-2-carboxylic acid derivatives inhibited yeast growth only occurred at a high concentration such as 10 mmol/L, and no obvious in vivo inhibition could be observed when the concentration decreased to 1 mmol/L (data not shown). It seemed that there was little relationship between the in vitro inhibitory potencies for the Co²⁺-ScMetAP1 and the in vivo inhibition against the wild-type or map2 deletion yeast strains. There are three main possibilities responsible for these differences. The first is the differences of these compounds in their ability of passing the cell wall, which we have not tested. The second is the diversity of these compounds in their solubility in the YPD culture medium, since these compounds could not be dissolved in water and all dissolved in Me₂SO for the tests. In addition, the in vivo physiological metal that located in the active site of MetAPs still remained to be defined and recent data suggested that MetAPs might have a metal other than cobalt at their active sites. We used Co²⁺-ScMetAP1 for screening and characterization of these inhibitors, and our previous data showed that the metal substitutions could bring on the changes in specificity toward substrates and inhibitors^[12]. And we are in the process of assessing these and other inhibitors on the other metal substituted ScMetAP1s with the hope of improving their in vivo potencies.

In summary, a large-scale screening assay for Co²⁺-ScMetAP1 was established and two groups of compounds were discovered. The pyridine-2-carboxylic acid derivatives selectively inhibited the growth of yeast and their in vivo target were demonstrated as ScMetAP1, while the 1,2,4-triazoles derivatives showed no evident inhibitions on yeast growths. ScMetAP1 is a potential target for antifungal drug screening, and inhibitors we discovered here need further modifications not only for the in vitro activity against ScMetAP1 but also for the in vivo efficacy on antifungal function.

ACKNOWLEDGEMENT We thank Prof Yie-hwa CHANG at St Louis University for his generous gift of yeast strains W303-1A (wild-type strain and Null ScMetAP1 and null ScMetAP2 strains) and we also thank Prof Kun-liang GUAN at Michigan University for his generous gift of expression vector pGEX-KG.

REFERENCES

• 1 Bradshaw RA, Brickey WW, Walker KW. N-terminal processing: the methionine aminopeptidase and N alpha-acetyl transferase families. Trends Biochem Sci 1998; 23:263-7.
• 2 Lowther WT, Matthews BW. Structure and function of the methionine aminopeptidases. Biochem Biophys Acta 2000; 1477: 157-67.
• 3 Arfin SM, Kendall RL, Bradshaw RA. Eukaryotic methionyl aminopeptidases: two classes of cobalt-dependent enzymes. Proc Natl Acad Sci USA 1995; 92: 7714-8.
• 4 Lowther WT, Orville AM, Madden DT, Lim S, Rich DH, Matthews BW. Escherichia coli methionine aminopeptidase: implications of crystallographic analyses of the native, mutant, and inhibited enzymes for the mechanism of catalysis. Biochemistry 1999; 38: 7678-88.
• 5 Liu S, Widom J, Kemp CW, Crews CM, Clardy J. Structure of human methionine aminopeptidase-2 complexed with fumagillin. Science 1998; 282: 1324-7.
• 6 Tahirov TH, Oki H, Tsukihara T, Ogasahara K, Yutani K, Ogata K, et al. Crystal structure of methionine aminopeptidase from hyperthermophile, Pyrococcus furiosus. J Mol Biol 1998; 284:101-24.
• 7 Walker KW, Bradshaw RA. Yeast methionine aminopeptidase I can utilize either Zn²⁺ or Co²⁺ as a cofactor: a case of mistaken identity? Protein Sci 1998; 7: 2684-7.
• 8 D'souza VM, Holz RC. The methionyl aminopeptidase from Escherichia coli can function as an iron(II) enzyme. Biochemistry 1999; 38: 11079-85.
• 9 Meng L, Ruebush S, D'souza VM, Copik AJ, Tsunasawa S, Holz RC. Overexpression and divalent metal binding properties of the methionyl aminopeptidase from Pyrococcus furiosus. Biochemistry 2002; 41: 7199-208.
• 10 D'souza VM, Swierczek SI, Cosper NJ, Meng L, Ruebush S, Copik AJ, et al. Kinetic and structural characterization of manganese (II)-loaded methionyl aminopeptidases. Biochemistry 2002; 41: 13096-105.
• 11 Wang J, Sheppard GS, Lou P, Kawai M, Park C, Egan DA, et al. Physiologically relevant metal cofactor for methionine aminopeptidase-2 is manganese. Biochemistry 2003; 42: 5035-42.
• 12 Li JY, Chen LL, Cui YM, Luo QL, Li J, Nan FJ, et al. Metal substitution of the methionine aminopeptidase from Escherichia coli changes specificity for substrates and inhibitors. Biochem Biophys Res Commun 2003; 307: 172-9.
• 13 Chang SY, McGary EC, Chang S. Methionine aminopeptidase gene of Escherichia coli is essential for cell growth. J Bacteriol 1989; 171: 4071-2.
• 14 Miller CG, Kukral AM, Miller JL, Movva NR. PepM is an essential gene in Salmonella typhimurium. J Bacteriol 1989; 171: 5215-7.
• 15 Li X, Chang YH. Amino-terminal protein processing in Saccharomyces cerevisiae is an essential function that requires two distinct methionine aminopeptidases. Proc Natl Acad Sci USA 1995; 92: 12357-61.
• 16 Vaughan MD, Sampson PB, Honek JF. Methionine in and out of proteins: targets for drug design. Curr Med Chem 2002; 9: 385-409.
• 17 Sin N, Meng L, Wang MQ, Wen JJ, Bornmann WG, Crews CM. The anti-angiogenic agent fumagillin covalently binds and inhibits the methionine aminopeptidase, MetAP-2. Proc Natl Acad Sci USA 1997; 94: 6099-103.
• 18 Chen SP, Vetro JA, Chang YH. The specificity in vivo of two distinct methionine aminopeptidases in Saccharomyces cerevisiae. Arch Biochem Biophys 2002; 398: 87-93.
• 19 Dummitt B, Micka WS, Chang YH. N-terminal methionine removal and methionine metabolism in Saccharomyces cerevisiae. J Cell Biochem 2003; 89: 964-74.
• 20 Griffith EC, Su Z, Niwayama S, Ramsay CA, Chang YH, Liu JO. Molecular recognition of angiogenesis inhibitors fumagillin and ovalicin by methionine aminopeptidase 2. Proc Natl Acad Sci USA 1998; 95: 15183-8.
• 21 Keding SJ, Dales NA, Lim S, Beauliu D, Rich DH. Synthesis of (3R)-amino-(2S)-hydroxy amino acids for inhibition of methionine aminopeptidase-1. Synth Commun 1998; 28:4463-70.
• 22 Luo QL, Li JY, Liu ZY, Chen LL, Shen Q, Li Y, et al. Discovery and structural modification of inhibitors of methionine aminopeptidases from Escherichia coli and Saccharomyces cerevisiae. J Med Chem 2003; 46: 2631-40.
• 23 Guan KL, Dixon JE. Eukaryotic proteins expressed in Escherichia coli: an improved thrombin cleavage and purification procedure of fusion proteins with glutathione S-transferase. Anal Biochem 1991; 192: 262-7.
• 24 Chang YH, Teichert U, Smith JA. Molecular cloning, sequencing, deletion, and overexpression of a methionine aminopeptidase gene from Saccharomyces cerevisiae. J Biol Chem 1992; 267: 8007-11.
• 25 Klinkenberg M, Ling C, Chang YH. A dominant negative mutation in Saccharomyces cerevisiae methionine aminopeptidase-1 affects catalysis and interferes with the function of methionine aminopeptidase-2. Arch Biochem Biophys 1997; 347: 193-200.
• 26 Zhou Y, Guo XC, Yi T, Yoshimoto T, Pei D. Two continuous spectrophotometric assays for methionine amino-peptidase. Anal Biochem 2000; 280: 159-65.
• 27 Walker KW, Yi E, Bradshaw RA. Yeast (Saccharomyces cerevisiae) methionine aminopeptidase I: rapid purification and improved activity assay. Biotechnol Appl Biochem 1999; 29: 157-63.
• 28 Zhuang SF, Zhou CH, Qian J, Qian Z, Shibuya M, Ye QZ. A new model for random screening inhibitors of vascular endo-thelial growth factor receptor 1 kinase. Acta Pharmacol Sin 2002; 23: 117-23.
• 29 Marino JP Jr, Thompson SK, Veber DF, inventors; Compounds and methods [P]. WO 0124796. 2001 Apr 12.