Inhibitory Effects of Diketopiperazines from Marine-Derived Streptomyces puniceus on the Isocitrate Lyase of Candida albicans

The glyoxylate cycle is a sequence of anaplerotic reactions catalyzed by the key enzymes isocitrate lyase (ICL) and malate synthase, and plays an important role in the pathogenesis of microorganisms during infection. An icl-deletion mutant of Candida albicans exhibited reduced virulence in mice compared with the wild type. Five diketopiperazines, which are small and stable cyclic peptides, isolated from the marine-derived Streptomyces puniceus Act1085, were evaluated for their inhibitory effects on C. albicans ICL. The structures of these compounds were elucidated based on spectroscopic data and comparisons with previously reported data. Cyclo(L-Phe-L-Val) was identified as a potent ICL inhibitor, with a half maximal inhibitory concentration of 27 μg/mL. Based on the growth phenotype of the icl-deletion mutants and icl expression analyses, we demonstrated that cyclo(L-Phe-L-Val) inhibits the gene transcription of ICL in C. albicans under C2-carbon-utilizing conditions.


Introduction
The glyoxylate cycle exists in a wide range of organisms, such as archaea, bacteria, fungi, plants, and nematodes. This cycle plays roles in cell metabolic processes as an anaplerotic pathway of the tricarboxylic acid cycle catalyzed by isocitrate lyase (ICL) and malate synthase, which is a key component of the glyoxylate cycle [1]. In microorganisms this cycle is essential for the uptake and utilization of non-fermentable carbon sources, such as ethanol, acetate, and fatty acids [2]. During host infection, pathogenic microorganisms including Mycobacterium tuberculosis and C. albicans cause up-regulation of the glyoxylate cycle [3,4]. In particular, the icl gene of Candida albicans is strongly activated when cells are exposed to macrophages. In mice, an icl-deletion mutant of C. albicans was unable to utilize C 2 carbon substrates and had diminished virulence compared with the wild-type strain [2,[5][6][7][8]. As this cycle does not exist in mammalian cells, ICL appears to be a prospective target for the development of antifungal drugs. Numerous studies have been conducted to find a potent ICL inhibitor [9][10][11].

50
As part of our ongoing search for inhibitors of C. albicans ICL, we encountered Streptomyces 51 puniceus Act1085, isolated from marine sediment from Jeju Island, Republic of Korea [20]. In this 52 study, an organic extract of the culture broth demonstrated inhibitory activity toward C. albicans ICL.

53
The subsequent activity-guided separation of the organic extract led to the isolation of five 54 diketopiperazines. Although these diketopiperazines have been reported to possess biological 55 properties, such as antimicrobial, antitumor, anti-inflammatory, and neuroprotective activities [21-56 23], the inhibitory activity of the isolated compounds toward ICL has not yet been explored. Thus, 57 we investigated the potential use of these compounds as inhibitors of C. albicans ICL.

60
The Act1085 strain was cultured for seven days in MTYB medium and fractionated using equal 61 volumes of n-hexane, ethylacetate (EtOAc), and n-butanol. Based on the results of the C. albicans ICL 62 activity assay, the EtOAc fraction was separated by reverse-phase C18 vacuum flash chromatography 63 and preparative high-performance liquid chromatography (HPLC) to yield five compounds. Using 64 combined spectroscopic analyses, including 1 H, 13 C nuclear magnetic resonance (NMR), 2D NMR spectral analyses (COSY, HMQC, and HMBC), and UV data, the isolated compounds were identified 66 as diketopiperazines: cyclo(L-Phe-L-Pro) [24], cyclo(L-Pro-L-Leu) [25], cyclo(L-Pro-L-Tyr) [24], 67 cyclo(L-Pro-L-Val) [25], and cyclo(L-Phe-L-Val) [26] (Figure 1). The spectroscopic data obtained from 68 the isolated compounds are consistent with previously reported values.   Isolated diketopiperazines were tested for ICL inhibitory activity according to methods reported 72 previously [27]. The inhibitory concentrations (IC50) values of the isolated compounds are shown in Table 1. Of the isolated diketopiperazines, cyclo(L-Pro-L-Leu) and cyclo(L-Pro-L-Val) demonstrated 74 weak inhibitory activity toward the ICL enzyme, with IC50 values of 533.79 μM and 516.28 μM, 75 respectively (Table 1). Cyclo(L-Phe-L-Val) exhibited the strongest inhibitory activity of the test 76 compounds but demonstrated inhibitory potency that was less than that of 3-nitropropionate, with

ICL Inhibitory Activity and Antifungal Activity of Diketopiperazines
Isolated diketopiperazines were tested for ICL inhibitory activity according to methods reported previously [27]. The inhibitory concentrations (IC 50 ) values of the isolated compounds are shown in Table 1. Of the isolated diketopiperazines, cyclo(L-Pro-L-Leu) and cyclo(L-Pro-L-Val) demonstrated weak inhibitory activity toward the ICL enzyme, with IC 50 values of 533.79 µM and 516.28 µM, respectively (Table 1). Cyclo(L-Phe-L-Val) exhibited the strongest inhibitory activity of the test compounds but demonstrated inhibitory potency that was less than that of 3-nitropropionate, with IC 50 values of 109.50 µM and 15.95 µM, respectively (Figure 2a). The others exhibited no inhibitory activity. To determine the type of inhibition, kinetic analysis was performed with cyclo(L-Phe-L-Val) (inhibitor) and phenylhydrazine (substrate). The inhibitor constant (Ki) was calculated from the Dixon plot. The results showed that cyclo(L-Phe-L-Val) behaved as a mixed inhibitor, with a K i value of 64.86 µM (Figure 2b). Fungal growth inhibition tests indicated that diketopiperazines at a concentration of 256 µg/mL did not exhibit inhibitory effects on SC5314 cultured in glucose (Table 1).   (Figure 2b). Fungal growth inhibition tests indicated that diketopiperazines at a concentration of 256 μg/mL did not exhibit inhibitory effects on SC5314 cultured in glucose (Table 1).

Inhibition of C 2 Substrate Utilization
The glyoxylate cycle is necessary for virulence in C. albicans, which can survive in macrophages under abundant carbon sources, such as fatty acids or their breakdown products [5,8]. When C. albicans is phagocytosed by a macrophage, the shift in metabolism from glycolysis to the glyoxylate cycle is activated so that the cells can utilize C 2 carbon sources. It was expected that ICL inhibitors would reduce the nutrient uptake capacity and impede survival of the pathogen in the macrophage. To determine whether cyclo(L-Phe-L-Val) affects C 2 substrate use, C. albicans strains SC5314, ATCC10231, ATCC10259, ATCC11006, and ATCC18804 were grown in YNB liquid broth containing either 2% glucose or 2% acetate as the sole carbon source. Cyclo(L-Phe-L-Val) exhibited a potent inhibitory effect on C. albicans in acetate (minimum inhibitory concentration of 32-64 µg/mL) but no inhibitory effect on C. albicans in glucose ( Table 2). These results demonstrate that cyclo(L-Phe-L-Val) affects ICL-mediated proliferation of the fungus under C 2 -carbon-utilizing conditions.

Effects of Cyclo(L-Phe-L-Val) on Growth Phenotype and icl Expression
To determine whether the cell phenotype of the icl-deletion mutant is affected by the presence of cyclo(L-Phe-L-Val), a growth assay was conducted using C. albicans SC5314 (wild type) and two icl-deletion mutants (MRC10 and MRC11). After pre-culture, these strains were streaked onto YNB agar containing 2% glucose or 2% potassium acetate with or without 32 µg/mL cyclo(L-Phe-L-Val). All strains grew normally on both the plates with glucose and with glucose plus the compound. However, MRC10 did not grow when acetate was the sole carbon source. Furthermore, none of the tested strains exhibited growth on the YNB agar plate with cyclo(L-Phe-L-Val) (Figure 3a). We further conducted semi-quantitative reverse-transcription (RT)-PCR to assess the effects of cyclo(L-Phe-L-Val) on ICL expression. No ICL-specific PCR product was detected in the cultures when SC5314 and MRC11 were grown in YNB liquid broth containing glucose. However, icl was strongly induced when these cells were cultured in YNB broth containing acetate.

135
The GPDH housekeeping gene was evaluated as a loading control.  , and MRC11 (∆icl + ICL) were cultured on YNB agar plates containing the indicated carbon source (2% glucose or 2% potassium acetate) with or without 32 µg/mL cyclo(L-Phe-L-Val) for 2 days at 28 • C. (B) Strains were grown until the mid-log phase in minimal YNB liquid medium containing 2% glucose. Cells were collected by centrifugation and transferred to the same medium containing 2% glucose (Glu), 2% potassium acetate (Ace), or 2% potassium acetate (Ace) plus cyclo(L-Phe-L-Val) (C1: 8 µg/mL; C2: 16 µg/mL; and C3: 32 µg/mL) and cultured for 4 h at 28 • C. Total RNA was prepared from these cells, and icl mRNA expression was analyzed by semi-quantitative RT-PCR. The GPDH housekeeping gene was evaluated as a loading control.

Discussion
The glyoxylate cycle, which is an anaplerotic pathway of the tricarboxylic acid cycle, is well documented in prokaryotes and eukaryotes. Diverse pathogenic fungi utilize the glyoxylate cycle during host infection. The function of this cycle has been confirmed by the analysis of mutant pathogenic microorganisms deficient in ICL and malate synthase, which are key components of the glyoxylate cycle. Research on candidiasis in mice has revealed that C. albicans, the most serious pathogenic fungus in humans, requires ICL to be fully virulent. As the glyoxylate cycle does not exist in mammalian cells, ICL is a promising target for antimicrobial agents. Therefore, we are interested in identifying an ICL inhibitor derived from natural products. In this study, five diketopiperazine compounds were isolated from the marine-derived S. puniceus Act1085, and their structures and inhibitory activities toward the C. albicans ICL enzyme were evaluated. Our work concluded that cyclo(L-Phe-L-Val) possesses ICL inhibitory activity. Analyses of the phenotype of icl-deletion mutants by growth assays and semi-quantitative RT-PCR demonstrated that this compound inhibits ICL expression in C. albicans under C 2 -carbon-utilizing conditions. Diketopiperazines possess various biological properties [12,[14][15][16][17][18][19]. Many research groups have focused on the pharmacological potential of diketopiperazines due to their advantages in medical chemistry, such as stability against proteolysis, mimicry of peptidic pharmacophoric groups, conformational rigidity, substituent group stereochemistry, and the existence of donor and acceptor groups for hydrogen bonding (which favor interactions with targets) [28]. Furthermore, diketopiperazines can be easily synthesized and isolated from natural products. In this study, five diketopiperazines, cyclo(L-Phe-L-Pro), cyclo(L-Pro-L-Leu), cyclo(L-Pro-L-Tyr), cyclo(L-Pro-L-Val), and cyclo(L-Phe-L-Val), were obtained from a marine actinomycete, S. puniceus, using activity-guided separation processes. Among these diketopiperazines, cyclo(L-Pro-L-Leu), cyclo(L-Pro-L-Val), and cyclo(L-Phe-L-Val) exhibited ICL inhibitory activity, with IC 50 values in the range of 27-112 µg/mL ( Table 1). The structures of the isolated diketopiperazines exhibiting ICL inhibition activity contained an isopropyl moiety, which suggests that the isopropyl moiety plays a role in the inhibition of ICL activity.
In M. tuberculosis, a bacterium, and C. albicans, a fungus, virulence led to expression of icl, which encodes a component of the glyoxylate cycle during persistent infection of macrophages [5,8]. These findings indicate that inhibitors of the glyoxylate cycle may reduce nutrient uptake capacity, leading to the death of these pathogens in macrophages. Thus, the effect of cyclo(L-Phe-L-Val) on C. albicans ICL was investigated by analyzing the growth phenotype and icl transcript levels in C. albicans SC5314 (wild-type) and icl-deletion mutants (MRC10 and MRC11) [8]. The growth assay revealed that cyclo(L-Phe-L-Val) specifically inhibits the ICL enzyme, because no growth of SC5314 or MRC11 was observed on the YNB agar plates containing acetate plus 32 µg/mL of the compound. Moreover, icl transcript levels declined as a result of treatment with cyclo(L-Phe-L-Val). Overall, this research uncovered a compound of the diketopiperazine class as a promising antifungal agent that acts by suppressing C. albicans pathogenicity.

General Experimental Procedure
The NMR instruments and mass spectrometry equipment have been described in a previous report [29]. HPLC analysis was conducted using the Shimadzu SCL-10A (Shimadzu, Tokyo, Japan) control system connected to the RID-10A refractive index detector (Shimadzu) and UV-Vis SPD-10A detector (Shimadzu). All organic solvents were of analytical reagent grade and purchased from Fisher Scientific (Fair Lawn, NJ, USA).

Fermentation and Isolation of Diketopiperazines
S. puniceus Act1085 was streaked onto GTYB agar plates and incubated at 28 • C for 5 days. The composition of the GTYB medium was 2 g/L tryptone, 1 g/L yeast extract, 1 g/L beef extract, 16 g/L agar, and 10 g/L glucose. The colonies were brown in color and produced white spores. Spores of Act1085 were cultivated in 2 L flasks containing 300 mL MTYB broth for fermentation. The MTYB medium composition was 2 g/L tryptone, 1 g/L yeast extract, 1 g/L beef extract, and 10 g/L mannitol. After cultivation for 1 week under static conditions at 28 • C, the culture broth (40 L) was filtered through filter paper to remove mycelia, and the filtered culture broth was sequentially fractionated using equal volumes (80 L) of n-hexane, EtOAc, and n-butanol to obtain 146 mg of n-hexane extract, 1.7 g of EtOAc extract, and 3.9 g of n-butanol extract. Based on the results of the ICL activity assay (IC 50 values of n-hexane, EtOAc, and n-butanol extracts were >256 µg/mL, 47 µg/mL, and >256 µg/mL, respectively), the EtOAc extract (1.7 g) was selected for further separation experiments. The EtOAc extract was subjected to reverse-phase flash column chromatography using a column packed with YMC Gel ODS-A (S-75 µm) and eluted with MeOH-H 2 O (gradient from 0:100 to 100:0), acetone, and EtOAc to produce seven fractions. Based on the bioactivity analyses, the H 2 O-MeOH (60:40, 325 mg) fraction was isolated and purified by semi-preparative reverse-phase HPLC (Agilient C 18

ICL Inhibition Assay
Before evaluation of the ICL inhibitory activity of the test compounds, the recombinant ICL from C. albicans was prepared according to methods described in a previous report [27]. The ICL inhibitory activity of the test compounds was determined according to a previously described method [30,31]. The ICL inhibition assay is based on the principle that glyoxylate phenylhydrazone forms in the reaction reagent after treatment with isocitrate and phenylhydrazine. Each test compound was dissolved in DMSO. The reaction mixture consisted of 20 mM sodium phosphate buffer (pH 7.0), 1.27 mM threo-DL(+)isocitrate, 3.75 mM MgCl 2 , 4.1 mM phenylhydrazine, and 2.5 µg/mL purified ICL and was incubated with or without the test compounds at a concentration range of 1-128 µg/mL at 37 • C for 30 min. The increase in intensity of absorbance resulting from the formation of glyoxylate phenylhydrazone was observed using a spectrophotometer (Shimadzu) at a wavelength of 324 nm. The inhibitory activity of the test compound was calculated relative to that of the DMSO control (n = 3). The methods for determination of the IC 50 value and ICL enzyme concentrations have been described in previous reports [29]. A known ICL inhibitor, 3-nitropropionate, was used as a reference control [32].

In Vitro Growth Assay
C. albicans strains were cultured in YNB broth containing 2% glucose at 28 • C for 24 h, centrifuged at 15,000× g for 1 min, and washed twice with sterile distilled water. Each test compound was dissolved in DMSO and diluted with YNB broth containing 2% glucose or 2% potassium acetate at concentrations ranging from 1 to 256 µg/mL. Additional DMSO was added to the medium at a final concentration of 0.5%. The fungal strain culture (20 µL) was poured into a 96-well assay plate at a final concentration of 1 × 10 4 cells/mL and total volume of 100 µL. The culture plates were incubated at 28 • C for 3 days. The positive control was amphotericin B, which is a known antifungal compound.

Growth Phenotype and Icl Expression Analysis
C. albicans SC5314 (wild type) and icl-deletion mutants (MRC10 and MRC11) were cultured in YNB broth containing 2% glucose at 28 • C on a rotary shaker for 24 h. Cells were collected and washed as described above. The harvested cells were streaked on YNB agar plates containing 2% glucose, 2% potassium acetate, or 2% potassium acetate plus 32 µg/mL cyclo(L-Phe-L-Val) and incubated at 28 • C for 2 days. For icl expression analysis, the harvested cells were added to YNB liquid broth containing 2% glucose, 2% potassium acetate, or 2% potassium acetate plus cyclo(L-Phe-L-Val) (8, 16, or 32 µg/mL) and incubated at 28 • C for 4 h. RNA extraction, cDNA synthesis, and semi-quantitative RT-PCR were performed using previously described methods [33]. GPDH was used as the loading control.