Furaquinocins K and L: Novel Naphthoquinone-Based Meroterpenoids from Streptomyces sp. Je 1-369

Actinomycetes are the most prominent group of microorganisms that produce biologically active compounds. Among them, special attention is focused on bacteria in the genus Streptomyces. Streptomycetes are an important source of biologically active natural compounds that could be considered therapeutic agents. In this study, we described the identification, purification, and structure elucidation of two new naphthoquinone-based meroterpenoids, furaquinocins K and L, from Streptomyces sp. Je 1-369 strain, which was isolated from the rhizosphere soil of Juniperus excelsa (Bieb.). The main difference between furaquinocins K and L and the described furaquinocins was a modification in the polyketide naphthoquinone skeleton. In addition, the structure of furaquinocin L contained an acetylhydrazone fragment, which is quite rare for natural compounds. We also identified a furaquinocin biosynthetic gene cluster in the Je 1-369 strain, which showed similarity (60%) with the furaquinocin B biosynthetic gene cluster from Streptomyces sp. KO-3988. Furaquinocin L showed activity against Gram-positive bacteria without cytotoxic effects.


Introduction
Actinomycetes are aerobic, spore-forming, Gram-positive bacteria that are characterized by a high G+C content in their genomes. These bacteria synthesize approximately twothirds of all natural antibiotics used in medicine, veterinary medicine, and agriculture [1]. On average, actinomycete strains have the genetic potential to produce approximately 10-20 secondary metabolites [2]. Secondary metabolites produced by actinomycetes have been widely studied since the 1950s and are the most economically and biotechnologically valuable source for discovering new biologically active compounds [3]. The analysis of actinomycete secondary metabolites resulted in the discovery of many biologically active compounds, which were later embodied in the development of antimicrobial (i.e., vancomycin, chloramphenicol, and tetracycline) anticancer (i.e., daunorubicin and bleomycin), and immunosuppressive (i.e., rapamycin) drugs among others [4]. However, due to the rapid spread of microbial pathogens and their frequent resistance to known antibiotics, the search for new biologically active compounds is still relevant.
Meroterpenoids are a large class of natural products derived from polyketide or nonpolyketide and terpenoid biosynthesis [5]. A significant amount of naphthoquinonebased meroterpenoids have been isolated from marine and soil-derived streptomycete bacteria [6,7]. The structure of naphthoquinone-based meroterpenoids contains a 1,3,6,8tetrahydroxynaphthalene core and a prenyl moiety, which are linked by aromatic prenyl-

Identification and Structure Elucidation of Furaquinocins K and L
The Streptomyces sp. Je 1-369 strain was isolated from the rhizosphere soil of J. excelsa (Crimea Peninsula, Ukraine). High-resolution LC-MS analysis of the crude extract of this strain revealed the presence of several peaks, two of which exhibited an interesting UV spectrum (Supplementary Materials Figure S1). These peaks were characterized by strong UV absorption signals at λ max 226, 264, 300, and 408 nm and at λ max 228, 278, 328, and 504 for peaks 1 and 2, respectively. Peak 1 had a retention time (tR) of 14.7 min and an m/z of 385.2005 Da [M+H] + (1), whereas for peak 2 they were tR = 15.3 min and m/z 443.2187 Da [M+H] + (2). The dereplication analysis of the identified monoisotopic masses 384.1929 (1) and 442.2114 (2) in the Dictionary of Natural Products Database (DNP) [14] yielded no matches. Thus, the lack of matches in the DNP database may indicate the novelty of these compounds.
To determine the structure of the identified compounds, the Je 1-369 strain was grown in 10 L of SG medium, and metabolites were extracted from the supernatant with ethyl acetate. The obtained extract was purified in three stages, including normal-phase chromatography on a silica gel column, size-exclusion chromatography through a Sephadex column, and preparative high-performance liquid chromatography (HPLC). As a result, 3.2 and 1.4 mg of compounds 1 and 2 were obtained.
Compound 2, C24H30N2O6, m/z 443.2187 Da ([M+H] + ), was present as a red solid. Its 1D and 2D NMR spectra (Table 2 and Figures S2 and S9-S14) revealed structural elements Furaquinocin K contains two stereo centres at C-2 and C-3, which are in the 2R and 3R configuration in all naturally occurring furaquinocins [18]. To confirm this finding, the optical rotation value of furaquinocin K, [ (Table 2 and Figures S2 and S9-S14) revealed structural elements already known from furaquinocin K (1), including the 8-Me (δ C 9.26 and δ H 2.06) and 7-OMe (δ C 60.70 and δ H 4.00) groups and the monoterpene unit as part of a cyclic ether. However, in contrast to 1, H-5 was missing, the five-membered ether ring was fused to a 1,2-benzoquinone derivative (δ C 179.22 and 139.94) rather than a phenol, and the carbonyls of the former 1,4-benzoquinone moiety at C-6 and C-9 were converted into a hydroquinone moiety (δ C 147.79 and 149.00). In addition, the molecular formula of 2 was enlarged by CH 2 N 2 O. The nature of the two nitrogens was further analysed by 1H,15N-HSQC and 1H,15N-HMBC (Figures S15 and S16). This identified an NH signal at δN 171.7 and δH 14.90 and an N signal at δN 301.9, in which the chemical shifts matched a hydrazone unit. This was supported by the correlation of the NH proton with C-5 (δC 139.94) of the benzoquinone in the 1H,13C-HMBC. In addition, correlations of the NH proton and a methyl group (δC 22.12, δH 2.25 s, C-17) with the carbons of a carbonyl (δC 179.22, C-16) were detected, which indicated a directly adjacent acetyl group. Consequently, the structural component was assigned to an acetylhydrazone.
The low-field-shifted proton signals of NH (δNH 14.90) and 6-OH (δOH 12.89) indicated intramolecular hydrogen bonds to the carbonyl oxygens and the imine nitrogen of hydrazine, respectively. This led to two possible constitutional isomers (2 and 2a) (Figure 2A), as C-4 and C-5 could not be distinguished by HMBC. To determine the correct isomer, the measured carbon shift of 2 was compared with predicted data for 2 and 2a, which were calculated using the NMR prediction tool of ACD Labs, version 2021.2.0 ( Figure 2B). As a result, the chemical shift values of isomer 2 showed a slightly better fit based on the regression coefficient values [20]. In addition, a 1H,15N-HMBC correlation from 6-OH to the imine nitrogen 5-N suggested a 1h J N , OH coupling caused by an O-H.... N intramolecular hydrogen bond, as also present in 2 [21], which strongly indicated that 2 was the correct structure.
2A), as C-4 and C-5 could not be distinguished by HMBC. To determine the correct isomer, the measured carbon shift of 2 was compared with predicted data for 2 and 2a, which were calculated using the NMR prediction tool of ACD Labs, version 2021.2.0 ( Figure 2B). As a result, the chemical shift values of isomer 2 showed a slightly better fit based on the regression coefficient values [20]. In addition, a 1H,15N-HMBC correlation from 6-OH to the imine nitrogen 5-N suggested a 1h JN, OH coupling caused by an O-H.... N intramolecular hydrogen bond, as also present in 2 [21], which strongly indicated that 2 was the correct structure.

Identification of the Furaquinocin Biosynthetic Gene Cluster
To identify the furaquinocin biosynthetic gene cluster, the genome of Je 1-369 was sequenced and analysed. The phylogenetic analysis based on the 16S rRNA gene sequence of the Je1-369 strain revealed this strain to belong to the genus Streptomyces ( Figure S17). The total genome size of Streptomyces sp. Je 1-369 strain is 8,820,026 bp with 71% G+C. Terminal inverted repeats (TIRs) of 160,307 bp are present at the ends of the chromosome. The genome annotation of this strain identified 7695 probable protein-coding genes, eighteen rRNA genes in six operons, and eighty-seven tRNA genes. A preliminary analysis of the complete genome using antiSMASH showed the presence of thirty-six predicted gene clusters (two of which were duplicated in TIRs) (Supplementary Table S1). Among them, there were nine clusters of terpenes, five PKS, and three NRPS; others included lanthipeptide, siderophores, ectoine, phenazine, bacteriocin, butyrolactone, melanine, clusters, and several hybrid clusters. Two secondary metabolite gene clusters were located at the edges of the chromosome in the TIRs. One of them showed a 60% homology to the furaquinocin B (fur) biosynthetic gene cluster.
A detailed analysis of the detected cluster in Streptomyces sp. strain Je 1-369 showed its significant difference from the described fur cluster from Streptomyces sp. KO-3988. The main difference between the studied clusters lay in their gene cluster organization. However, the key furaquinocin-forming genes involved in naphthoquinone core formation (open reading frame (Orf) 1-4), polyprenyl synthetase (Orf22), and prenyltrasferase (Orf26) were presented. In addition, five of the six mevalonate pathway genes were missing from the identified cluster from the Je 1-369 strain, but these genes were detected outside the cluster. In addition, a complete set of genes (Orf8-Orf12) for the recycling of S-adenosylhomocysteine to S-adenosylmethionine were contained between the furaquinocine-forming genes in the cluster from the Je 1-369 strain ( Figure 3 and Table

Identification of the Furaquinocin Biosynthetic Gene Cluster
To identify the furaquinocin biosynthetic gene cluster, the genome of Je 1-369 was sequenced and analysed. The phylogenetic analysis based on the 16S rRNA gene sequence of the Je1-369 strain revealed this strain to belong to the genus Streptomyces ( Figure S17). The total genome size of Streptomyces sp. Je 1-369 strain is 8,820,026 bp with 71% G+C. Terminal inverted repeats (TIRs) of 160,307 bp are present at the ends of the chromosome. The genome annotation of this strain identified 7695 probable protein-coding genes, eighteen rRNA genes in six operons, and eighty-seven tRNA genes. A preliminary analysis of the complete genome using antiSMASH showed the presence of thirty-six predicted gene clusters (two of which were duplicated in TIRs) (Supplementary Table S1). Among them, there were nine clusters of terpenes, five PKS, and three NRPS; others included lanthipeptide, siderophores, ectoine, phenazine, bacteriocin, butyrolactone, melanine, clusters, and several hybrid clusters. Two secondary metabolite gene clusters were located at the edges of the chromosome in the TIRs. One of them showed a 60% homology to the furaquinocin B (fur) biosynthetic gene cluster.
A detailed analysis of the detected cluster in Streptomyces sp. strain Je 1-369 showed its significant difference from the described fur cluster from Streptomyces sp. KO-3988. The main difference between the studied clusters lay in their gene cluster organization. However, the key furaquinocin-forming genes involved in naphthoquinone core formation (open reading frame (Orf) 1-4), polyprenyl synthetase (Orf22), and prenyltrasferase (Orf26) were presented. In addition, five of the six mevalonate pathway genes were missing from the identified cluster from the Je 1-369 strain, but these genes were detected outside the cluster. In addition, a complete set of genes (Orf8-Orf12) for the recycling of S-adenosylhomocysteine to S-adenosylmethionine were contained between the furaquinocine-forming genes in the cluster from the Je 1-369 strain ( Figure 3 and Table 3). An identical gene arrangement occurred in the furanonaphthoquinone I gene cluster [22], which was a regioisomer of furaquinocin C.
FOR PEER REVIEW 6 of 12 3). An identical gene arrangement occurred in the furanonaphthoquinone I gene cluster [22], which was a regioisomer of furaquinocin C.

Biological Activity of Novel Furaquinocin Analogues
The structures of the furaquinocin analogues identified in this study contained significant differences from those already described. Therefore, these differences in structure may have a significant effect on their activities. Thus, the identified furaquinocins K and

Biological Activity of Novel Furaquinocin Analogues
The structures of the furaquinocin analogues identified in this study contained significant differences from those already described. Therefore, these differences in structure may have a significant effect on their activities. Thus, the identified furaquinocins K and L were tested against a wide range of test strains of bacteria, yeast, and fungi, including Bacillus subtilis DSM 10, Staphylococcus aureus Newman, Mycobacterium smegmatis mc2155, Escherichia coli BW25113 (wt), E. coli JW0451-2 (∆acrB), Pseudomonas aeruginosa PA14, Acinetobacter baumannii DSM 30008, Citrobacter freundii DSM 30039, Candida albicans DSM 1665, Cryptococcus neoformans DSM 11959, Pichia anomala DSM 6766, and Mucor hiemalis DSM 2656. Furaquinocin K showed no antagonistic activity against the tested strains but  (Table 4).

Discussion
The numbers of cases involving antibiotic-resistant bacterial pathogens are increasing; as a result, mankind is facing a dilemma and screening for new antibiotics is urgently needed [23]. Microbial secondary metabolites are the predominant source of biologically active products that can be used as therapeutic agents to preserve human life and health [24]. Meroterpenoids can play a significant role in this endeavour, since most of them show an impressive range of biological activity [9]. In this study, we described the identification, purification, and structure elucidation of two new naphthoquinone-based meroterpenoids, furaquinocins K and L, from the Streptomyces sp. Je 1-369 strain isolated from the rhizosphere soil of J. excelsa. Furaquinocins are a small family of meroterpenoids, including furaquinocins A-J [15][16][17], furanonaphthoquinone I, which is a regioisomer of furaquinocin C [25], furaquinocin derivatives PI-220 [26], and JBIR-136 [27]. The furaquinocins A-J and their homologues differ only in the modification of the terpene side chains. However, the furaquinocins K and L isolated in this study contained a modification in the naphthoquinone moiety. In addition, the structure of furaquinocin L included an acetylhydrazone moiety, which is quite rare among natural products [28]. Not surprisingly, relatively little is known about hydrazone biosynthesis. A recent study showed that the hydrazone group is formed by nonenzymatic Japp-Klingemann coupling between the electrophilic diazotated alkyl 5-hydroxylanthranilate and a β-keto aldehyde-containing cyclic peptide precursor during tazicamide biosynthesis [29]. Given the cardinal difference in the structure between tazicamides and furaquinocin L, we believe that the formation of the hydrazone group between these structures will also be different.
The gene cluster of furaquinocin B was identified in the course of a study by Kawasaki et al. [30] due to its promising cytotoxic activity against human cancer cells. Genome analysis of the Je 1-369 strain revealed a gene cluster that showed a 60% homology with the cluster of furaquinocin B biosynthesis. Moreover, this cluster was located in the TIRs at the ends of the chromosome. The presence of gene clusters that are duplicated in streptomycete chromosomes is called "superclusters" and can often affect production levels [31]. However, we did not find any genes similar to the genes involved in N-N bond formation in the identified furaquinocin gene cluster. Thus, we suggest that extracluster genes or genes in which the function has not yet been characterized may be involved in the formation of the furaquinocin L structure. Thus, further studies on the biosynthesis of furaquinocin L, namely, the formation of acetylhydrazone, will clarify the nature and mechanisms of hydrazone-containing compound formation.
As mentioned above, the furaquinocins showed promising cytotoxic activity but did not exhibit antimicrobial activity. Furaquinocin K showed no antimicrobial activity but demonstrated cytotoxicity against HepG2 cells in the same manner as the described furaquinocins and their analogues. This was explained by the slight difference in the chemical structures, in which the difference between the C and K furaquinocins lay in the methoxy group at C-4 ( Figure 1). We also hypothesized that the presence of acetylhydrazone in the structure of furaquinocin L may have a significant effect on its activity, since N-Ncontaining natural compounds exhibit a diverse spectrum of biological activities [28]. As a result, furaquinocin L containing acetylhydrazone showed activity against Gram-positive bacteria in the absence of cytotoxicity. Therefore, to the best of our knowledge, this is the first furaquinocin with antibacterial activity. Thus, our study expands the possibilities of using furaquinocins not only as anticancer agents but also as potential antibacterial agents.

General Experimental Procedures
The strain Je 1-369 isolated from the rhizosphere soil of J. excelsa was used in this study. This strain was isolated by direct inoculation of an aqueous suspension of soil on OM medium (20 g/L oat flour and 20 g/L agar; pH 7.2) and incubated at 28 • C for 14 days. Spores and mycelial suspensions of Je 1-369 strain were stored in 20% (v/v) glycerol solution at −20 • C and deposited in the Microbial Culture Collection of Antibiotic Producers (MCCAP) of Ivan Franko National University of Lviv (collection number Lv 391).
OM medium was used to cultivate the actinomycete strain. Liquid tryptic soy broth (TSB) (Sigma-Aldrich, St. Louis, MO, USA) was used for the extraction of total DNA. SG medium (20.0 g/L glucose, 10 g/L soy peptone, and 2 g/L CaCO 3 ; pH 7.2) was used to produce the secondary metabolites.

Secondary Metabolite Extraction and Analysis
To extract the secondary metabolites, the Streptomyces sp. Je 1-369 strain was grown in 15 mL of TSB in a 100 mL flask for 2 days at 28 • C and 180 rpm, and 1 mL of pre-culture was inoculated into 100 mL of production SG medium in a 500 mL flask. The Je 1-369 strain was grown for 7 days at 28 • C and 180 rpm in an Infors multitron shaker (Infors AG, Basel, Switzerland). The secondary metabolites of the Je 1-369 strain were then extracted from the culture supernatant with an equal amount of ethyl acetate and acetone:methanol (1:1) mixture from the culture biomass. The obtained extracts were evaporated using an IKA RV-8 rotary evaporator (IKA, Staufen, Germany) at 40 • C and were dissolved in methanol. The extracts were analysed on a Dionex Ultimate 3000 UPLC system (ThermoFisher Scientific, Waltham, MA, USA) coupled to a PDA detector using a 100 mm ACQUITY UPLC BEH

Secondary Metabolite Purification
Secondary metabolites were extracted from the Streptomyces sp. Je 1-369 strain grown in 10 L of SG medium. The ethyl acetate extract from the culture supernatant of this strain was dissolved in methanol and purified in three stages. The first purification stage was normal-phase chromatography on a silica gel column with hexane (solvent A), chloroform (solvent B), ethyl acetate (solvent C), and methanol (solvent D) as the mobile phase at a flow rate of 100 mL/min. A triple linear gradient of each solvent pair A/B (10 column volumes (CV)), B/C (15 CV) and C/D (15 CV) was used, and fractions were collected every 18 mL. The separation was performed on a Biotage Isolera One LC-system (Biotage, Uppsala, Sweden) using a SNAP Ultra 50 g column HP-Sphere (Biotage, Uppsala, Sweden). The fractions containing the compounds of interest were pooled together, concentrated, and used for the second separation stage, which was size-exclusion chromatography on a Sephadex

Antimicrobial Susceptibility Test and Cytotoxicity Assay
MICs were determined according to standard procedures. Single colonies of the bacterial strains were suspended in cation-adjusted Müller-Hinton broth to obtain a final inoculum of 10 4 CFU mL −1 . Serial dilutions of furaquinocins (0.5 to 64 µg/mL) were prepared in sterile 96-well plates, and the bacterial suspension was added. Growth inhibition was assessed after overnight incubation at 30-37 • C for 16-18 h. A total of 5 µL of thiazolyl blue tetrazolium bromide (MTT, 10 mg/mL) solution was added to each well and plates were incubated at 30 • C for 1 h. MICs were evaluated visually as the concentration of furaquinocins in a well, in which the compound colour did not change from yellow to dark blue. The following microbial test cultures were used: B. subtilis DSM 10, S. aureus Newman, M. smegmatis mc2155, E. coli BW25113 (wt), E. coli JW0451-2 (∆acrB), P. aeruginosa PA14, A. baumannii DSM 30008, C. freundii DSM 30039, C. albicans DSM 1665, C. neoformans DSM 11959, P. anomala DSM 6766, and M. hiemalis DSM 2656.
The HepG2 cell line was used to evaluate the cytotoxic activity of the isolated furaquinocins. The cell line was obtained from the German Collection of Microorganisms and Cell Cultures (Deutsche Sammlung für Mikroorganismen und Zellkulturen (DSMZ)) and cultured under the conditions recommended by the depositor. Cells were grown and diluted to 5 × 10 4 per well of 96-well plates in 180 µL of complete RPMI-1640 medium (+10% fetal bovine serum (FBS)). After 2 h of equilibration (37 • C and 5% CO 2 ), the cells were treated with a serial dilution of furaquinocins in methanol. A total of 20 µL of 5 mg/mL MTT in phosphate-buffered saline (PBS) was added to each well after the cells were grown for 5 days (37 • C and 5% CO 2 ). The cells were further incubated for 2 h at 37 • C before the supernatant was discarded. Subsequently, the cells were washed with 100 µL of PBS and treated with 100 µL of 2-propanol/10 N HCl (250:1) to dissolve formazan granules. Cell viability was measured as a percentage relative to the respective methanol control by measuring the absorbance at 570 nm with a microplate reader (Tecan Infinite 200 PRO). GraphPad Prism was used for sigmoidal curve fitting to determine the IC50 values as well as the calculation of confidence intervals.

Genome Sequencing and Bioinformatics Analysis
To extract the total DNA, the Streptomyces sp. Je 1-369 strain was grown in TSB medium for four days at 28 • C with a shaking rate of 180 rpm. A salting-out procedure was used to obtain the total DNA [32]. The RNA-free genomic DNA of strain Je 1-369 was sequenced using an Illumina paired-end sequencing library (TruSeq sample preparation kit; Illumina, USA) as recommended by the manufacturer. The Illumina sequencing data were de novo assembled using Newbler v2.8 (454 Life Sciences, Branford, CT, USA) with default settings. Gene prediction and genome annotation were carried out using the prokka v1.11 and GenDB 2.0 platforms [33,34]. The secondary metabolite gene clusters were analysed using the antiSMASH genome mining tool [35]. The analysis of genetic data was performed using Geneious 9.1.2 software [36]. The whole genome sequences of Streptomyces sp. Je 1-369 were deposited under accession number CP101750 (PRJNA860344 for BioProject and SAMN29841164 for BioSample) into the GenBank database.

Conclusions
In summary, two new naphthoquinone-based meroterpenoids, furaquinocin K and L, were isolated from the Streptomyces sp. Je 1-369 strain. Their structures were elucidated by NMR and found to contain modifications in the polyketide naphthoquinone skeleton that have not yet been described for furaquinocins. Deciphering the biosynthesis of these furaquinocin analogues will expand the knowledge about the biosynthesis of meroterpenoids as well as the formation of hydrazones in natural compounds.