Identification of the Gene Responsible for Lignin-Derived Low-Molecular

G C A T T A C G G C A T genes

Article Identiﬁcation of the Gene Responsible for Lignin-Derived Low-Molecular-Weight Compound Catabolism in Pseudomonas sp. Strain LLC-1

Jun Hirose *, Ryusei Tsukimata, Munetoshi Miyatake and Haruhiko Yokoi

Faculty of Engineering, University of Miyazaki, Miyazaki 889-2192, Japan; [email protected] (R.T.); [email protected] (M.M.); [email protected] (H.Y.) * Correspondence: [email protected]; Tel.: +81-985-58-7322

Received: 30 October 2020; Accepted: 25 November 2020; Published: 27 November 2020

Abstract: Pseudomonas sp. strain LLC-1 (NBRC 111237) is capable of degrading lignin-derived low-molecular-weight compounds (LLCs). The genes responsible for the catabolism of LLCs were characterized in this study using whole-genome sequencing. Despite the close phylogenetic relationship with Pseudomonas putida, strain LLC-1 lacked the genes usually found in the P. putida genome, which included fer, encoding an enzyme for ferulic acid catabolism, and vdh encoding an NAD+-dependent aldehyde dehydrogenase specific for its catabolic intermediate, vanillin. Cloning and expression of the 8.5 kb locus adjacent to the van operon involved in vanillic acid catabolism revealed the bzf gene cluster, which is involved in benzoylformic acid catabolism. One of the structural genes identified, bzfC, expresses the enzyme (BzfC) having the ability to transform vanillin and syringaldehyde to corresponding acids, indicating that BzfC is a multifunctional enzyme that initiates oxidization of LLCs in strain LLC-1. Benzoylformic acid is a catabolic intermediate of (R,S)-mandelic acid in P. putida. Strain LLC-1 did not possess the genes for mandelic acid racemization and oxidation, suggesting that the function of benzoylformic acid catabolic enzymes is different from that in P. putida. Genome-wide characterization identified the bzf gene responsible for benzoylformate and vanillin catabolism in strain LLC-1, exhibiting a unique mode of dissimilation for biomass-derived aromatic compounds by this strain.

Keywords: benzoylformic acid; bzf gene; lignin; mandelic acid; Pseudomonas; vanillin; vanillin dehydrogenase

1. Introduction Microorganisms capable of catabolizing aromatic compounds play an important role in degrading aromatic biomass compounds in the environment, major components of which are lignin-derived low-molecular-weight compounds (LLCs) produced by lignin depolymerization. They are composed of aldehydes, ketones, and organic acids. Aromatic LLCs are typically catabolized through the upper and lower ring fission pathways, where the aromatic rings are cleaved and subsequent metabolites enter the central carbon metabolism [1]. Many genes coding for the LLC metabolic pathways have been isolated and characterized. The upper and lower ring fission pathways are encoded by different loci and are controlled by independent transcriptional regulators [2,3]. The genes coding for lower pathways, such as potocatechate, benzoate, and gentisate, are ubiquitous and well conserved in many aerobic bacteria [2]. Recent advances in DNA sequencing technology have enabled the identification of additional genes involved in LLC catabolism; however, most of these remain to be studied in terms of their function and regulation.

Genes 2020, 11, 1416; doi:10.3390/genes11121416 www.mdpi.com/journal/genes Genes 2020, 11, 1416 2 of 13

Vanillin is one of the major components of LLCs. Both bio-production [4,5] and bio-degradation [6] of vanillin are of interest. Vanillin itself is a flavor that is used in the food industry. Many attempts have been made to produce dicarboxylic bio-plastic monomer compounds from LLC using engineered bacteria [7–9]. Several pathways involving bacterial catabolism of vanillin have been reported. Vanillin is a catabolic intermediate of eugenol [10,11], isoeugenol [12], or ferulic acid [13,14] in Pseudomonas and other bacteria. They are transformed into vanillin via ligation by coenzyme A and subsequent cleavage of the propanoic side chain. The vanillin catabolic pathway is initiated by oxidation to vanillic acid and undergoes monooxygenation to form protocatechuate. This is followed by dioxygenation to form ring-cleavage intermediate compounds that enter the tricarboxylic acid cycle via the β-ketoadipate pathway [2]. We identified a gram-negative aerobic bacterial isolate, Pseudomonas sp. strain LLC-1 (here referred to as strain LLC-1) using enrichment culture in a medium containing LLCs and inorganic salts [15]. This bacterial strain can utilize vanillin and vanillic acid as sole carbon sources, and co-metabolizes syringaldehyde, o-vanillin, and isovanillin. Here, we present the identification and characterization of a gene responsible for the catabolic pathway of LLCs using whole-genome sequencing.

2. Materials and Methods

2.1. Bacterial Strains, Plasmid, and Cultivation The bacterial strains used in this study are presented in Table1. Pseudomonas sp. strain LLC-1 (NBRC 111237) was isolated from soil in Miyazaki, Japan [15]. P. putida NBRC 14164 was obtained from the Biological Resource Center of the National Institute of Technology and Evaluation (NBRC), Tokyo, Japan. For the growth of Pseudomonas strains, a basal salts medium (BSM) containing (in grams per liter) K HPO , 4.3; KH PO , 3.4; (NH ) SO , 2.0; MgCl 6H O, 0.34; MnCl 4H O, 0.001; FeSO 7H O, 2 4 2 4 4 2 4 2· 2 2· 2 4· 2 0.0006; CaCl 2H O, 0.026; and Na MoO 2H O, 0.002 (pH 7.0) was used. BSM was supplemented 2· 2 2 4· 2 with 0.1% (w/v) of various aromatic compounds. The bacterial strains were grown with shaking (120 rpm) at 30 ◦C and the growth was monitored by optical density measurements at 600 nm. For DNA isolation or freezing stock preparation, Luria–Bertani (LB) medium (Bacto Trypton, 10 g; yeast extract, 5 g; and NaCl, 10 g per liter, pH 7.0) was used. The Escherichia coli cells were grown with shaking (160 rpm) at 37 ◦C in Terriﬁc broth (TB) medium (1.2% Bacto Tryptone, 2.4% yeast extract, 0.5% glycerol, 0.17 M KH2PO4-0.72 M K2HPO4 10%(v/v)) supplemented with ampicillin sodium (100 µg/mL). Isopropyl-β-D-thiogalactpyranoside (IPTG) was added to a ﬁnal concentration of 0.2 mM to induce gene expression under the control of the lac promoter in the plasmid vector.

2.2. Genome Sequencing and Computational Analysis The whole genome sequence of strain LLC-1 was determined as previously reported [16]. The reads obtained by the two systems were assembled using Newbler v.2.8 (Roche, Basel, Switzerland). The genome sequences were annotated using the NCBI Prokaryotic Genome Annotation Pipeline (PGAP) [17] and Rapid Annotations using Subsystems Technology (RAST) server v.2.0 [18]. Identiﬁcation of coding genes was performed using BLAST and BLASTX searches [19]. Sequence comparisons were performed using EasyFig v.2.1 [20], and a map was generated using drawGeneArrows3. The 16S rRNA gene sequences were aligned computationally using the ClustalW algorithm, as described. Phylogenetic trees were generated using the neighbor-joining method with the MEGA X program [21]. The trees were evaluated through bootstrap resampling (500 replicates). Genes 2020, 11, 1416 3 of 13

Table 1. Bacterial Strains and Plasmids Used in this Study.

Strain or Plasmid Relevant Characteristic(s) Source or Reference Bacterial strains + Pseudomonas sp. strain VNL MDL− [15,16] LLC-1 (NBRC 111237) P. putida MDL+ [22] NBRC 14164 recAl endAl gyrA96 thi hsdRl7 E. coli JM109 supE44 relA1∆(lac proAB) [F’proAB Laboratory stock laclq DM15 traD36] Plasmid pBluescript KS+ lacZ Apr Agilent Technologies/Stratagene pMD20 lacZ Apr Takara Bio pMDBzfC pMD20-bzfC This study pKSBzfC pBluescript KS+-bzfC This study pKSBzf1 pBluescript KS+-bzfABCD This study VNL, vanillin utilization; MDL, mandelic acid utilization; Apr, ampicillin resistance.

2.3. Cloning and Expression of bzf Gene Cluster DNA isolation, Southern blot analysis, PCR, DNA sequencing, and other DNA manipulations were performed according to standard procedures as described [23]. A sequence encoding the bzfC gene (1.58 kb) was ampliﬁed by PCR using primers VSDH-6 (50-GGCCATTGCCCTTATCCAGGTG-30) and VSDH-7 (50-GTGGTCGAAAGGCTGCATGAATG-30), and inserted into pMD20, using the Mighty TA-cloning Kit (Takara Bio Inc., Kusatsu, Japan), yielding the plasmid pMDBzfC. A 1.58 kb XbaI-BamHI fragment was excised from pMDBzfC and inserted into pBluescriptKS+ using the Ligation-Convenience Kit (Nippon Gene, Tokyo, Japan), yielding the plasmid pKSBzfC. An 8.5 kb fragment was ampliﬁed from LLC-1 genomic DNA using the primers VSDH-13 (50-GGTATCGATAAGCTTGATCCAGTCATGAGTGGTTTTCCAGG-30) and VSDH-14 (50-GGGCTGCAGGAATTCGATCTGTTGGCTTGTTCGGGCAATAC-30) and PrimeSTAR GXL DNA Polymerase (Takara Bio Inc.). The PCR fragment was treated with Cloning Enhancer and inserted into pBluescriptKS+, digested with EcoRV, using an In-Fusion® HD Cloning Kit (Clontech Laboratories, Mountain View, CA, USA) to yield the plasmid pKSBzf1. The consistency of the sequence of each plasmid with the genomic sequence was checked by DNA sequencing. All plasmids were transformed into E. coli JM109 cells.

2.4. Metabolism of LLCs by Bacterial Strains E. coli cells grown in TB were harvested by centrifugation and suspended to a turbidity of 2.0 at 600 nm in 50 mM phosphate buﬀer (pH 7.5) containing 20 mM glucose. Cell suspensions (20 mL) in 200 mL ﬂasks were incubated at 30 ◦C on a rotary shaker (120 rpm) in the presence of substrates (1 mM). After 12 h incubation, the pH of the sample was adjusted to 2.0–3.0 with 1 M HCl and then extracted with 1.5-fold volumes of ethyl acetate. The organic extracts were dried over anhydrous sodium sulfate and concentrated under vacuum at 37 ◦C. The residual extracts were dried with anhydrous sodium sulfate and concentrated using a rotary evaporator. The samples were dissolved in methanol and subsequently subjected to gas chromatography-mass spectrometry (GC-MS) analysis.

2.5. Analytical Procedures GC-MS analysis was performed on a Shimadzu model QP-5050A (Shimadzu Co., Kyoto, Japan) platform equipped with a Phenomenex ZB-5 capillary column (inside diameter: 0.25 mm; ﬁlm thickness: 0.25 mm; length: 30 m) (Phenomenex, Torrance, CA, USA). The column temperature was programmed at 80 ◦C for 2 min, from 80 to 200 ◦C at 15.0 ◦C/min, from 200 to 240 ◦C at 10.0 ◦C/min. For trimethylsilyl (TMS) derivatization, samples were dried under vacuum and treated with 30 µL of Genes 2020, 11, 1416 4 of 13 Genes 2020, 11, x FOR PEER REVIEW 4 of 13

programmed at 80 °C for 2 min, from 80 to 200 °C at 15.0 °C/min, from 200 to 240 °C at 10.0 °C/min. N-Methyl-N-trimethylsilyltriﬂuoroacetamide (MSTFA) and 7.5 µL of pyridine at room temperature For trimethylsilyl (TMS) derivatization, samples were dried under vacuum and treated with 30 μL of for 30N-Methyl- min. OxygenN-trimethylsilyltrifluoroacetamide consumption during the (MSTFA) metabolism and 7.5 of μ aromaticL of pyridine compounds at room temperature by bacteria was measuredfor 30 min. as described Oxygen consumption previously [during15], and the dissolved metabolism dioxygen of aromatic was compounds monitored by using bacteria a Clark-type was polarographicmeasured as O described2 electrode previously (Yellow [15], Springs and diss Instrumentsolved dioxygen Model was 5300A, monitored Yellow using Springs, a Clark-type OH, USA). To achievepolarographic this, 3 mLO2 electrode of cell suspension (Yellow Springs in a chamberInstruments was Mo incubateddel 5300A, at Yellow 30 ◦C Springs, on a magnetic OH, USA). stirrer in the presenceTo achieve of this, the 3 substrate. mL of cell suspension in a chamber was incubated at 30 °C on a magnetic stirrer in the presence of the substrate. 3. Results 3. Results 3.1. Phylogenetic Analysis of Strain LLC-1 3.1. Phylogenetic Analysis of Strain LLC-1 A phylogenetic tree of strain LLC-1 and various Pseudomonas species is shown in Figure1. The 16S A phylogenetic tree of strain LLC-1 and various Pseudomonas species is shown in Figure 1. The rRNA of strain LLC-1 was almost identical to that of P. putida GB-1 and P. putida F1 (99.9%), with one or 16S rRNA of strain LLC-1 was almost identical to that of P. putida GB-1 and P. putida F1 (99.9%), with two base substitutions, compared with that of P. putida GB-1 or P. putida F1. From the topology of the one or two base substitutions, compared with that of P. putida GB-1 or P. putida F1. From the topology phylogeneticof the phylogenetic tree, strain tree, LLC-1 strain clustered LLC-1 clustered clearly withclearly the withP. putida the P. speciesputida species (Figure (Figure1), with 1), thewith exception the of P.exception putida NBRC14164, of P. putida NBRC14164 which is distant, which from is distant other fromP. putida otherstrains P. putida [22 strains]. Comparison [22]. Comparison of the genomeof sequencesthe genome available sequences in the available RAST data in th setse RAST revealed data thatsets P.revealed putida GB-1that P. is putida the closest GB-1 neighboris the closest of strain LLC-1,neighbor followed of strain by P.LLC-1, putida followedF1 [16 by]. TheseP. putida results F1 [16]. are These consistent results are with consistent the 16S with rRNA the 16S phylogenetic rRNA analysis.phylogenetic The 16S analysis. rRNA of The other 16S species rRNA of exhibited other specie< 99.4%s exhibited similarity < 99.4% to that similarity of strain to that LLC-1, of strain showing a distantLLC-1, relationship showing a with distant strain relati LLC-1.onship with strain LLC-1.

FigureFigure 1. 16S 1. rRNA16S rRNA phylogenetic phylogenetic tree oftreePseudomonas of Pseudomonasstrains strains (GenBank (GenBank/ENA/DDBJ/ENA/DDBJ Accession Accession Number); PseudomonasNumber);sp. Pseudomonas strain LLC-1 sp. strain (NZ_LUVY00000000), LLC-1 (NZ_LUVY00000000),P.putida KT2440 P. putida (AE015451), KT2440 (AE015451),P.putida F1 P. (CP000712),putida P.putidaF1 (CP000712),GB-1 (CP000926), P. putidaP.putida GB-1NBRC14164 (CP000926), (AP013070), P. putida P.aeruginosaNBRC14164 PAO1(NC_002516),(AP013070), P. aeruginosaP.mendocina ympPAO1(NC_002516), (CP000680), P. ﬂuorescens P. mendocinaPf-5 (CP000076),ymp (CP000680),P. stutzeri P. fluorescensA1501 (NC_009434),Pf-5 (CP000076),P. syringaeP. stutzeripv A1501 tomato str DC3000(NC_009434), (AE016853) P. syringae and P. emtomophilapv tomato strL48 DC3000 (CT573326). (AE016853) The and numbers P. emtomophila in parentheses L48 (CT573326). indicate The variants foundnumbers in one in strain. parentheses Multiple indicate sequence variants alignment found in outputsone strain. of Multiple the 16S sequence rRNA sequence alignment (1528 outputs bp) were of the 16S rRNA sequence (1528 bp) were used to generate neighbor-joining phylogenetic trees using used to generate neighbor-joining phylogenetic trees using MEGA X [21]. The bar indicates expected MEGA X [21]. The bar indicates expected nucleotide substitutions per site. Numbers indicate the nucleotide substitutions per site. Numbers indicate the percentage occurrence of the branch in the percentage occurrence of the branch in the bootstrapped trees of 500 replicates. bootstrapped trees of 500 replicates.

Genes 2020, 11, 1416 5 of 13

3.2. Detection of Catabolic Genes for Aromatic Compounds in Strain LLC-1 Catabolic genes for various aromatic acids were identified from the draft genome sequence of strain LLC-1: anthranilate catabolic atn gene, benzoate catabolic ben gene, catechol catabolic cat gene, 4-hydroxybenzoic acid catabolic pob gene, 4-hydroxyphenylacetic acid catabolic, hpa gene, homogentisate catabolic hmg gene, phenylacetate catabolic paa gene, protocatechuate catabolic pca gene, and vanillate catabolic van genes were found. Aromatic compound catabolic genes found in the genome sequence of strain LLC-1 are shown in Table S1. Protocatechuate, homogentisate, and benzoate are known as key intermediates of funneling pathways for aromatic ring degradation. The van gene encodes a two-component vanillate-O-demethylase (VanAB), and a regulatory component involved in oxidative demethylation of vanillic acid. The pca gene encodes β-ketoadipate pathway enzymes following ring-cleavage reactions, which subsequently enter the tricarboxylic acid cycle. Despite the close phylogenetic relationship to P. putida, strain LLC-1 lacked certain genes usually found in the P. putida genome; strain LLC-1 does not possess the fer gene, which encodes the ferulic acid metabolic pathway widely distributed in the genome of P. putida [14], and the vdh gene encoding an NAD+-dependent aldehyde dehydrogenase that specifically oxidizes its catabolic intermediate, vanillin. In addition, strain LLC-1 was deficient in genes encoding the isoeugenol (iem) catabolic pathways detected in the genome of a minor group of P. putida [12]. We attempted to search the draft genome sequence of the LLC-1 strain for a gene encoding vanillin dehydrogenase, which converts vanillin to vanillic acid. Annotation by RAST detected 47 open reading frames (ORFs) identified as aldehyde dehydrogenase. Two of these were benzaldehyde dehydrogenases. A BLASTX search was performed on one of the open reading frames present in the vicinity of the van operon encoding vanillate demethylase. The ORF showed 65% identity with the vdh gene encoding vanillin dehydrogenase of P. putida strain IE27 [12], which is a component of the isoeugenol catabolic gene cluster. The ORF identified as benzaldehyde dehydrogenase (ORF4) was adjacent to the LysR family of transcriptional regulators (ORF1), benzoylformate decarboxylase (ORF2), putative major facilitator superfamily (MFS) transporters that are related to benzoate transporter BenK of Acinetobacter sp. [24] (ORF3), and outer membrane OprD family porin protein (ORF5), which is related to the putative phenylacetic acid uptake porin PhaK of P. putida [25] (Figure2). A BLASTX search showed that ORF2 exhibited 84% and 63% identity at the amino acid level to thiamine diphosphate-dependent benzoylformate decarboxylase from P. fluorescens (PfBFDC) [26] and P. putida (MdlC) [27], respectively. MdlC is known to be involved in mandelic acid metabolism in mandelic acid-degrading P. putida. From these results, it was presumed that this gene cluster is involved in the transport and catabolism of benzoylformic acid. The gene clusters encoded by these five ORFs were named bzfR, bzfA, bzfB, bzfC, and bzfD, as shown in Figure2. Genes closely related to the bzf gene cluster in P. fluorescens Pf-5 [26] and P. aeruginosa [28] have been reported, including benzoylformate decarboxylase, MFS transporter, aldehyde dehydrogenase, and porin protein. The gene organization of the regions corresponding to bzfRABCD of strain LLC-1 is shown in Figure2. The five components of the bzf gene cluster showed 78–81% identity with those of P. fluorescens Pf-5, and 73–80% identity with those of P. aeruginosa PAO1, respectively, with the exception of bzfB whose homolog was not found for Pf-5. From the genome sequences of P. putida F1, P. putida GB-1, and P. putida KT2440, a gene cluster showing identity with the bzf gene cluster was not detected (Figure2). Genes 2020, 11, x FOR PEER REVIEW 6 of 13

Genes 2020, 11, 1416 6 of 13 Genes 2020, 11, x FOR PEER REVIEW 6 of 13

Figure 2. Physical map of benzoylformate catabolic bzf gene cluster and vanillate cetabolic van operon of Pseudomoas sp. strain LLC-1 and other pseudomonas strains. 1, vanK; 2, bzfR (ORF1); 3, bzfA (ORF2); Figure 2. Physical map of benzoylformate catabolic bzf gene cluster and vanillate cetabolic van operon 4, bzfB (ORF3); 5, bzfC (ORF4); 6, bzfD (ORF5); 7, vanR; 8, vanA; 9, vanB;. ORFs with conserved sequence ofFigurePseudomoas 2. Physicalsp. strain map LLC-1of benzoylformate and other pseudomonas catabolic bzf strains.gene cluster 1, vanK and; 2, vanillatebzfR (ORF1); cetabolic 3, bzfA van(ORF2); operon (>70% identity with those of LLC-1 within bzf gene cluster or van operon) are filled with the same 4,ofbzfB Pseudomoas(ORF3); 5,sp.bzfC strain(ORF4); LLC-1 6, andbzfD othe(ORF5);r pseudomonas 7, vanR; 8, vanA strains.; 9, vanB1, vanK;. ORFs; 2, bzfR with (ORF1); conserved 3, bzfA sequence (ORF2); color. The shading in pink to red shows the identity (74–100%) of the gene clusters as indicated on the (>4,70% bzfB identity (ORF3); with 5, bzfC those (ORF4); of LLC-1 6, bzfD within (ORF5);bzf gene 7, vanR cluster; 8, vanA or van; 9,operon) vanB;. ORFs are filled with withconserved the same sequence color. left of the figure. The(>70% shading identity in pink with to those red shows of LLC-1 the identitywithin bzf (74–100%) gene cluster of the or gene van clusters operon) as are indicated filled with on the the left same of thecolor. figure. The shading in pink to red shows the identity (74–100%) of the gene clusters as indicated on the 3.3. Biotransformation of LLCs by E.coli Cells Expressing bzf Gene Cluster left of the figure. 3.3. Biotransformation of LLCs by E.coli Cells Expressing bzf Gene Cluster We cloned the bzfABCD gene cluster (8.5 kb locus adjacent to the van operon) and expressed it in3.3. E. WecoliBiotransformation clonedto investigate the bzfABCD of the LLCs functiongene by E. clustercoli of Cellsthe (8.5 bzf Expressing kb gene locus products. adjacentbzf Gene The Cluster to the resultsvan operon) of GC-MS and analysis expressed of itthe in reactionE. coli Weto investigateproducts cloned the after the bzfABCD functionthe degradation gene of the clusterbzf geneof (8.5 various products. kb locus aromatic Theadjacent results carboxylic to ofthe GC-MS van acids operon) analysis by restingand of theexpressed reactioncells of it recombinantproductsin E. coli afterto E.investigate thecoli degradation JM109 the (pKSBzf1) function of various ofexpressing the aromatic bzf gene the carboxylic products.bzfABCD acidsThegene results byare resting shown of GC-MS cells in Figure of analysis recombinant 3. When of the benzoylformicE.reaction coli JM109 products (pKSBzf1) acid wasafter expressing usedthe degradationas a thesubstrate,bzfABCD of variousin genethe vector arearomatic shown control carboxylic in Figureobtained 3acids. Whenby incubatingby benzoylformicresting E.cells coli of carryingacidrecombinant was pBluescript used E. as acoli substrate, KS+,JM109 the (pKSBzf1) in degradation the vector expressing control of benzoylformic obtained the bzfABCD by acid incubating gene did notareE. occur.shown coli carrying Inin contrast,Figure pBluescript 3. E. When coli cellsKSbenzoylformic+ ,carrying the degradation pKSBzf1 acid was ofcompletely benzoylformic used as converted a substrate, acid didbenzoylformic in not the occur. vector Into contrast,controlbenzoic obtainedacid.E. coli Thecells bzfby carryinggeneincubating cluster pKSBzf1 E. was coli suggestedcompletelycarrying pBluescriptto converted be involved KS+, benzoylformic thein benzoylformicdegradation to benzoic of acidbenzoylformic acid. catabolism. The bzf acid geneAlthough did clusternot occur.BzfA was Inis suggested contrast,presumed toE. to becoli decarboxylateinvolvedcells carrying in benzoylformic vanillic pKSBzf1 acid completely to acid guaiacol, catabolism. converted the reaction Although benzoylformic products BzfA vanillic to is benzoic presumed acid acid. and to theThe decarboxylate by-product bzf gene cluster vanillyl vanillic was alcoholacidsuggested to guaiacol,were todetected, be the involved reaction whereas productsin guaiacolbenzoylformic vanillic was not acid aciddetected and catabolism. the by-productwhen E. Althoughcoli vanillyl cells carrying BzfA alcohol is pKSbzf1 were presumed detected, were to incubatedwhereasdecarboxylate guaiacol with vanillicvanillin. was not acid Therefore, detected to guaiacol, whendecarboxylation theE. reaction coli cells products carryingby BzfA vanillic pKSbzf1is likely acid wereto andbe incubatedbenzoylformatic the by-product with vanillin. vanillyl acid- specific.Therefore,alcohol were decarboxylation detected, whereas by BzfA guaiacol is likely was to benot benzoylformatic detected when E. acid-specific. coli cells carrying pKSbzf1 were incubated with vanillin. Therefore, decarboxylation by BzfA is likely to be benzoylformatic acid- specific.

Figure 3. Conversion of benzoylformate by E. coli cells carrying pKSBzf1 (black line) or E. coli cells Figure 3. Conversion of benzoylformate by E. coli cells carrying pKSBzf1 (black line) or E. coli cells carrying pBluescriptKS+ (yellow line). Peak 1, benzoic acid; 2, benzoylformic acid. The products were carrying pBluescriptKS+ (yellow line). Peak 1, benzoic acid; 2, benzoylformic acid. The products were derivatized with MSTFA and identified comparing the retention time and mass spectra of authentic derivatized with MSTFA and identified comparing the retention time and mass spectra of authentic trimethylsilylatedFigure 3. Conversion standards. of benzoylformatem/z values of by each E. peakcoli cells are listedcarrying in SupplementalpKSBzf1 (black Table line) S2. or E. coli cells trimethylsilylated standards. m/z values of each peak are listed in Supplemental Table S1. carrying pBluescriptKS+ (yellow line). Peak 1, benzoic acid; 2, benzoylformic acid. The products were BzfCderivatized, a structural with MSTFA gene and of theidentifiedbzf gene comparing cluster, th hase retention been identifiedtime and mass as anspectra NAD of+ authentic-dependent BzfC, a structural gene of the bzf gene cluster, has been identified as an NAD+-dependent dehydrogenase.trimethylsilylated It is proposedstandards. m/z tobe values involved of each in peak the are oxidation listed in Supplemental of benzaldehyde Table producedS1. by the dehydrogenase. It is proposed to be involved in the oxidation of benzaldehyde produced by the decarboxylation of benzoylformic acid. Strain LLC-1 lacked the ferulic acid catabolic fer gene, and the decarboxylation of benzoylformic acid. Strain LLC-1 lacked the ferulic acid catabolic fer gene, and the vdh geneBzfC encoding, a structural an NAD gene+-dependent of the bzf aldehyde gene cluster, dehydrogenase has been specific identified for vanillin. as an NAD It was+-dependent confirmed dehydrogenase. It is proposed to be involved in the oxidation of benzaldehyde produced by the decarboxylation of benzoylformic acid. Strain LLC-1 lacked the ferulic acid catabolic fer gene, and the

Genes 2020, 11, 1416 7 of 13 that bzfC shows 65% identity with the vdh gene, encoding the vanillin dehydrogenase component of the isoeugenolGenes 2020 catabolic, 11, x FOR PEER gene REVIEW cluster in P. putida IE27, as described above. Therefore,7 of we 13 assumed that bzfC vdhis involvedgene encoding in the an oxidizationNAD+-dependent of bothaldehyde benzaldehyde dehydrogenase and specific vanillin. for vanillin. We investigatedIt was the transformationconfirmed of benzaldehyde,that bzfC shows 65% vanillin, identity isovanillin, with the vdh and gene, syringaldehyde encoding the vanillin in E. dehydrogenase coli cells expressing the bzfC gene.component In the vector of the control,isoeugenolE. catabolic coli carrying gene cluster pBluescript in P. putida KS IE27,+ reduced as described all aldehydesabove. Therefore, investigated to we assumed that bzfC is involved in the oxidization of both benzaldehyde and vanillin. We the corresponding alcohol completely (Figure4), a reaction assumed to be catalyzed by benzaldehyde investigated the transformation of benzaldehyde, vanillin, isovanillin, and syringaldehyde in E. coli reductasecells from expressingE. coli hostthe bzfC cells gene. [29 In]. the After vector the control, conversion E. coli carrying of benzylaldehyde pBluescript KS+ and reduced vanillin all for 12 h, in reactionaldehydes mixtures investigated containing to the recombinant correspondingE. colialcoholcarrying completely plasmid (Figure pKSBzfC, 4), a reaction benzoic assumed acid, to be and vanillic acid insteadcatalyzed of benzyl by benzaldehyde alcohol and reductase vanillyl from alcohol E. coli were host detected cells [29]. (Figure After the4A,B). conversion The conversion of of benzylaldehyde and vanillin for 12 h, in reaction mixtures containing recombinant E. coli carrying isovanillin by bzfC-expressing E. coli resulted in the formation of the corresponding carboxylic acids, plasmid pKSBzfC, benzoic acid, and vanillic acid instead of benzyl alcohol and vanillyl alcohol were whereas partdetected of syringaldehyde (Figure 4A, B). The was conversion transformed of isovanillin into syringic by bzfC-expressing acid (Figure E. coli4C,D). resulted From in thesethe results, it was conﬁrmedformation thatof the oxidization corresponding by carboxylic aldehyde acids, dehydrogenase whereas part of syringaldehyde encoded by bzfCwas transformed(BzfC) is not limited to benzaldehydeinto syringic produced acid (Figure by 4C, benzoylformic D). From these results, acid decarboxylization,it was confirmed that oxidization and is expanded by aldehyde to vanillin, dehydrogenase encoded by bzfC (BzfC) is not limited to benzaldehyde produced by benzoylformic isovanillin, and syringaldehyde. acid decarboxylization, and is expanded to vanillin, isovanillin, and syringaldehyde.

(A) (×1,000,000) 10.0 TIC 7.5 3 4

5.0

Intensity 2.5

0.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0 10.0 11.0 12.0 13.0 Retention time [min] (B) (×1,000,000) TIC 6.0 5 6 4.0

Intensity 2.0

0.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0 10.0 11.0 12.0 13.0 Retention time [min]

Intensity 2.0

0.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0 10.0 11.0 12.0 13.0 (D) Retention time [min] (×1,000,000) TIC 7.5 9 10 5.0

Intensity 2.5

0.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0 10.0 11.0 12.0 13.0 Retention time [min] Figure 4. Conversion of bezylaldehyde (A), vanillin (B), isovanillin (C), and syringaldehyde (D) by

E. coli cells carrying pKSBzfC (black line) or E. coli cells carrying pBluescriptKS+ (yellow line). Peak 3, benzyl alcohol; 4, benzoic acid; 5, vanillyl alcohol; 6, vanillic acid; 7, isovanillyl alcohol, 8; isovanillic acid; 9, syringic alcohol; 10, syringic acid. The products were derivatized with MSTFA and identiﬁed by comparing the retention time and mass spectra of authentic standards. m/z values of each peak are listed in Supplemental Table S2. Genes 2020, 11, x FOR PEER REVIEW 8 of 13

Figure 4. Conversion of bezylaldehyde (A), vanillin (B), isovanillin (C), and syringaldehyde (D) by E. coli cells carrying pKSBzfC (black line) or E. coli cells carrying pBluescriptKS+ (yellow line). Peak 3, benzyl alcohol; 4, benzoic acid; 5, vanillyl alcohol; 6, vanillic acid; 7, isovanillyl alcohol, 8; isovanillic acid; 9, syringic alcohol; 10, syringic acid. The products were derivatized with MSTFA and identified by comparing the retention time and mass spectra of authentic standards. m/z values of each peak are Genes 11 listed2020, in, Supplemental 1416 Table S1. 8 of 13

3.4.3.4. Induction Induction of of Benzoylformic Benzoylformic Acid Acid Metabolism Metabolism The metabolic ability of resting cells of strain LLC-1 grown in a medium containing The metabolic ability of resting cells of strain LLC-1 grown in a medium containing benzoylformic benzoylformic acid or pyruvic acid as the sole carbon source was compared by measuring oxygen acid or pyruvic acid as the sole carbon source was compared by measuring oxygen consumption rates consumption rates (Figure 5). Strain LLC-1 grown on benzoylformic acid as the sole carbon source (Figure5). Strain LLC-1 grown on benzoylformic acid as the sole carbon source showed a higher oxygen showed a higher oxygen uptake rate in samples incubated with benzoylformate, benzaldehyde, and uptake rate in samples incubated with benzoylformate, benzaldehyde, and benzoic acid, compared to benzoic acid, compared to cells grown on pyruvate as a carbon source. This result indicated that cells grown on pyruvate as a carbon source. This result indicated that strain LLC-1 can metabolize strain LLC-1 can metabolize benzoylformic acid and its catabolic intermediate benzaldehyde and benzoylformic acid and its catabolic intermediate benzaldehyde and benzoic acid more eﬃciently benzoic acid more efficiently under induction by benzoylformic acid. under induction by benzoylformic acid.

g/min) μ

Oxygen consumption rate ( Oxygen BFA BAH BA Substrate

Figure 5. Oxygen consumption by strain LLC-1 grown with benzoylformic acid (white box) or pyruvic Figure 5. Oxygen consumption by strain LLC-1 grown with benzoylformic acid (white box) or pyruvic acid (black box). The cell suspensions (at OD600 of 2.0) were incubated with 150 µM benzoylformic acid acid (black box). The cell suspensions (at OD600 of 2.0) were incubated with 150 μM benzoylformic (BFA), benzaldehyde (BAH), or benzoic acid (BA). Dissolved dioxygen in a chamber was monitored acid (BFA), benzaldehyde (BAH), or benzoic acid (BA). Dissolved dioxygen in a chamber was as described in Materials and Methods. Experiments were performed in triplicates, and standard monitored as described in Materials and Methods. Experiments were performed in triplicates, and deviations are displayed as error bars. standard deviations are displayed as error bars. 3.5. Growth on Mandelic Acid 3.5. Growth on Mandelic Acid Benzoylformic acid, a substrate for benzoylformate decarboxylase encoded by bzfA, is known to be a catabolicBenzoylformic intermediate acid, ina substrate the mandelate for benzoylformate pathway encoded decarboxylase by the mdl geneencoded of P. by putida bzfA[, 27is ].known The bzfA to begenes a catabolic showed intermediate 63% identity in at the the mandelate amino acid pathway level with encoded the mdlC bygenes, the mdl but gene the of homologous P. putida [27]. gene The for bzfAmdlA genesencoding showed mandelate 63% identity racemase at the and amino the mdlB acidgene level encoding with the ( SmdlC)-mandelate genes, but dehydrogenase the homologous were genenot present for mdlA in the encoding genome of strainmandelate LLC-1. racemase The growth and of strainthe mdlB LLC-1 gene and the encoding mandelic acid-degrading(S)-mandelate dehydrogenasebacterial strain, wereP. putida not NBRCpresent 14164, in the ingenome a medium of strain containing LLC-1. ( R,SThe)-mandelic growth of acidstrain or LLC-1 benzoylformic and the mandelicacid as the acid-degrading sole carbon source bacterial were examinedstrain, P. (Figureputida 6NBRC). Strain 14164 LLC-1, in and a mediumP. putida containingNBRC 14164 (R,S were)- mandelicable to grow acid on or benzoylformicbenzoylformic acidacid asas athe sole sole carbon carbon source, source but were strain examined LLC-1was (Figure not able6). Strain to grow LLC- on 1mandelic and P. putida acid. NBRC From these14164 results, were able it was to grow concluded on benzoylformic that strain LLC-1 acid doesas a sole not producecarbon source, mandelate but strainracemase LLC-1 and was mandelate not able dehydrogenase,to grow on mandelic which acid transforms. From these mandelate results, to it benzoylformicwas concluded acid.that strain LLC-1 does not produce mandelate racemase and mandelate dehydrogenase, which transforms mandelate to benzoylformic acid.

Genes 2020, 11, 1416 9 of 13 Genes 2020, 11, x FOR PEER REVIEW 9 of 13

(A)

1.2 600 0.8

0.4 Growth OD Growth 0 0481216 Time (h) (B) 1.2

600 0.8

0.4

Growth OD Growth 0 0481216 Time (h)

Figure 6. Growth of strain LLC-1 and P.putida NBRC 14164 in culture medium containing benzoylformic Figure 6. Growth of strain LLC-1 and P. putida NBRC 14164 in culture medium containing acid (A) and (R.S)-mandelic acid (B). The circles (red) represent the growth profile of strain LLC-1 and benzoylformic acid (A) and (R.S)-mandelic acid (B). The circles (red) represent the growth profile of squares (black) represent the growth profile of P. putida NBRC 14164. Experiments were performed in strain LLC-1 and squares (black) represent the growth profile of P. putida NBRC 14164. Experiments triplicates, and standard deviations are displayed as error bars. were performed in triplicates, and standard deviations are displayed as error bars. 4. Discussion 4. Discussion In this study, the draft genome sequence of strain LLC-1 was used to characterize the bzf gene clusterIn involvedthis study, in the the draft catabolism genome of sequence LLCs. The of sequence strain LLC-1 of the was 16S used rRNA to genecharacterize indicated the that bzf straingene clusterLLC-1 isinvolved phylogenetically in the catabolism close to ofP. LLCs. putida The(Figure sequence1), but of the the overall 16S rRNA genomic gene indicated structure that exhibited strain LLC-1variation is phylogenetically between strain LLC-1 close to and P.P. putida putida. (FigureLLC-1 1), does but notthe possessoverall genomic the fer gene, structure involved exhibited in the variationferulic acid between metabolic strain pathway LLC-1 widelyand P. distributedputida. LLC-1 in thedoes genome not possess of P. putidathe fer[ 11gene,], and involved the vdh ingenes the ferulicencoding acid a dehydrogenasemetabolic pathway that widely specifically distributed oxidizes in itsthe catabolic genome intermediate,of P. putida [11], vanillin. and the In vdh addition, genes encodingstrain LLC a wasdehydrogenase deficient in that genes specifically encoding oxidizes the isoeugenol its catabolic (iem) catabolicintermediate, pathways vanillin. detected In addition, in the straingenome LLC of awas sub-group deficient of inP. putidagenes [encoding12], as well the as isoeugenol the mdl gene (iem encoding) catabolic the mandelicpathways acid detected degradation in the genomepathway of [22 a,27 sub-group]. On the other of P. hand, putida strain [12], LLC-1 as well carried as the the bzfmdlgene gene cluster encoding encoding the mandelic benzoylformic acid degradationacid catabolic pathway enzymes [22,27]. and transporters On the other that hand, do not strain exist LLC-1 in the carriedP. putida thegenome. bzf gene cluster encoding benzoylformicIn this study, acid the catabolicbzf gene enzymes cluster and of strain transporters LLC-1 wasthat showndo not exist to be in involved the P. putida in the genome. conversion of benzoylformicIn this study, acidthe bzf to benzoicgene cluster acid of (Figure strain7 LLC-1). Benzoic was acidshown can to be be degraded involvedto intricarboxylic the conversion acid of benzoylformiccycle intermediates acid to via benzoic the β-ketoadipate acid (Figure pathway7). Benzoic encoded acid can by be the degradedpca gene to intricarboxylic the genome acid of straincycle intermediatesLLC-1 (Table S1) via [ 30the]. β The-ketoadipate first ORF atpathway the end encoded of the bzf bygene the cluster pca gene was in identified the genome as aof transcriptional strain LLC-1 (Tableregulator S1) ([30].bzfR ).The Benzoylformic first ORF at acid-metabolizingthe end of the bzf gene enzymes cluster were was induced identified by benzoylformicas a transcriptional acid regulatorin strain LLC-1(bzfR). (FigureBenzoylformic5), and acid-metabolizingbzfR is a putative enzymes transcriptional were induced regulator by ofbenzoylformic benzoylformic acid acid in straincatabolic LLC-1 enzymes. (Figure We 5), previously and bzfR reportedis a putative that vanillin-catabolizingtranscriptional regulator enzymes of benzoylformic are not induced acid by catabolicvanillin in enzymes. strain LLC-1 We [previously15]. This may reported be due that to poor vanillin-catabolizing interaction between enzymes the putative are not transcriptional induced by vanillinregulator in BzfR strain and LLC-1 vanillin [15]. or This its metabolite,may be due vanillicto poor acid.interaction This transcriptional between the putative regulator transcriptional is likely to be regulatorbenzoylformic BzfR acid-specific.and vanillin or The its second metabolite, ORF ( bzfAvanillic) in theacid.bzf Thisgene transcriptional cluster showed regulator 84% or 63% is likely identity to beat thebenzoylformic amino acid levelacid-spe withcific. thiamine The second diphosphate-dependent ORF (bzfA) in the bzf benzoyldormate gene cluster showed decarboxylase 84% or from63% identityP. putida (MdlC)at the or aminoP. fluorescence acid (PfBFDC),level with respectively thiamine [26diphosphate-depe,27]. Three-dimensionalndent structuresbenzoyldormate of both decarboxylaseMdlC (PDB entry from code: P. 1BFD)putida [31(MdlC)] and PfBFDCor P. fluorescence (6QSI) [26 ](PfBFDC), are available. respectively They are enzymes[26,27]. Three- whose dimensionalreaction mechanisms structures have of beenboth preciselyMdlC (PDB investigated entry code: based 1BFD) on their [31] structure and PfBFDC [32,33 ].(6QSI) The results[26] are of available.the biotransformation They are enzymes study (Figure whose3 reaction) and comparison mechanisms of sequences have been with precisely these known investigated benzoylformate based on theirdecarboxylases structure [32,33]. (Figure 2The) supported results of the the hypothesis biotransformation that BzfA study is an (Figure enzyme 3) thatand decarboxylatescomparison of sequencesbenzoylformic with acid.these The know fourthn benzoylformate ORF (bzfC) located decarbox betweenylasesbzfB (Figureand 2)bzfD supportedshowed the 65% hypothesis identity at that the BzfAamino is acidan enzyme level with that the decarboxylatesmdlD gene encoding benzoylformic benzaldehyde acid. The dehydrogenase fourth ORF (bzfC of P.) located putida [between27]. It is bzfB and bzfD showed 65% identity at the amino acid level with the mdlD gene encoding benzaldehyde dehydrogenase of P. putida [27]. It is obvious that bzfA and bzfC are homologous to

Genes 2020, 11, x FOR PEER REVIEW 10 of 13 Genes 2020, 11, 1416 10 of 13 mdlA and mdlC, respectively. However, the overall structures of the bzf and mdl gene clusters are completely different. MdlA and mdlC are encoded in the opposite direction to each other and are locatedobvious on that a clearlybzfA and differentbzfC are transcri homologousption unit. to mdlAIt is presumedand mdlC that, respectively. the mdl gene However, cluster underwent the overall reorganizationstructures of the duringbzf and evolution.mdl gene Transporter clusters are membra completelyne proteins different. are usuallyMdlA requiredand mdlC forare the encoded uptake ofin aromatic the opposite carboxylic direction acids to into each cells other through and are the located cell membrane on a clearly [34]. di Thefferent BenK transcription-like MFS transporter unit. It is encodedpresumed by that the the thirdmdl geneORF, cluster bzfB, underwentand the PhaK-like reorganization outer duringmembrane evolution. OprD Transporterfamily porin membrane protein encodedproteins by are the usually fifth ORF, required bzfD for, are the putative uptake transporters of aromatic carboxylicspecific for acids benzoylformic into cells throughacid (Figure the cell7). Theymembrane may transport [34]. The BenK-likebenzoylformic MFS transporteracid in strain encoded LLC-1, by as the well third as in ORF, recombinantbzfB, and the E. PhaK-likecoli expressing outer themembrane bzf gene OprD cluster. family On the porin other protein hand, encoded vanillin by is the proposed fifth ORF, to bzfDbe incorporated, are putative via transporters simple diffusion specific infor strain benzoylformic LLC-1 and acid recombinant (Figure7). E. They coli cells may due transport to its hydrophobic benzoylformic nature acid in[35]. strain LLC-1, as well as in recombinant E. coli expressing the bzf gene cluster. On the other hand, vanillin is proposed to be incorporated via simple diffusion in strain LLC-1 and recombinant E. coli cells due to its hydrophobic nature [35].

O O O O O H OH OH OH BzfB BzfA BzfC

OH 4-Hydroxy benzoic acid Benzoylformic BzfD Benzaldehyde Benzoic acid acid

O O H O OH OH BzfC VanAB TCA cycle OH OCH OCH3 β-ketoadipate 3 OH OH OH pathway Vanillic acid Protocatechuic acid Vanillin om im

CH2 O OH CH3 O H

OCH 3 OCH3 OCH3 OCH3 OH OH OH OH Eugenol Ferulic acid Vanillin Isoeugenol

Figure 7. Putative metabolic pathways of benzoylformic acid and vanillin in strain LLC-1. Enzymes: Figure 7. Putative metabolic pathways of benzoylformic acid and vanillin in strain LLC-1. Enzymes: BzfA, benzoylformate carboxylase; BzfC, benzaldehyde dehydrogenase/vanillin dehydrogenase; VanAB, BzfA, benzoylformate carboxylase; BzfC, benzaldehyde dehydrogenase/vanillin dehydrogenase; vanillin-O-demethylase. Transporters: BzfB, BenK-like MFS transporter; BzfD, PhaK-like outer VanAB, vanillin-O-demethylase. Transporters: BzfB, BenK-like MFS transporter; BzfD, PhaK-like membrane porin. Structures: om, outer membrane; im, inner membrane. BzfC which convert both outer membrane porin. Structures: om, outer membrane; im, inner membrane. BzfC which convert benzaldehyde and vanillin was highlighted within boxes. Catabolic pathways for eugenol, ferulic acid both benzaldehyde and vanillin was highlighted within boxes. Catabolic pathways for eugenol, and isoeugenol of P. putida which are described in the ‘Introduction’ section, are represented in gray. ferulic acid and isoeugenol of P. putida which are described in the 'Introduction' section, are representedBenzoylformic in gray acid (Table is known S2). to be an intermediate of the mandelic acid catabolic pathway of P. putuda. Benzoylformic acid is produced through the isomerization and oxidation of mandelic Benzoylformic acid is known to be an intermediate of the mandelic acid catabolic pathway of P. acid and then undergoes decarboxylation and oxidization to form benzoic acid. Strain LLC-1 did putuda. Benzoylformic acid is produced through the isomerization and oxidation of mandelic acid not utilize (R,S)-mandelates as a carbon source (Figure6). In addition, ORFs coding for enzymes and then undergoes decarboxylation and oxidization to form benzoic acid. Strain LLC-1 did not involved in the metabolism of mandelic acid was not found. Therefore, the role of benzoylformic acid utilize (R,S)-mandelates as a carbon source (Figure 6). In addition, ORFs coding for enzymes involved catabolic enzymes of strain LLC-1 is different from that of P. putida, where strain LLC-1 assimilates in the metabolism of mandelic acid was not found. Therefore, the role of benzoylformic acid catabolic benzoylformic acid, and not as a metabolic intermediate of other carbon sources. Wei et al. recently enzymes of strain LLC-1 is different from that of P. putida, where strain LLC-1 assimilates studied the role of the benzoylformate decarboxylase of P. fluorescens Pf-5 in the degradation of lignin benzoylformic acid, and not as a metabolic intermediate of other carbon sources. Wei et al. recently depolymerization fragments. It was confirmed that benzoylformate decarboxylase does not act on studied the role of the benzoylformate decarboxylase of P. fluorescens Pf-5 in the degradation of lignin guaiacil-glycerol derivatives, an allyl C3 lignin-degrading fragment, and specifically decarboxylates

Genes 2020, 11, 1416 11 of 13 benzoylformic acid or 4-hydroxybenzoylformic acid, which represent aryl C2 substrates [26]. It was proposed that 4-hydroxybenzoylformic acid is formed during the degradation of lignin via oxidation of a keto-aldehyde precursor. Our results provided insight into the role of the benzoylformate pathway encoded by the bzf gene cluster of strain LLC-1 in lignin degradation via the formation of 4-hydroxybenzoylformic acid. The result that E. coli expressing the bzfC gene oxidized vanillin and syringaldehyde to the corresponding acids indicated that BzfC is a multifunctional enzyme that initiates oxidization of LLCs in strain LLC-1. The substrates of BzfC were not limited to benzaldehyde, which is a catabolic intermediate of benzoylformic acid. This result was consistent with our previous study in which the same substrates were transformed into their corresponding aromatic acids as dead-end products by strain LLC-1 [15]. Alternative use of BfzC in benzoylformate and vanillin catabolism may be due to adaptation to environmental niches. There are two other reasons to explain the role of BzfC in vanillin catabolism. First, the amino acid sequence encoded by the bzfC gene showed relatively high identity (65%) with that of vanillin dehydrogenase (Vdh) of P. putida IE27 [12]. Second, the bzf gene cluster is flanked by the van operon, which encodes a downstream vanillate-metabolizing enzyme (vanAB) and its transcriptional regulator (vanR). Cross-regulation between bzf and van genes may be feasible when the two genes are located at short distances from each other [36,37]. There are examples of studies that attempted to produce vanillin from ferulic acid by constructing a vanillin dehydrogenase-deficient microbial strain [13,14]. Construction of mutant strains deficient in bzfB, bzfC, or bzfD will be useful to determine the role of bzf in vanillin catabolism and benzoylformic acid transportation.

5. Conclusions In this study, we revealed that BzfC has a wide substrate range and is an important enzyme that initiates the oxidation of aromatic biomass (LLC) in strain LLC-1. Genome-wide characterization demonstrated that the strain LLC-1 catabolized benzoylformic acid and vanillin via a metabolic pathway that is independent of ferulic acid, eugenol, isoeugenol, or mandelic acid catabolic pathways, exhibiting a unique mode of dissimilation for the metabolic pathways related to biomass-derived aromatic compounds.

Supplementary Materials: The following are available online at http://www.mdpi.com/2073-4425/11/12/1416/s1, Table S1: Catabolic genes of aromatic compounds found in the draft genome of strain LLC-1. Table S2: List of data from GC/MS analysis. Author Contributions: Conceptualization and project administration, J.H.; Investigation, J.H. and R.T.; Data curation, M.M. and H.Y.; Visualization, R.T.; Writing original draft, J.H.; Review and editing, M.M. and H.Y. All authors have read and agree to the submitted version of the manuscript. Funding: This work was partly supported by the budget of the University of Miyazaki. Acknowledgments: We thank Jun Shimodaira for helpful discussions. Conﬂicts of Interest: The authors declare no conﬂict of interest.

References

1. Brink, D.P.; Ravi, K.; Lidén, G.; Gorwa-Grauslund, M.F. Mapping the diversity of microbial lignin catabolism: Experiences from the eLignin database. Appl. Microbiol. Biotechnol. 2019, 103, 3979–4002. [CrossRef] [PubMed] 2. Kamimura, N.; Takahashi, K.; Mori, K.; Araki, T.; Fujita, M.; Higuchi, Y.; Masai, E. Bacterial catabolism of lignin-derived aromatics: New ﬁndings in a recent decade: Update on bacterial lignin catabolism. Environ. Microbiol. Rep. 2017, 9, 679–705. [CrossRef][PubMed] 3. Calisti, C.; Ficca, A.G.; Barghini, P.; Ruzzi, M. Regulation of ferulic catabolic genes in Pseudomonas ﬂuorescens BF13: Involvement of a MarR family regulator. Appl. Microbiol. Biotechnol. 2008, 80, 475–483. [CrossRef] [PubMed] 4. Walton, N.J.; Mayer, M.J.; Narbad, A. Vanillin. Phytochemistry 2003, 63, 505–515. [CrossRef] Genes 2020, 11, 1416 12 of 13

5. Priefert, H.; Rabenhorst, J.; Steinbüchel, A. Biotechnological production of vanillin. Appl. Microbiol. Biotechnol. 2001, 56, 296–314. [CrossRef][PubMed] 6. Chen, H.-P.; Chow, M.; Liu, C.-C.; Lau, A.; Liu, J.; Eltis, L.D. Vanillin catabolism in Rhodococcus jostii RHA1. Appl. Environ. Microbiol. 2012, 78, 586–588. [CrossRef] 7. Linger, J.G.; Vardon, D.R.; Guarnieri, M.T.; Karp, E.M.; Hunsinger, G.B.; Franden, M.A.; Johnson, C.W.; Chupka, G.; Strathmann, T.J.; Pienkos, P.T.; et al. Lignin valorization through integrated biological funneling and chemical catalysis. Proc. Natl. Acad. Sci. USA 2014, 111, 12013–12018. [CrossRef] 8. Sonoki, T.; Morooka, M.; Sakamoto, K.; Otsuka, Y.; Nakamura, M.; Jellison, J.; Goodell, B. Enhancement of protocatechuate decarboxylase activity for the effective production of muconate from lignin-related aromatic compounds. J. Biotechnol. 2014, 192, 71–77. [CrossRef] 9. Niu, W.; Willett, H.; Mueller, J.; He, X.; Kramer, L.; Ma, B.; Guo, J. Direct biosynthesis of adipic acid from lignin-derived aromatics using engineered Pseudomonas putida KT2440. Metab. Eng. 2020, 59, 151–161. [CrossRef] 10. Overhage, J.; Priefert, H.; Rabenhorst, J.; Steinbüchel, A. Biotransformation of eugenol to vanillin by a mutant of Pseudomonas sp. strain HR199 constructed by disruption of the vanillin dehydrogenase (vdh) gene. Appl. Microbiol. Biotechnol. 1999, 52, 820–828. [CrossRef] 11. Plaggenborg, R.; Overhage, J.; Steinbüchel, A.; Priefert, H. Functional analyses of genes involved in the metabolism of ferulic acid in Pseudomonas putida KT2440. Appl. Microbiol. Biotechnol. 2003, 61, 528–535. [CrossRef][PubMed] 12. Yamada, M.; Okada, Y.; Yoshida, T.; Nagasawa, T. Biotransformation of isoeugenol to vanillin by Pseudomonas putida IE27 cells. Appl. Microbiol. Biotechnol. 2007, 73, 1025–1030. [CrossRef][PubMed] 13. García-Hidalgo, J.; Brink, D.P.; Ravi, K.; Paul, C.J.; Lidén, G.; Gorwa-Grauslund, M.F. Vanillin production in Pseudomonas: Whole-genome sequencing of Pseudomonas sp. strain 9.1 and reannotation of Pseudomonas putida CalA as a vanillin reductase. Appl. Environ. Microbiol. 2020, 86.[CrossRef][PubMed] 14. Overhage, J.; Priefert, H.; Steinbüchel, A. Biochemical and genetic analyses of ferulic acid catabolism in Pseudomonas sp. Strain HR199. Appl. Environ. Microbiol. 1999, 65, 4837–4847. [CrossRef] 15. Hirose, J.; Nagayoshi, A.; Yamanaka, N.; Araki, Y.; Yokoi, H. Isolation and characterization of bacteria capable of metabolizing lignin-derived low molecular weight compounds. Biotechnol. Bioproc. Eng. 2013, 18, 736–741. [CrossRef] 16. Hirose, J.; Tsuda, N.; Miyatake, M.; Yokoi, H.; Shimodaira, J. Draft genome sequence of Pseudomonas sp. strain LLC-1 (NBRC 111237), capable of metabolizing lignin-derived low-molecular-weight compounds. Genome Announc. 2018, 6.[CrossRef] 17. Tatusova, T.; DiCuccio, M.; Badretdin, A.; Chetvernin, V.; Nawrocki, E.P.; Zaslavsky, L.; Lomsadze, A.; Pruitt, K.D.; Borodovsky, M.; Ostell, J. NCBI prokaryotic genome annotation pipeline. Nucleic Acids Res. 2016, 44, 6614–6624. [CrossRef] 18. Aziz, R.K.; Bartels, D.; Best, A.A.; DeJongh, M.; Disz, T.; Edwards, R.A.; Formsma, K.; Gerdes, S.; Glass, E.M.; Kubal, M.; et al. The RAST Server: Rapid Annotations using Subsystems Technology. BMC Genom. 2008, 9, 75. [CrossRef] 19. Johnson, M.; Zaretskaya, I.; Raytselis, Y.; Merezhuk, Y.; McGinnis, S.; Madden, T.L. NCBI BLAST: A better web interface. Nucleic Acids Res. 2008, 36, W5–W9. [CrossRef] 20. Sullivan, M.J.; Petty, N.K.; Beatson, S.A. Easyfig: A genome comparison visualizer. Bioinformatics 2011, 27, 1009–1010. [CrossRef] 21. Kumar, S.; Stecher, G.; Li, M.; Knyaz, C.; Tamura, K. MEGA X: Molecular evolutionary genetics analysis across computing platforms. Mol. Biol. Evol. 2018, 35, 1547–1549. [CrossRef][PubMed] 22. Ohji, S.; Yamazoe, A.; Hosoyama, A.; Tsuchikane, K.; Ezaki, T.; Fujita, N. The complete genome sequence of Pseudomonas putida NBRC 14164T confirms high intraspecies variation. Genome Announc. 2014, 2.[CrossRef] [PubMed] 23. Sambrook, J.; Russell, D.W. Molecular Cloning: A Laboratory Manual, 3rd ed.; CSHL Press: Cold Spring Harbor, NY, USA, 2001. 24. Collier, L.S.; Nichols, N.N.; Neidle, E.L. benK encodes a hydrophobic permease-like protein involved in benzoate degradation by Acinetobacter sp. strain ADP1. J. Bacteriol. 1997, 179, 5943–5946. [CrossRef] [PubMed] Genes 2020, 11, 1416 13 of 13

25. Olivera, E.R.; Miñambres, B.; García, B.; Muñiz, C.; Moreno, M.A.; Ferrández, A.; Díaz, E.; García, J.L.; Luengo, J.M. Molecular characterization of the phenylacetic acid catabolic pathway in Pseudomonas putida U: The phenylacetyl-CoA catabolon. Proc. Natl. Acad. Sci. USA 1998, 95, 6419–6424. [CrossRef] 26. Wei, Z.; Wilkinson, R.C.; Rashid, G.M.M.; Brown, D.; Fülöp, V.; Bugg, T.D.H. Characterization of thiamine diphosphate-dependent 4-hydroxybenzoylformate decarboxylase enzymes from Rhodococcus jostii RHA1 and Pseudomonas ﬂuorescens Pf-5 Involved in degradation of aryl C2 lignin degradation fragments. Biochemistry 2019, 58, 5281–5293. [CrossRef] 27. Tsou, A.Y.; Ransom, S.C.; Gerlt, J.A.; Buechter, D.D.; Babbitt, P.C.; Kenyon, G.L. Mandelate pathway of Pseudomonas putida: Sequence relationships involving mandelate racemase, (S)-mandelate dehydrogenase, and benzoylformate decarboxylase and expression of benzoylformate decarboxylase in Escherichia coli. Biochemistry 1990, 29, 9856–9862. [CrossRef] 28. Tamber, S.; Ochs, M.M.; Hancock, R.E.W. Role of the novel OprD family of porins in nutrient uptake in Pseudomonas aeruginosa. J. Bacteriol. 2006, 188, 45–54. [CrossRef] 29. Lapthorn, A.J.; Zhu, X.; Ellis, E.M. The diversity of microbial aldo/keto reductases from Escherichia coli K12. Chem. Biol. Interact. 2013, 202, 168–177. [CrossRef] 30. Harwood, C.S.; Parales, R.E. The β-ketoadipate pathway and the biology of self-identity. Annu. Rev. Microbiol. 1996, 50, 553–590. [CrossRef] 31. Hasson, M.S.; Muscate, A.; McLeish, M.J.; Polovnikova, L.S.; Gerlt, J.A.; Kenyon, G.L.; Petsko, G.A.; Ringe, D. The crystal structure of benzoylformate decarboxylase at 1.6 Å resolution: diversity of catalytic residues in thiamin diphosphate-dependent rnzymes. Biochemistry 1998, 37, 9918–9930. [CrossRef] 32. Yep, A.; Kenyon, G.L.; McLeish, M.J. Saturation mutagenesis of putative catalytic residues of benzoylformate decarboxylase provides a challenge to the accepted mechanism. Proc. Natl. Acad. Sci. USA 2008, 105, 5733–5738. [CrossRef][PubMed] 33. Bruning, M.; Berheide, M.; Meyer, D.; Golbik, R.; Bartunik, H.; Liese, A.; Tittmann, K. Structural and kinetic studies on native intermediates and an intermediate analogue in benzoylformate decarboxylase reveal a least motion mechanism with an unprecedented short-lived predecarboxylation intermediate. Biochemistry 2009, 48, 3258–3268. [CrossRef][PubMed] 34. Choudhary, A.; Purohit, H.; Phale, P.S. Benzoate transport in Pseudomonas putida CSV86. FEMS Microbiol. Lett. 2017, 364.[CrossRef][PubMed] 35. De Wulf, O.; Thonart, P. Bioconversion of vanillin to vanillyl alcohol in a two-phase reactor. Appl. Biochem. Biotechnol. 1989, 20, 165–180. [CrossRef] 36. Fujihara, H.; Yoshida, H.; Matsunaga, T.; Goto, M.; Furukawa, K. Cross-regulation of biphenyl- and salicylate-catabolic genes by two regulatory systems in Pseudomonas pseudoalcaligenes KF707. J. Bacteriol. 2006, 188, 4690–4697. [CrossRef] 37. Tsipa, A.; Koutinas, M.; Pistikopoulos, E.N.; Mantalaris, A. Transcriptional kinetics of the cross-talk between the ortho-cleavage and TOL pathways of toluene biodegradation in Pseudomonas putida mt-2. J. Biotechnol. 2016, 228, 112–123. [CrossRef][PubMed]

Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional aﬃliations.

© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).