Introduction

Lignin, an aromatic polymer comprising plant cell walls, has attracted attention as a promising aromatic alternative to fossil resources. The effective use of lignin remains unestablished, mainly because of its complex structure. However, if we can produce value-added chemicals such as polymer building blocks from lignin, we will move significantly closer to achieving a decarbonized society1,2,3. In light of this background, there has been increased interest in the production of building blocks for polymers from lignin through a combination of chemical depolymerization of lignin and bacterial catabolic systems4,5,6,7,8.

Sphingobium lignivorans SYK-6 is a gram-negative bacterium that utilizes various lignin-derived aromatic compounds as a sole carbon source, and its catabolic system of lignin-derived aromatic compounds has been best characterized9,10. Research has been conducted on the production of value-added chemicals using the SYK-6 catabolism genes11. An intermediate metabolite of SYK-6, 2-pyrone-4,6-dicarboxylic acid (PDC) is a promising polymer building block for synthesizing biodegradable polymers12,13. In addition to the bacterial catabolic enzyme genes, uptake systems for lignin-derived aromatic compounds have also been investigated, but much remains unclear14,15,16,17,18. Hydrophobic aromatic compounds generally permeate biological membranes through passive diffusion. However, aromatic acids are considered to minimally passively diffuse because of the ionization of their carboxy groups at physiological pH (Table S1; the fractions of carboxylate of protocatechuic acid (PCA), 4-hydroxybenzoic acid (HBA), vanillic acid (VA), syringic acid (SA), and ferulic acid (FA) are 99.87%, 99.86%, 99.86%, 99.95%, and 99.86%)19,20,21. To date, aromatic acid: H+ symporter (AAHS) family transporters, which belong to the major facilitator superfamily (MFS), have been identified as playing a role in the uptake of aromatic acids22,23. For example, PcaK, a transporter of PCA and HBA; VanK, a transporter of VA; and HcnK, a transporter of FA and p-coumaric acid, have been reported in gram-negative bacteria such as Pseudomonas strains21,24,25. PcaK of Acinetobacter sp. ADP1 exhibits transport activity not only for PCA and HBA but also for VA, 3-hydroxybenzoic acid, and 2,4-dihydroxybenzoic acid26. This finding suggests that PcaK has broad substrate specificity for lignin-derived aromatic acids22,26. However, the pathway for the uptake of SA, a key aromatic monomer derived from hardwood and herbaceous lignin, remains unknown. Our group reported that PcaK is essential for the growth of SYK-6 on PCA and the uptake of PCA across the inner membrane17. PcaK of SYK-6 is phylogenetically distant from the other PcaK reported in Pseudomonas and Acinetobacter strains and is more closely related to VanK, which mediates VA uptake in these bacteria17. Heterologous expression analysis demonstrated that PcaK of SYK-6 has the capacity to take up VA in addition to PCA. Moreover, the uptake capacity of PcaK for PCA was significantly reduced in the presence of SA or HBA, implying that PcaK of SYK-6 can take up SA and HBA (Fig. 1)17. However, the exact uptake pathway for these substrates in SYK-6 and the role of PcaK in this process remain unclear.

Fig. 1
figure 1

The uptake and metabolic pathway of lignin-derived aromatic acids in S. lignivorans SYK-6. Enzymes: LigM, VA/3MGA O-demethylase; LigA and B, small and large subunits of PCA 4,5-dioxygenase; LigC, CHMS dehydrogenase; DesA, SA O-demethylase; DesZ, 3MGA 3,4-dioxygenase; PobA, HBA hydroxylase. Compounds: FA, ferulic acid; SA, syringic acid; 3MGA, 3-O-methylgallic acid; GA, gallic acid; VA, vanillic acid; PCA, protocatechuic acid; HBA, p-hydroxybenzoic acid; PDC, 2-pyrone-4,6-dicarboxylic acid; OMA, 4-oxalomesaconic acid.

Full size image

Novosphingobium aromaticivorans DSM 12444 is a gram-negative bacterium that utilizes various lignin-derived aromatic compounds as its sole carbon source27. DSM 12444 catabolizes VA and SA and produces PDC as an intermediate metabolite28,29. Engineered DSM 12444 strains producing cis,cis-muconic acid and carotenoids have been created, making this strain a promising host for producing value-added chemicals from lignin-derived aromatic compounds30,31. Although many degraders of lignin-derived aromatic acids, including SYK-6 and DSM 12444, have been identified in Sphingomonadaceae, only two inner membrane transporters in SYK-6, PcaK and the 5,5’-dehydrodivanillate (DDVA) transporter DdvK, have been identified18. Since the increased expression of pcaK and ddvK in SYK-6 enhanced PDC production from PCA and DDVA, respectively, elucidation of the uptake system in the inner membrane will be helpful for establishing efficient production of value-added products from lignin-derived aromatic acids.

This study aimed to gain insight into the uptake system for lignin-derived aromatic acids in the inner membranes of SYK-6 and DSM 12444. We examined the driving forces required for the uptake of lignin-derived aromatic acids to infer the transporters involved, and we investigated the role of PcaK, which is highly conserved in Sphingomonadaceae. Additionally, we examined how the increased expression of pcaK affects the conversion rate of lignin-derived aromatic acids.

Results and discussion

PcaK is involved in the uptake of lignin-derived aromatic acids in S. lignivorans SYK-6

PcaK of SYK-6 plays a major role in the uptake of PCA and has the capacity to take up VA (Fig. 1)17. To investigate the involvement of PcaK in the uptake of other lignin-derived aromatic acids, we examined the PCA, HBA, VA, SA, and FA conversion capacities of resting SYK-6 and pcaK-disrupted mutant (∆pcaK) cells cultured in Wx-SEMP (10 mM sucrose, 10 mM glutamate, 0.13 mM methionine, and 10 mM proline)32 medium containing 5 mM PCA, HBA, VA, SA, or FA, respectively. ∆pcaK cells had essentially no PCA conversion capacity, and their capacity to convert HBA was reduced to 46% of that of the wild type (WT) after 90 min of incubation (Fig. S1). The conversion rates for VA, SA, and FA decreased to 72%, 46%, and 83% of those of the WT, respectively (Fig. 2A-C, S2).

Fig. 2
figure 2

Characterization of ∆pcaK at various pH. Conversion of VA, SA, and FA by ∆pcaK at various pH (A–C). The cells of SYK-6 and ∆pcaK grown in Wx-SEMP with VA, SA, or FA were incubated in 50 mM Tris-HCl buffer with 100 µM VA, SA, or FA, respectively, at pH 7.5, 8.0, 8.5, or 9.0. The conversion rates at pH 7.5/8.0 and 8.5/9.0 were calculated at 1 and 2 h of reaction, respectively. Conversion rates by SYK-6 of VA at pH 7.5, 8.0, 8.5, and 9.0 (67.9 ± 4.3 µM/h, 69.8 ± 1.3 µM/h, 34.4 ± 2.5 µM/h, 22.3 ± 1.5 µM/h, respectively), SA at pH 7.5, 8.0, 8.5, and 9.0 (68.3 ± 1.4 µM/h, 65.3 ± 7.4 µM/h, 35.5 ± 1.6 µM/h, 24.2 ± 0.5 µM/h), and FA at pH 7.5, 8.0, 8.5, and 9.0 (83.3 ± 4.3 µM/h, 58.8 ± 1.3 µM/h, 28.0 ± 2.5 µM/h, 24.3 ± 1.5 µM/h) were set as 100% rates (control). Growth of ∆pcaK on VA, SA, and FA at pH 7.5 (D–F) and 9.0 (G–I). SYK-6 and ∆pcaK cells were cultured in Wx media supplemented with 5 mM VA, SA, or FA at pH 7.5 and pH 9.0. Cell growth was monitored by measuring the OD600. Each value is the average ± the standard deviation of three independent experiments.

Full size image

The pKa values of the carboxy and phenolic hydroxy groups of PCA, HBA, VA, SA, and FA have been reported to be around pH 4.0 and 9.0, respectively (Table S1). Increasing the pH value enhances the ionization of these functional groups33. As a result, the uptake of hydroxyaromatic acids largely depends on transporters under high pH conditions21. Thus, we evaluated the conversion of VA, SA, and FA by WT and ∆pcaK cells at pH 8.0, 8.5, and 9.0 (Fig. 2A–C, S2). The conversion rates of VA, SA, and FA by ∆pcaK cells at pH 9.0 were reduced to 30%, 14%, and 41% of those of WT cells, respectively. These conversion rates were recovered by introducing a plasmid carrying pcaK, confirming that the reduction in the conversion rate was due to the disruption of pcaK (Fig. S3).

Next, we evaluated the growth of WT and ∆pcaK cells on 5 mM VA, SA, or FA as the sole carbon source at pH 7.5 and 9.0 (Fig. 2D–I). At pH 7.5, the growth capacity of the ∆pcaK cells on VA and FA was comparable to that of the WT cells, but the growth capacity decreased at pH 9.0. WT and ∆pcaK cells did not grow on SA at pH 9.0, but a decrease in growth was observed for ∆pcaK at pH 7.5. These results suggest that PcaK mediates the uptake of PCA, VA, HBA, SA, and FA. Disruption of pcaK had only a minor effect on the growth and conversion of these compounds at pH 7.5, indicating the presence of other transporters for VA, SA, and FA (Fig. 2)21,34. In contrast, PcaK plays a significant role as a transporter for VA, SA, and FA at pH 9.0. This may be attributed to the complete elimination of any residual passive diffusion of these aromatic acids at pH 9.0, or it could be that the other transporters are simply less active at this pH, as was contemplated previous analysis of the transporters of P. putida KT244021. Further verification is needed in the future. In addition, the reason why the loss of growth at pH 9.0 was specific to SA is unclear. The increase in pH may have significantly affected the enzyme activities involved in the conversion of SA metabolites.

Involvement of transporters other than PcaK in S. lignivorans SYK-6

To identify which types of transporters are involved in the uptake of VA, SA, and FA at pH 9.0, we first evaluated the conversion capacities of resting WT and ∆pcaK cells in the presence of cyanide m-chlorophenylhydrazone (CCCP), an inhibitor of proton motive force (PMF) formation (Fig. 3). Examining the conversion capacity of each substrate in the presence of inhibitors of PMF formation and ATP synthesis will provide insight into the driving force required for the uptake of each substrate despite the possibility of side effects. At pH 7.5, the VA, SA, and FA conversion capacities of ∆pcaK cells decreased in the presence of CCCP, suggesting that MFS transporters other than PcaK are also involved in the uptake of VA, SA, and FA at pH 7.5 (Fig. 3A–C). In comparison, the presence of CCCP did not affect the conversion capacities of ∆pcaK cells for these substrates at pH 9.0 (Fig. 3D–F). These results suggest that other MFS transporters, with the exception of PcaK, are not involved in the uptake of VA, SA, and FA at pH 9.0. The involvement of AAHS family transporters, which belong to the MFS transporter and are known to mediate the uptake of aromatic acids, under physiological pH was further investigated. Gene disruption mutants were generated for all 15 AAHS family transporter genes, other than pcaK, and their ability to convert VA and SA was evaluated at pH 7.5 (Fig. 4, S4)17. The results showed that the conversion ability of these gene disruption mutants did not differ from that of the WT. Thus, PcaK plays the most significant role in the uptake of lignin-derived aromatic acids among the AAHS family transporters.

Fig. 3
figure 3

Conversion of VA, SA, and FA by ∆pcaK in the presence of CCCP. SYK-6 and ∆pcaK cells grown in Wx-SEMP with VA, SA, or FA were incubated in 50 mM Tris-HCl buffer with 100 µM VA, SA, or FA in the presence or absence of 100 µM CCCP at pH 7.5 (A–C) or pH 9.0 (D–F). Portions of the reaction mixtures were collected over time, and the amount of substrate was measured using HPLC. Each value is the average ± the standard deviation of three independent experiments.

Full size image
Fig. 4
figure 4

Conversion of VA and SA by 16 AAHS family transporter mutants of SYK-6. The cells of SYK-6, ∆pcaK, and other mutants grown in LB with VA or SA were incubated in 50 mM Tris-HCl buffer (pH 7.5) with 100 µM VA or SA, respectively. The conversion rates of VA and SA were calculated after 1 h of reaction. The conversion rates of VA (90.8 ± 4.1 µM/h) and SA (69.3 ± 2.2 µM/h) by SYK-6 were set as 100% rates (control). Each value is the average ± the standard deviation of three independent experiments.

Full size image

Next, to examine whether ATP-binding cassette (ABC) transporters are involved in the uptake of VA, SA, and FA, we evaluated the conversion of these substrates by resting WT and ∆pcaK cells at pH 7.5 and 9.0 in the presence of N,N’-dicyclohexylcarbodiimide (DCCD), an inhibitor of ATP synthesis (Fig. 5). The addition of DCCD decreased the VA, SA, and FA conversion capacities of WT and ∆pcaK cells at pH 7.5. In particular, FA conversion was lost in WT and ∆pcaK cells and was significantly inhibited compared with that in the presence of CCCP. Based on these results, we concluded that PcaK plays a significant role in the uptake of VA and SA at pH 7.5, but several MFS and ABC transporters are also redundantly involved. In addition, the ABC transporter’s contribution to the uptake of FA was considered significant. On the other hand, at pH 9.0, the conversion capacities of ∆pcaK cells for VA, SA, and FA in the presence of DCCD were extremely low. Thus, the uptake of VA, SA, and FA at pH 9.0 was suggested to be mediated mainly by PcaK and ABC transporters. SYK-6 has 14 genes that encode putative substrate-binding proteins (SBP) of the ABC transporters. However, it was difficult to predict which SBPs would be involved in the uptake because no genes showed high amino acid sequence similarity (< 22%) to SBPs such as CouP and RPA0668 that bind to lignin-derived aromatic compounds found in Rhodopseudomonas palustris35,36,37.

Fig. 5
figure 5

Conversion of VA, SA, and FA by ∆pcaK in the presence of DCCD. SYK-6 and ∆pcaK cells grown in Wx-SEMP with VA, SA, or FA were incubated in 50 mM Tris-HCl buffer with 100 µM VA, SA, or FA in the presence or absence of 100 µM DCCD at pH 7.5 (A–C) or pH 9.0 (D–F). Portions of the reaction mixtures were collected over time, and the amount of substrate was measured using HPLC. Each value is the average ± the standard deviation of three independent experiments.

Full size image

Saro_0804 is involved in the uptake of lignin-derived aromatic acids in N. aromaticivorans DSM 12444

pcaK of SYK-6 (Sl-pcaK) is widely conserved in Sphingomonadaceae17. N. aromaticivorans DSM 12444, which can catabolize a variety of lignin-derived aromatic compounds, also has genes showing 58% and 26% amino acid sequence identity (Saro_0804 and Saro_1588), respectively, with Sl-PcaK. While SYK-6 has 16 AAHS family genes, DSM 12444 has only two genes, Saro_0804 and Saro_1588, and the involvement of these genes in the uptake of lignin-derived aromatic acids was inferred. A phylogenetic tree based on the amino acid sequence similarity of these genes with known AAHS family transporters revealed that Saro_0804 and Saro_1588 are closely related to Sl-PcaK. (Fig. S5, Table S4).

We created disruption mutants of Saro_0804 and Saro_1588 and evaluated their growth and conversion capacities for PCA, HBA, VA, SA, and FA at pH 7.5 (Fig. 6, S6). A Saro_0804 mutant (∆Saro_0804) showed a notable decrease in growth on PCA, but this disruption did not affect growth on other substrates (Fig. 6A–E). These results were consistent with a previous RB-TnSeq analysis of DSM 1244429. In contrast, conversion experiments using resting cells of DSM 12444 (WT) and each disrupted mutant showed that the conversion rates of PCA, HBA, VA, and SA by ∆Saro_0804 cells cultured in LB were reduced to 20%, 26%, 51%, and 36% of those of WT cells (WT activities: PCA, 127 ± 31.8 µM/h; HBA, 112 ± 9.75 µM/h; VA, 66.3 ± 23.2 µM/h; SA, 52.0 ± 13.8 µM/h; FA, 60.2 ± 21.1 µM/h), respectively (Fig. 6F–I). However, the FA conversion rate of the ∆Saro_0804 cells was almost the same as that of the WT cells (Fig. 6J). The conversion rates of the Saro_0804-complemented ∆Saro_0804 cells for PCA, HBA, VA, and SA were restored to the same level as those of the WT (Fig. S7). These results showed that Saro_0804 is essential for optimal growth on PCA and is involved in the uptake of PCA, HBA, VA, and SA; thus, this gene was designated pcaK (Na-pcaK). On the other hand, the disruption of Saro_1588 did not affect growth and conversion.

Fig. 6
figure 6

Characterization of ∆Saro_0804 and ∆Saro_1588. Growth of ∆Saro_0804 and ∆Saro_1588 on PCA, HBA, VA, SA, and FA (A–E). The DSM 12444, ∆Saro_0804, and ∆Saro_1588 cells were cultured in Wx media supplemented with 5 mM PCA, HBA, VA, SA, and FA at pH 7.5. Cell growth was monitored by measuring the OD600. Conversion of PCA, HBA, VA, SA, and FA by ∆Saro_0804 and ∆Saro_1588 (F–J). The DSM 12444, ∆Saro_0804, and ∆Saro_1588 cells grown in LB were incubated in 50 mM Tris-HCl buffer (pH 7.5) with 100 µM PCA, HBA, VA, SA, or FA, respectively. The conversion rates of PCA and HBA were calculated by conversion at 20 and 30 min, respectively. The conversion rates of VA, SA, and FA were calculated after 1 h of reaction. Portions of the reaction mixtures were collected over time, and the amount of substrate was measured using HPLC. Each value is the average ± the standard deviation of three independent experiments.

Full size image

The disruption of Na-pcaK significantly but not completely reduced the ability to convert VA and SA, suggesting that multiple transporters are involved in the uptake of these substrates in DSM 12444 (Fig. 6H, I). In addition, FA conversion by ΔNa-pcaK cells was almost the same as that by WT cells, indicating the presence of transporters other than Na-PcaK for FA uptake. To evaluate the involvement of other MFS transporters in VA, SA, and FA uptake, we measured the VA, SA, and FA conversion capacities of WT and ∆Na-pcaK cells in the presence of CCCP (Fig. 7A–C). When the resting WT and ∆Na-pcaK cells cultured in LB were incubated with 100 µM VA, SA, or FA in the presence or absence of CCCP, both cells lost most of their ability to convert VA, SA, and FA under CCCP-containing conditions. In contrast, inhibition by DCCD was not significant (Fig. 7D–F). These results suggest that the uptake of VA, SA, and FA in DSM 12444 is likely facilitated primarily by MFS transporters, including Na-PcaK.

Fig. 7
figure 7

Conversion of VA, SA, and FA by ∆Na-pcaK in the presence of CCCP and DCCD. The DSM 12444 and ∆Na-pcaK cells grown in LB were incubated in 50 mM Tris-HCl buffer (pH 7.5) with 100 µM VA, SA, or FA in the presence or absence of 100 µM CCCP (A–C) or 100 µM DCCD (D–F). Portions of the reaction mixtures were collected over time, and the amount of substrate was measured using HPLC. Each value is the average ± the standard deviation of three independent experiments.

Full size image

A DGXD motif containing Asp, which is essential for protonation, and two GXXXD[R/K]XGR[R/K] motifs, which are essential for substrate uptake, are conserved in the AAHS family38,39,40. In Na-PcaK, the DGXD motif and the first GXXXD[R/K]XGR[R/K] motif are conserved, as in Sl-PcaK (Fig. S8)17. The fifth Asp in the second GXXXD[R/K]XGR[R/K] motif is not conserved as it is in Sl-PcaK, but the eighth Gly is conserved, unlike in Sl-PcaK. In contrast, the first Gly of both GXXXD[R/K]XGR[R/K] motifs is not present in Saro_1588. Therefore, it is unlikely that Saro_1588 functions as an AAHS family transporter.

Increased expression of pcaK in SYK-6 improved the conversion rate of lignin-derived aromatic acids

It has been reported that the increased expression of transporter genes improves the production efficiency of value-added products. For example, the increased expression of pcaK in SYK-6 resulted in a 1.24-fold increase in PDC production efficiency from PCA17. ddvK-expressing SYK-6 showed a 1.30-fold PDC production efficiency from DDVA18. Here, we investigated whether the increased expression of pcaK in SYK-6 cells would increase the conversion rates of VA, SA, and FA (100 µM) in SYK-6(pJBSl-pcaK) and SYK-6(pJBNa-pcaK) cells. The results revealed that when expression of Sl-pcaK was elevated in SYK-6, the conversion rates for VA, SA, and FA increased 1.31-, 1.33-, and 1.13-fold, respectively, compared with those of the WT [SYK-6(pJB861)] (WT activities: VA, 19.3 ± 1.75 µM/h; SA, 25.7 ± 2.55 µM/h; FA, 66.9 ± 1.52 µM/h) (Fig. 8). Similarly, increased expression of Na-pcaK resulted in increases of 1.17-fold for VA, 1.24-fold for SA, and 1.12-fold for FA. These results indicate that the increased expression of pcaK also improves the conversion efficiency of VA, SA, and FA.

Fig. 8
figure 8

Conversion of VA, SA, and FA by SYK-6 cells introduced with Sl-pcaK or Na-pcaK via plasmid. The cells of SYK-6(pJB861), SYK-6(pJBSl-pcaK), and SYK-6(pJBNa-pcaK) grown in LB containing 1 mM m-toluate were incubated in 50 mM Tris-HCl buffer (pH 7.5) with 100 µM VA (A), SA (B), or FA (C). The conversion rates of VA, SA, and FA were calculated after 3 h, 2 h, and 1 h of reaction, respectively. Portions of the reaction mixtures were collected over time, and the amount of substrate was measured using HPLC. Each value is the average ± the standard deviation of three independent experiments. *, P < 0.05; **, P < 0.01; ***, P < 0.001 (one-way ANOVA with Dunnett’s multiple comparisons post-test).

Full size image

Conclusions

PcaK was shown to play a significant role in the uptake of PCA, HBA, VA, and SA in S. lignivorans SYK-6 and N. aromaticivorans DSM 12444, which are valuable platform bacteria for producing value-added chemicals from lignin. Among the 16 AAHS family transporter genes in SYK-6, only the disruption of Sl-pcaK had a noticeable effect on VA and SA conversion, supporting a central role for Sl-pcaK in the uptake of lignin-derived aromatic acids. Additionally, Sl-PcaK is involved in the uptake of FA. The increased expression of both pcaK genes improved the conversion rates of lignin-derived aromatic acids. These transporter genes could serve as tools to improve the production efficiency of target metabolites.

Methods

Bacterial strains, plasmids, culture conditions, and substrates

The strains and plasmids used in this study are listed in Table S2. S. lignivorans SYK-6 (NBRC 103272/JCM 17495) and its mutants were grown at 30 °C with shaking (160 rpm) in LB or Wx minimal media supplemented with SEMP. The media for the SYK-6 transformants and mutants were supplemented with 50 mg/L kanamycin (Km). N. aromaticivorans DSM 12444 (NBRC16084) and its mutants were grown under the same conditions as SYK-6. E. coli strains were cultured in LB at 37 °C. The media for the E. coli transformants harboring antibiotic resistance markers were supplemented with 25 mg/L Km. Lignin-derived aromatic compounds were purchased from Sigma-Aldrich and Tokyo Chemical Industry Co., Ltd.

Mutant construction

Plasmids for gene disruption were constructed by amplifying approximately 1-kb fragments carrying upstream and downstream regions of each gene by PCR with SYK-6 or DSM 12444 genomic DNA as a template and primer pairs (Table S3). The resulting PCR fragments were inserted into the BamHI site in pAK405 by the NEBuilder HiFi DNA assembly cloning kit (New England Biolabs, Inc.). These plasmids were independently introduced into SYK-6 or DSM 12444 cells by triparental mating, and candidate mutants were isolated using the same method previously described41. The disruption of the genes was confirmed by colony PCR using primer pairs (Table S3). The plasmids for gene complementation (Table S2) were introduced into the mutant cells of SYK-6 and DSM 12444 by electroporation or triparental mating. The plasmids for increased gene expression (Table S2) were introduced into SYK-6 cells by electroporation or triparental mating.

Sequence analysis

DNA sequencing of the PCR amplification products was conducted by Eurofins Genomics. Sequence analysis was performed using the MacVector program v.15.5.2. Sequence similarity searches and multiple alignments were conducted using the BLAST42 and Clustal Omega43respectively. The phylogenetic tree was generated using the MEGA X program based on multiple sequence alignment generated by the Clustal Omega program and the neighbor-joining algorithm44.

Growth measurement

The cells of SYK-6, DSM 12444, and their derivatives were grown in LB for 24 h. The cells were harvested by centrifugation at 14,100 × g for 5 min, washed twice with Wx buffer [12.5 mM KH2PO4, 27.4 mM Na2HPO4, and 7.6 mM (NH4)2SO4 (pH 7.1)], and resuspended in 1 mL of the same buffer. The cells were then inoculated into Wx medium (pH 7.5 or 9.0) containing 5 mM PCA, HBA, VA, SA, and FA to an optical density at 600 nm (OD600) of 0.1. The pH of the medium was adjusted using NaOH. The cells were incubated at 30 °C with shaking (567 rpm), and cell growth was monitored every hour by measuring the OD600 with an EPOCH2 microplate reader (BioTek Instruments, Inc.). For the growth test of the complemented strains, the cells were cultured in Wx medium containing Km and 1.0 mM m-toluate.

Resting cell assay

The cells of SYK-6 and its pcaK mutant were grown in LB for 24 h, harvested by centrifugation at 14,100 × g for 5 min, washed twice with Wx buffer, and resuspended in the same buffer. The cells were then inoculated into Wx-SEMP medium to an OD600 of 0.2 and incubated for 6 h, after which 5 mM PCA, HBA, VA, SA, or FA was added, and the cells were incubated for an additional 6 h. The cells were harvested by centrifugation at 14,100 × g for 5 min, washed twice with 50 mM Tris-HCl buffer (pH 7.5, 8.0, 8.5, or 9.0), and resuspended in the same buffer. The cells of SYK-6 and the pcaK mutant (OD600, 1.0) were then incubated in 50 mM Tris-HCl buffer (pH 7.5, 8.0, 8.5, or 9.0) with 100 µM PCA, HBA, VA, SA, or FA. For the analysis of disruption mutants of AAHS family transporter genes, cells were grown in LB for 24 h, then inoculated in fresh LB containing 10 mM VA or SA to an OD600 of 0.2 and incubated for 24 h. The cells were harvested by centrifugation at 14,100 × g for 5 min, washed twice with 50 mM Tris-HCl buffer (pH 7.5), and resuspended in the same buffer. The cells of SYK-6 and its mutants (OD600, 2.0) were then incubated in 50 mM Tris-HCl buffer (pH 7.5) with 100 µM VA or SA. The cells of DSM 12444 and its mutants were grown in LB for 24 h, harvested by centrifugation at 14,100 × g for 5 min, washed twice with 50 mM Tris-HCl buffer (pH 7.5), and resuspended in the same buffer. The cells (OD600, 1.0 or 3.0) were then inoculated with 100 µM PCA, HBA, VA, SA, or FA in 50 mM Tris-HCl buffer (pH 7.5). The samples were collected periodically, and the reactions were stopped by centrifugation at 18,800 × g for 10 min. The supernatants were diluted 6-fold in water, filtered, and analyzed using high-performance liquid chromatography (HPLC). For the analysis of SYK-6, DSM 12444, and their mutant cells harboring complementary plasmids, the cells were grown in LB containing Km, and the addition of 1.0 mM m-toluate induced gene expression. For the inhibitor assay, 100 µM CCCP or DCCD dissolved in ethanol was added to the assay mixtures. Ethanol was added to the assay mixtures without inhibitor (1% [v/v]) as a control.

HPLC conditions

HPLC analysis was performed using an Acquity UPLC system (Waters Corporation) equipped with a TSKgel ODS-140HTP column (2.1 by 100 mm; Tosoh Corporation). All HPLC analyses were conducted at a flow rate of 0.5 mL/min. The mobile phase was a mixture of water (90%) and acetonitrile (10%) containing 0.1% formic acid. PCA, HBA, VA, SA, and FA were detected at 259.6, 255.4, 260.3, 274.4, and 322.5 nm, respectively.

Statistics

Statistic tests were performed with Graphpad Prism10 (Graphpad software). Statistical significance was assessed with one-way ANOVA with Dunnett’s multiple comparisons post-test. P < 0.05 was considered statistically significant.