Responsiveness of Aromatoleum aromaticum EbN1^T to Lignin-Derived Phenylpropanoids

Lignin is a ubiquitous heterobiopolymer built from a suite of 3-phenylpropanoid subunits. It accounts for more than 30% of the global plant dry material, and lignin-related compounds are increasingly released into the environment from anthropogenic sources, i.e., by wastewater effluents from the paper and pulp industry.

ABSTRACT

The betaproteobacterial degradation specialist Aromatoleum aromaticum EbN1^T utilizes several plant-derived 3-phenylpropanoids coupled to denitrification. In vivo responsiveness of A. aromaticum EbN1^T was studied by exposing nonadapted cells to distinct pulses (spanning 100 µM to 0.1 nM) of 3-phenylpropanoate, cinnamate, 3-(4-hydroxyphenyl)propanoate, or p-coumarate. Time-resolved, targeted transcript analyses via quantitative reverse transcription-PCR of four selected 3-phenylpropanoid genes revealed a response threshold of 30 to 50 nM for p-coumarate and 1 to 10 nM for the other three tested 3-phenylpropanoids. At these concentrations, transmembrane effector equilibration is attained by passive diffusion rather than active uptake via the ABC transporter, presumably serving the studied 3-phenylpropanoids as well as benzoate. Highly substrate-specific enzyme formation (EbA5316 to EbA5321 [EbA5316-21]) for the shared peripheral degradation pathway putatively involves the predicted TetR-type transcriptional repressor PprR. Accordingly, relative transcript abundances of ebA5316-21 are lower in succinate- and benzoate-grown wild-type cells than in an unmarked in-frame ΔpprR mutant. In trans-complementation of pprR into the ΔpprR background restored wild-type-like transcript levels. When adapted to p-coumarate, the three genotypes had relative transcript abundances similar to those of ebA5316-21 despite a significantly longer lag phase of the pprR-complemented mutant (∼100-fold higher pprR transcript level than the wild type). Notably, transcript levels of ebA5316-21 were ∼10- to 100-fold higher in p-coumarate- than succinate- or benzoate-adapted cells across all three genotypes. This indicates the additional involvement of an unknown transcriptional regulator. Furthermore, physiological, transcriptional, and (aromatic) acyl-coenzyme A ester intermediate analyses of the wild type and ΔpprR mutant grown with binary substrate mixtures suggest a mode of catabolite repression of superior order to PprR.

IMPORTANCE Lignin is a ubiquitous heterobiopolymer built from a suite of 3-phenylpropanoid subunits. It accounts for more than 30% of the global plant dry material, and lignin-related compounds are increasingly released into the environment from anthropogenic sources, i.e., by wastewater effluents from the paper and pulp industry. Hence, following biological or industrial decomplexation of lignin, vast amounts of structurally diverse 3-phenylpropanoids enter terrestrial and aquatic habitats, where they serve as substrates for microbial degradation. This raises the question of what signaling systems environmental bacteria employ to detect these nutritionally attractive compounds and to adjust their catabolism accordingly. Moreover, determining in vivo response thresholds of an anaerobic degradation specialist such as A. aromaticum EbN1^T for these aromatic compounds provides insights into the environmental fate of the latter, i.e., when they could escape biodegradation due to too low ambient concentrations.

INTRODUCTION

Lignin accounts for more than 30% of the global plant dry mass (1), providing compressive strength and protection against pathogens (2, 3). The high heterogeneity of this biopolymer arises from the structural diversity of its phenylpropanoid building blocks, which, moreover, are interconnected by various bond types at multiple positions (4). Due to its global abundance and renewability, lignocellulosic biomass is regarded as a promising alternative to fossil fuels in the production of biofuels (3) and industrial bulk chemicals (5). Hence, pronounced research interest is focused on the identification of novel bacterial enzymes for depolymerization of lignin and respective metabolic engineering (3, 4). In addition to its natural abundance, millions of tons of lignin and lignin-related products are presently released into the environment through waste effluents of the paper and pulping industries (6). White-rot basidiomycetes were found to be mainly accountable for natural lignin depolymerization, employing different ligninolytic enzymes (7–9). However, recent studies showed that bacterial consortia also play a role in degradation of heteropolymeric lignin (4, 10–12). According to the so-called rhizosphere effect, phenylpropanoids are exudated by plant roots (13, 14), enriching the complex soluble organic matter in the rhizosphere soil. This may be rationalized by phenylpropanoids belonging to the dominant metabolites of root core cell types (15). Furthermore, members of the intestinal microbiome of animals can convert dietary flavonoids into phenylpropanoids (16), which may be excreted into the environment. Thus, phenylpropanoids enter terrestrial and aquatic habitats from a variety of sources. Apart from their utilization as a carbon source, recent studies have proven environment-derived 3-phenylpropanoids (i.e., p-coumarate) to also play a role in quorum-sensing mechanisms, allowing bacteria to communicate intercellularly and to adjust global gene expression according to environmental changes and cell density (1, 17).

Many different soil and sediment bacteria have been shown to harbor the metabolic capacity to degrade 3-phenylpropanoids under oxic or anoxic conditions (18–21). The betaproteobacterial degradation specialist Aromatoleum aromaticum EbN1^T is a well-studied model organism capable of oxidizing >20 different aromatic growth compounds completely to CO₂ coupled to denitrification (22–24). Differential proteogenomic analyses revealed a complex catabolic network composed of distinct peripheral degradation pathways subject to highly substrate-specific regulation (23, 25–27). The peripheral degradation pathway for 3-phenylpropanoids in A. aromaticum EbN1^T comprises 5 enzymes and one predicted solute binding protein (SBP), with the coding genes (ebA5316 to ebA5321 [ebA5316-21]; pprA to pprF [pprA-F]) arranged in an operon-like structure (Fig. 1) (20, 23). The latter is part of a larger cluster comprising genes for an ABC-type transport system and the central anaerobic benzoyl-coenzyme A (CoA) pathway (20).

FIG 1.

Scheme of the proposed transcriptional regulation of anaerobic 3-phenylpropanoid degradation in the denitrifying bacterium Aromatoleum aromaticum EbN1^T. Gene expression is proposed to be mediated by the predicted one-component transcriptional repressor PprR. PprR is assumed to bind shared degradation intermediates of different 3-phenylpropanoids to its N-terminal TetR domain. Upon effector binding, PprR disengages from the promoter region of the 3-phenylpropanoid catabolic gene cluster, allowing initiation of transcription (ebA5316-21; pprA-F). Compound names are the following: 1, benzoate; 2, 3-phenylpropanoate or 3-(4-hydroxyphenyl)propanoate; 3, cinnamate or p-coumarate; 4, hydrocinnamoyl-CoA or 3-(4-hydroxyphenyl)propanoyl-CoA; 5, cinnamoyl-CoA or p-coumaroyl-CoA; 6, 3-hydroxy-3-phenylpropanoyl-CoA or 3-hydroxy-3-(4-hydroxyphenyl)propanoyl-CoA; 7, benzoylacetyl-CoA or 4-hydroxybenzoylacetyl-CoA; 8, 4-hydroxybenzoyl-CoA; 9, benzoyl-CoA; 10, cyclohexa-1,5-diene-1-carbonyl-CoA. R, H or OH. green−, presence or absence of double bond. OM, outer membrane; PP, periplasm; IM, inner membrane.

Substrate-specific pathway regulation in A. aromaticum EbN1^T is proposed to be mediated on the transcriptional level by a variety of one- or two-component signaling systems (23), with recent studies pointing to remarkably high in vivo sensitivities in the low-nanomolar range (28, 29). Gene expression for anaerobic degradation of p-ethylphenol was shown to be regulated by the σ⁵⁴-dependent one-component transcriptional activator EtpR (30). The two-component system PcrSR was demonstrated to be essential for the expression of genes encoding enzymes of the anaerobic p-cresol catabolism (29). For the substrate-dependent regulation of the peripheral 3-phenylpropanoid degradation pathway, the one-component system EbA5314 (here renamed PprR, affiliated with the TetR family) was proposed to act as a transcriptional repressor (23). Similar modes of regulation have been identified for the degradation of p-cymene in Pseudomonas putida (CymR) (31) and p-coumarate in Rhodopseudomonas palustris (CouR) and Acinetobacter species (21, 32, 33). Based on these studies, one may assume that the catabolic intermediate p-coumaroyl-CoA binds to the C-terminal TetR domain of PprR, which thereupon disengages from the promoter upstream region, allowing DNA-dependent RNA polymerase to transcribe the 3-phenylpropanoid gene cluster (Fig. 1).

The present study investigates the transcriptional regulation of 3-phenylpropanoid utilization by A. aromaticum EbN1^T on several levels. To begin with, the in vivo responsiveness towards four different 3-phenylpropanoids across a concentration range from 100 µM down to 0.1 nM was determined by means of time-resolved, targeted transcript profiling (Fig. 2). Differential proteomics allocated a nearby encoded ABC transporter to 3-phenylpropanoid uptake (Fig. 3). However, theoretical considerations suggest this system does not play a role in threshold concentrations, where transmembrane effector equilibrium is attained by diffusion (see Fig. 8). Furthermore, we used unmarked in-frame deletion to verify the proposed role of PprR as an appendant transcriptional repressor: a ΔpprR mutant showed an increased transcript level of 3-phenylpropanoid genes compared to the wild type when anaerobically grown with succinate or benzoate (see Fig. 5A and B). In accordance with this, in trans-complementation of pprR in the ΔpprR background restored wild-type-like transcript levels. When grown with p-coumarate, equal transcript levels were observed for all three genotypes (see Fig. 5C). Finally, we provide evidence that 3-phenylpropanoid utilization is, besides being controlled by PprR, also subject to catabolite repression, revealing a multilayered transcriptional regulation of the respective gene cluster (see Fig. 6 and 7).

FIG 2.

Time-resolved, quantitative transcript profiles of A. aromaticum EbN1^T in response to different extracellular effector concentrations. Tested 3-phenylpropanoid effectors were 3-phenylpropanoate, 3-(4-hydroxyphenyl)propanoate, cinnamate, and p-coumarate. The selected transcripts represent genes (Fig. 1) coding for enzymes involved in the anaerobic degradation of 3-phenylpropanoids (ebA5316, ebA5317, ebA5319, and ebA5321). Relative transcript abundances were determined by means of qRT-PCR, with the time point of 5 min prior to effector addition serving as a reference. Each data point is based on 3 biological replicates with 3 technical replicates analyzed for each. Growth data of the cultures providing the RNA samples for all tested conditions are shown in detail in Fig. S1 to S4.

FIG 3.

Relationship between 3-phenylpropanoid and benzoate utilization by A. aromaticum EbN1^T in terms of uptake, catabolism, and transcriptional regulation. (A) Structural and functional representation of the chromosomal locus (27) comprising the genes related to anaerobic catabolism of 3-phenylpropanoids and benzoate (20). Note the proposed shared role of the predicted ABC-type transporter for the uptake of these aromatic carboxylates. (B) Differential profiles of the encoded proteins are in accord with the proposed polyspecificity of this ABC uptake system. Proteomic profiling was based on membrane protein-enriched as well as soluble protein fractions. In case of protein identification in both fractions, the one yielding the highest MASCOT score is shown here; further details are provided in Table S7.

FIG 8.

Correlation of response threshold with K_d values of known SBPs. (A) Response threshold (red) for 3-phenylpropanoids determined in this study. K_d values (gray) theoretically inferred for the predicted SBP EbA5316 from the in vivo response thresholds. Abbreviations: 3PP, 3-phenylpropanoate; 3HP, 3-(4-hydroxyphenyl)propanoate; Cin, cinnamate; p-Cou, p-coumarate. (B) K_d values of known SBPs as retrieved from published experimental studies (K_d values, further details, and references are compiled in Table S10). Note that the category “others” comprises SBPs for vitamins, cofactors, trace elements, and signaling molecules. Black lines, medians.

FIG 5.

Relative transcript abundances of 3-phenylpropanoid genes in A. aromaticum EbN1^T grown with succinate (A), benzoate (B), or p-coumarate (C). Cells were harvested at half-maximum optical density, and transcripts of target genes were normalized against gyrB. Each data point is based on 3 biological replicates with 3 technical replicates analyzed for each. Colors: blue, wild type; pink, ΔpprR mutant; gray-blue, pprR-complemented mutant.

FIG 6.

Anaerobic cultivation of A. aromaticum EbN1^T (wild type) for targeted transcript analysis in response to a binary mixture of 1 mM p-coumarate and either 1 mM benzoate (A) or 3 mM succinate (B). Growth was monitored by measuring the optical density at 660 nm (OD₆₆₀). Sampling time points for transcript analysis are indicated by gray dashed lines. Fold changes of each gene at each time point were normalized against gyrB. Each data point is based on 3 biological replicates with 3 technical replicates analyzed for each.

FIG 7.

Quantification of (aromatic) acyl-CoA metabolites in cells of A. aromaticum EbN1^T (wild type and ΔpprR mutant) adapted to anaerobic growth with p-coumarate, benzoate, or succinate. Shades of red represent the number of molecules per cell. For each substrate condition and genotype at least three biological replicates were analyzed. Compound names: 5 to 10, see the legend to Fig. 1; 11, 6-hydroxycyclohex-1-ene-1-carbonyl-CoA; 12, 6-oxocyclohex-1-ene-1-carbonyl-CoA; 13, 3-hydroxypimelyl-CoA.

RESULTS

In vivo response threshold toward 3-phenylpropanoids.

To assess the response threshold for structural variants of 3-phenylpropanoids, first benzoate-adapted cells of A. aromaticum EbN1^T were anaerobically grown with a limiting supply of benzoate (1 mM). Upon its complete consumption, a single pulse of one 3-phenylpropanoid was immediately administered per individual culture, yielding initial concentrations from 100 µM down to 0.1 nM. Tested 3-phenylpropanoids were 3-phenylpropanoate, 3-(4-hydroxyphenyl)propanoate, cinnamate, and p-coumarate. Highly reproducible growth curves with sampling time points for each of the eight tested effector concentrations and a negative control (no addition of effector) are provided in Fig. S1 to S4 in the supplemental material and fold changes of transcript abundances in Tables S1 to S4. For transcript profiling, samples were retrieved 5 min prior to as well as 5, 15, 30, 60, and 120 min after the pulse (zoom-ins in Fig. S1 to S4). Differential expression of the four selected genes was related to the individual expression levels directly prior to the effector pulse. Analyzed genes were selected according to their location within the 3-phenylpropanoid gene cluster, covering its start (ebA5316), middle (ebA5317 and ebA5319), and end (ebA5321).

For all tested 3-phenylpropanoids, only the highest concentration of 100 µM was sufficient to support an increase in optical density. At all other tested concentrations, the optical density of the cultures remained constant after addition of the respective effector. Time-resolved transcript profiles revealed throughout a correlation between transcriptional response time and the gene position within the operon-like structure (Fig. 2). Accordingly, genes near the end of the gene cluster had delayed and lower transcript abundances than genes located closer to the putative transcriptional start. Moreover, an apparent correlation between effector concentration and overall response intensity was observed.

Upon addition of 3-phenylpropanoate, 3-(4-hydroxyphenyl)propanoate, or cinnamate at final concentrations of ≥10 nM, an increase in relative ebA5316 transcript abundance was already observed after 5 min (≥30 nM for p-coumarate). Altogether, the maximal transcript abundance increase of ∼300-fold was observed for ebA5317 in response to 100 µM 3-phenylpropanoate. For the three other 3-phenylpropanoids, the maxima were likewise recorded at 100 µM effector: 100-fold for cinnamate and 135-fold for p-coumarate and 3-(4-hydroxyphenyl)propanoate. The lowest effector concentration allowing for a transcriptional response was 10 nM in the case of 3-phenylpropanoate, 3-(4-hydroxyphenyl)propanoate, and cinnamate. Therefore, an apparent in vivo response threshold between 1 nM and 10 nM is obvious for these three effectors. In the case of p-coumarate, the lowest concentration yielding a transcriptional response was 50 nM, suggesting a higher in vivo response threshold in the range of 30 to 50 nM.

3-Phenylpropanoid uptake presumably via ABC transporter.

The first gene in the phenylpropanoid gene cluster of A. aromaticum EbN1^T, ebA5316, encodes a solute-binding protein (SBP) of ABC-type transporters that shows highest similarity to SBPs from various betaproteobacteria (Table S5). Furthermore, comparison to experimentally verified aromatic compound-binding SBPs from R. palustris revealed higher similarity of EbA5316 to p-coumarate-binding RPA1789 (identity 49%) than to benzoate- or p-hydroxybenzoate-binding SBPs (RPA0668, 24% identity; RPA4029, 25% identity) (34) (Table S6). This suggests EbA5316 acts as a 3-phenylpropanoid-specific SBP serving an ABC-type transport system. Notably, such a system is encoded in direct proximity of the 3-phenylpropanoid gene cluster, comprising permease (EbA5304/6) and ATP-binding subunits (EbA5307/9) (Fig. 3A). Since an SBP (EbA5303) is encoded directly upstream of ebA5304-9 on the same strand, this system was previously assigned to benzoate uptake (20). In accordance with this, EbA5303 is markedly more similar to benzoate-binding RPA0668 (identity, 54%) than to p-coumarate-binding RPA1789 (identity, 26%) of R. palustris. Finally, this ABC transporter (EbA5304-9), together with the two SBPs (EbA5316 and EbA5303 [EbA5316/03]), is encoded between the gene clusters for anaerobic catabolism of 3-phenylpropanoids and benzoate (Fig. 3A).

To scrutinize the implicated polyspecificity of this ABC transporter, the membrane protein-enriched fractions of 3-(4-hydroxyphenyl)propanoate-, benzoate-, and acetate-adapted cells of A. aromaticum EbN1^T were analyzed comparatively. The predicted ABC transporter permease (EbA5304/6) as well as ATP-binding subunits (EbA5307/9) were similarly abundant in the case of the two aromatic substrates but not with acetate (Fig. 3B, Table S7). While the benzoate-specific SBP (EbA5303) showed similar abundance with both aromatic substrates, the 3-phenylpropanoid-specific SBP (EbA5316) was abundantly formed only in 3-(4-hydroxyphenyl)propanoate-adapted cells. Thus, the predicted permease (EbA5304/6) apparently accommodates both types of aromatic carboxylates, while the two SBPs should confer substrate specificity (EbA5316 for 3-phenylpropanoids and EbA5303 for benzoate).

Generation of ΔpprR and pprR-complemented mutants.

The predicted one-component transcriptional repressor PprR is encoded directly upstream of the 3-phenylpropanoid gene cluster (Fig. 3A) on the complementary strand. To validate its function in repressing expression of ebA5316-21, an unmarked, in-frame ΔpprR mutant (ΔpprR) was generated. In this mutant, the start and stop codons of pprR as well as a BamHI restriction site (6 nucleotides) were preserved to maintain the reading frame (Fig. S5A). Accordingly, gene-specific primers for pprR did not yield a PCR product in the deletion mutant, whereas primers hybridizing up- and downstream of the knockout region resulted in a small 316-bp region compared to a large 1,207-bp product in the wild type (Fig. S5B). Deletion of pprR led to derepression of the 3-phenylpropanoid gene expression in cells anaerobically grown with benzoate (see below). The ΔpprR mutant was in trans-complemented (ΔpprR-compl.) using a pprR-containing pBBR1MCS-2 broad-host-range vector (Fig. S5B), and this complementation restored repression of the 3-phenylpropanoid-specific gene expression in benzoate-grown cells (see below). Moreover, the correctness of both mutants was verified by nucleotide sequencing.

PprR involved in substrate-specific expression of 3-phenylpropanoid genes.

To investigate the ΔpprR-specific phenotype, all three genotypes (wild type, ΔpprR, and ΔpprR-compl.) were anaerobically grown with succinate, benzoate, or p-coumarate (Fig. 4A to C). Furthermore, for all three genotypes and substrate conditions (at half-maximal optical density, or ½OD_max), abundance profiles of transcripts (via quantitative reverse transcription-PCR [qRT-PCR], normalizing against gyrB) (Fig. 5) and proteins (via shotgun proteomics; Fig. S6) of the 3-phenylpropanoid gene cluster were determined.

FIG 4.

Characterization of the wild type (blue), ΔpprR mutant (pink), and pprR-complemented mutant (gray-blue) of A. aromaticum EbN1^T. All three genotypes were grown with 8 mM succinate (A), 4 mM benzoate (B), or 2 mM p-coumarate (C). (D) Increased transcript level of pprR in the pprR-complemented mutant (gray-blue) compared to the wild type (blue) of A. aromaticum EbN1^T; transcript levels are normalized against gyrB.

(i) Succinate and benzoate.

All three genotypes showed highly similar growth behavior, maximum growth rates (µ_max), and substrate consumption profiles (Fig. 4AB). In accordance with the prediction that PprR acts as a transcriptional repressor of the 3-phenylpropanoid gene cluster, transcript abundance of the respective genes was markedly increased (derepressed) in the ΔpprR mutant compared to the wild type (Fig. 5AB). The strongest relative increase in transcript abundance (10-fold) was observed for ebA5316, which is localized closest to the putative transcriptional start. Analogous increases in relative transcript abundance (2.3- to 4.1-fold) were observed for ebA5317, ebA5318, and ebA5319. In the case of ebA5320 and ebA5321, located at the end of the 3-phenylpropanoid gene cluster, such an increase was only observed for succinate-grown cells. The described effects on gene expression in the ΔpprR mutant could be canceled by in trans-complementation of the pprR gene, restoring relative transcription to levels similar to those of the wild type, except for ebA5316 (2.8-fold increase with benzoate).

(ii) p-Coumarate.

µ_max as well as OD_max were similar for all three genotypes (Fig. 4C). However, the pprR-complemented mutant had a significantly longer lag phase than the wild type and ΔpprR mutant. Correspondingly, complete substrate consumption by the pprR-complemented mutant was achieved approximately 30 h later. This could be due to the markedly higher transcript level of pprR (∼100-fold; Fig. 4D) in the pprR-complemented mutant, since it carries the pprR gene on a high expression vector. This further supports PprR to function as a transcriptional repressor, with higher pprR expression levels resulting in prolonged repression of 3-phenylpropanoid gene cluster expression.

In contrast to succinate- and benzoate-grown cells, transcript profiles of p-coumarate-grown cells revealed no differences between the three genotypes. Interestingly, transcript levels in all three genotypes were at least 1 order of magnitude higher with p-coumarate than those (derepressed) of the ΔpprR mutant with either succinate or benzoate (Fig. 5). Accordingly, the respective proteins were exclusively detected (Fig. 3B and Fig. S6B and C) in p-coumarate cells of all three genotypes.

Benzoate-induced diauxie.

The above-described transcript profiles of the three genotypes (Fig. 5) indicated the presence of additional regulatory mechanisms controlling transcription of the 3-phenylpropanoid gene cluster (ebA5316-21). To investigate this assumption, cells of A. aromaticum EbN1^T (wild type and ΔpprR mutant) were supplied with a binary mixture of p-coumarate (1 mM) and either benzoate (1 mM) or succinate (3 mM). With the mixture of p-coumarate and benzoate, both genotypes clearly showed diauxic growth accompanied by preferential utilization of benzoate (Fig. 6A and Fig. S7A). For both genotypes, maximum relative ebA5316-21 transcript abundances were observed shortly after benzoate was completely consumed (Fig. 6A and Fig. S7A).

Interestingly, the wild type and ΔpprR mutant showed monophasic growth and concomitant utilization of substrates when supplied with a mixture of p-coumarate and succinate (Fig. 6B and Fig. S7B). However, both genotypes consumed succinate at a higher rate than p-coumarate. The wild type reached maximum relative ebA5316-21 transcript abundances only after complete consumption of succinate (Fig. 6B). In contrast, this was achieved by the ΔpprR mutant already when only half of the initially supplied succinate was consumed (Fig. S7B).

Benzoyl-CoA was previously reported as the effector of transcriptional repressor BzdR in Azoarcus sp. strain CIB (which needs reclassification as Aromatoleum species) (24). Therefore, the formation of CoA-ester intermediates involved in anaerobic degradation of p-coumarate and benzoate was analyzed in cells of A. aromaticum EbN1^T (wild type and ΔpprR mutant). Under these substrate conditions, both genotypes contained equal amounts of benzoyl-CoA, ∼700 × 10³ molecules cell⁻¹, corresponding to ∼1.5 µmol g_CDW⁻¹, respectively (CDW, cell dry weight) (Fig. 7 and Tables S8 and S9). Notably, the amount of benzoyl-CoA in these cells is ∼50-fold higher than that of all other identified (aromatic) acyl-CoA esters (Fig. 7). In contrast, succinate-grown cells of both genotypes harbored significantly smaller amounts of benzoyl-CoA (∼30 × 10³ molecules cell⁻¹ or 0.05 µmol g_CDW⁻¹). As expected, the p-coumarate-specific CoA-ester intermediates could only be detected in cells of the wild type and ΔpprR mutant upon growth with this phenylpropanoid (Fig. 7).

DISCUSSION

In vivo response threshold toward 3-phenylpropanoids.

Recently, Vagts et al. (29) revealed highly sensitive responsiveness of A. aromaticum EbN1^T toward alkylphenols. There, it was concluded that such low (lower nanomolar range) response thresholds could have implications for microbial viability at low substrate availability as well as for the mechanisms underlying the persistence of dissolved organic matter (DOM) in a variety of environmental settings. The diverse chemical structures and properties of the tested 3-phenylpropanoids prompted a 2-fold reflection on their determined in vivo response thresholds, as discussed below.

(i) Factors controlling attainment of transmembrane effector equilibrium at response threshold concentrations.

In contrast to uncharged alkylphenols, 3-phenylpropanoids represent charged molecules (dissociated carboxyl group) at ambient pH. Hence, besides passive diffusion across the cell envelope, there is another means of uptake, in which an ABC transporter in conjunction with the predicted SBP EbA5316 (see above) is assumed to be involved (Fig. 1). In the following paragraphs, we evaluate to what degree active uptake versus passive diffusion of membrane-permeable solutes contributes to attaining transmembrane equilibrium of 3-phenylpropanoids at the response threshold concentrations determined in this study.

As a first step, we estimated the abundance of EbA5316 in A. aromaticum EbN1^T actively growing with ample provision of p-coumarate. For this purpose, we used the proteomic data of the mutant characterization, generated at ½OD_max, when approximately 1 mM p-coumarate was still available (Fig. 4C). We evaluated these data by referring our measured peptide counts of selected ribosomal proteins against reported copy numbers per cell of the respective homologues in E. coli growing in minimal medium with 22 mM glucose and harvested in early logarithmic phase (35) (see Fig. S8 in the supplemental material). We obtained an estimate of 5 × 10³ to 10 × 10³ copies of EbA5316 per actively growing cell of A. aromaticum EbN1^T. Given the limit of accuracy of our approach (50% variability in multiplier $\overline{M}$; see Fig. S8), our estimate is on the same order of magnitude as the copy numbers (22 × 10³ and 45 × 10³) determined for the maltose-binding protein (MalE) in logarithmically growing E. coli supplemented with 5.8 mM (36) and 11.6 mM (37) maltose, respectively. In A. aromaticum EbN1^T, 10,000 copies of EbA5316 in the periplasm (estimated volume, 10⁻¹⁶ liters [29]) correspond to a concentration of ∼0.14 mM.

From dialysis experiments (38, 39) and theoretical reasoning (40), it was recognized early on that SBP copy numbers per cell in the range of 10³ to 10⁴ afford two major functions for the cell. To begin with, they allow for an efficient collection of ligands from the solute flux impinging on the cell. Accordingly, 3 × 10³ SBPs already well dispersed in the periplasmic space, only covering 0.1% of the cell surface, would suffice to collect 50% of the flux, thereby enhancing the effective absorption cross section 1,000-fold (40). Furthermore, SBPs at high periplasmic concentrations relative to their dissociation constant (K_d) values act as a retention buffer to maintain the periplasmic solute concentration for a long time (38, 41). Lastly, a highly abundant SBP promotes the transfer of the solute from its SBP-bound state to the permease-mediated uptake and avoids unproductive dissociation of the solute back into the periplasm or extracellular space.

As a second step, we estimated the basal level of EbA5316 in benzoate-adapted cells of A. aromaticum EbN1^T, viz. under noninduced conditions. For this purpose, we are considering two observations: (i) fully induced cells of A. aromaticum EbN1^T have an ∼100-fold increased ebA5316 transcript level (Table S4), and (ii) according to the study by Lu et al. (35), the ratio of synthesized protein per mRNA molecule typically is at least 100 in E. coli. Taken together, we estimate the EbA5316 abundance in noninduced cells of A. aromaticum EbN1^T to be ∼10⁴-fold lower than that of p-coumarate-adapted cells. Considering that we estimated the latter to carry about 5 × 10³ to 10 × 10³ copies of EbA5316, this SBP might be present in single-figure copy numbers, if at all, in noninduced cells. Such an estimate is in accord with the lack of proteomic detection of EbA5316 in succinate- or acetate-adapted cells (Fig. 3B and Fig. S6). With benzoate-adapted cells, we have detected EbA5316, but not consistently (absence, Fig. S6; presence, Fig. 3B). This may be due to a certain degree of unspecificity of the involved regulators for aromatic acyl-CoA effectors.

As a third step, we assessed whether an SBP present at low copy numbers in noninduced cells could contribute to attaining effector equilibrium at the low-nanomolar threshold concentrations. In this context, the affinity of a given SBP to its effector molecule is decisive. Therefore, we compiled, from the literature, 89 experimentally determined K_d values of purified SBPs associated with ABC and TRAP transporters specific for different classes of solutes (Fig. 8 and Table S10). For growth-supporting organic substrates, K_d values range from 8 nM to 716 µM with a median value between 0.5 and 1.0 µM. At threshold substrate concentrations, [S], of 10 nM, as determined in the present study (Fig. 2), the ratio of unbound SBP ([P]) to bound SBP ([P·S]) ranges between 50:1 and 100:1 according to the following equation:

$$\frac{K_d}{\left[\text{S}\right]}=\frac{\left[\text{P}\right]}{\left[\text{P}\cdot \text{S}\right]}$$ (1)

At such ratios, an SBP-mediated uptake is not efficient. Even when considering the lowest K_d value reported for an aromatic carboxylate (95 nM for p-hydroxybenzoate [34]), the ratio of free to bound SBP is still 10:1. For EbA5316 to play a role in substrate uptake at response threshold concentrations, its K_d value would have to be distinctly lower than the experimentally determined threshold concentrations (Fig. 8A), i.e., in the subnanomolar range, which is currently known only for compounds such as vitamins and trace elements (Fig. 8B). Therefore, we conclude that passive diffusion of 3-phenylpropanoids across the cell envelope apparently drives the attainment of the transmembrane effector equilibrium, as recently described for alkylphenols (29).

(ii) Rationalizing higher threshold observed for p-coumarate.

Among the four tested 3-phenylpropanoids, 3-phenylpropanoate yields the highest overall transcriptional response (∼300-fold for ebA5317). Additional structural features [ring hydroxylation in 3-(4-hydroxyphenyl)propanoate versus alkenoyl chain in cinnamate] lead to a lower overall transcriptional response intensity (∼100-fold for ebA5316), albeit to a response threshold (1 to 10 nM) similar to that for 3-phenylpropanoate (Fig. 2). In contrast, p-coumarate, characterized by a para-hydroxyl group as well as an alkenoyl chain, delivers a lower overall transcriptional response as well as a higher response threshold (30 to 50 nM).

Even though one may speculate that EbA5316 has a lower affinity to p-coumarate, this should be irrelevant for its threshold concentration, since SBP-mediated active uptake does not play a role in attainment of the effector equilibrium, as outlined above. Alternatively, the higher threshold for p-coumarate may reflect a lower turnover by CoA-ligase EbA5317, yielding a lower pool size of p-coumaroyl-CoA than the CoA-esters formed from the other three structurally different 3-phenylpropanoids. This may be plausible, since EbA5317 represents the first common enzyme during degradation of all four tested 3-phenylpropanoids, and the respective CoA-esters most likely act as the true effectors for the repressor PprR. The latter would be in accord with a previous report on the analogous transcriptional repressor CouR (R. palustris), which is not responsive to p-coumarate but instead binds p-coumaroyl-CoA as its effector (32).

Multilayered transcriptional regulation of 3-phenylpropanoid genes.

Characterization of the ΔpprR mutant underpinned the predicted function of PprR as a transcriptional repressor of ebA5316-21 expression in a 3-phenylpropanoid-dependent manner. Unexpectedly, however, even in the ΔpprR mutant, maximum relative transcript abundances of ebA5316-21 were only observed with p-coumarate rather than with benzoate or succinate (Fig. 5). This suggests the involvement of additional transcriptional control circuits, namely, a 3-phenylpropanoid-specific activation and/or an additional benzoate/succinate-mediated repression. A specific transcriptional activator for ebA5316-21 expression could not be inferred from analyzing the genome of A. aromaticum EbN1^T. Furthermore, while some TetR-family members are known to act as repressor as well as activator (42), such a dual function does not appear likely for PprR of A. aromaticum EbN1^T, according to our present data and the close relatedness to the proven TetR-type repressors KstR and KstR2, controlling “cholesterol degradation” genes in mycobacteria (43). One can speculate about catabolite repression, since anaerobic utilization of C₄-dicarboxylates by A. aromaticum EbN1^T is repressed in the presence of benzoate (44). In accordance with this, catabolite repression was previously reported to play a role in the catabolism of aromatic compounds in Thauera aromatica K172^T (45), Azoarcus sp. strain CIB (46), Acinetobacter baylyi (47), and Escherichia coli (48, 49).

(i) Benzoate-induced diauxie.

While the two applied binary substrate mixtures (p-coumarate and benzoate versus p-coumarate and succinate) yielded converse growth courses (diauxic versus monophasic) with the wild type under both substrate conditions, maximal ebA5316-21 transcript abundances were observed upon complete consumption of benzoate and succinate (Fig. 6). In the case of the ΔpprR mutant, the growth courses were more pronounced and maximal transcript abundances were reached when benzoate was completely and succinate half-completely consumed (Fig. S7). These findings suggest the presence of a regulatory mechanism hindering full transcription of 3-phenylpropanoid genes in the presence of both benzoate and succinate. Furthermore, this mechanism seems to be superior to PprR, as its effect was observed not only with the wild type but also with the ΔpprR mutant. However, benzoate appears to more strongly repress ebA5316-21 transcription than succinate, as it not only delayed full ebA5316-21 transcription but also completely repressed p-coumarate utilization until benzoate depletion.

(ii) Role of benzoyl-CoA in diauxie?

Anaerobic degradation of most monoaromatic compounds feeds into the central benzoyl-CoA pathway. Benzoyl-CoA represents not only the entry point into this central degradation pathway but also the effector for BzdR (K_d of ∼150 µM), a transcriptional repressor controlling gene expression of the anaerobic benzoyl-CoA pathway in Azoarcus sp. strain CIB (50, 51). Thus, intracellular benzoyl-CoA levels might function as a control switch to impose preferential utilization of benzoate in the presence of other aromatic or aliphatic growth substrates in A. aromaticum EbN1^T (44) and Magnetospirillum sp. strain pMbN1 (52). However, the observed similar intracellular pool sizes of benzoyl-CoA in wild-type and ΔpprR mutant cells of A. aromaticum EbN1^T grown with either benzoate or p-coumarate (Fig. 7) argues against this aromatic acyl-CoA conveying benzoate-induced diauxie. Furthermore, these congruent benzoyl-CoA levels suggest a previously reported preference of benzoate over succinate in A. aromaticum EbN1^T (44) likewise to not be mediated by benzoyl-CoA, since p-coumarate and succinate were concomitantly utilized (Fig. 6 and Fig. S7). The ∼25-fold lower but nevertheless detectable level of intracellular benzoyl-CoA in succinate-grown cells most likely reflects its additional role in anabolic processes.

In conclusion, our findings support the assumption of a higher-order transcriptional regulator being responsible for tuning the studied degradation pathways of A. aromaticum EbN1^T in a benzoate- and succinate-dependent manner comparable to the ones described for E. coli (53) and Pseudomonas (54, 55).

MATERIALS AND METHODS

Bacterial strains and cultivation conditions.

A. aromaticum EbN1^T was cultivated under nitrate-reducing conditions in a defined, ascorbate-reduced, and bicarbonate-buffered mineral medium at 28°C as previously described (22). Organic substrates [acetate, succinate, benzoate, p-coumarate, 3-phenylpropanoate, 3-(4-hydroxyphenyl)propanoate, and cinnamate] were added from aqueous stock solutions sterilized by filtration. Each experimental line (response threshold determination, mutant characterization, and differential proteomics) was preceded by the same cultivation steps with benzoate as the substrate for anaerobic growth as described before (28, 29). Essentially, glycerol stocks of A. aromaticum EbN1^T were used to inoculate a dilution series (10⁻¹ to 10⁻⁶) as the starting point for two successive precultures (80-ml culture volume), followed by a main culture (400-ml culture volume), which served as the inoculum for each actual experiment. For each experimental line (see below), cultivation was performed in 500-ml flat-bottomed glass bottles (400-ml culture volume) sealed with butyl rubber stoppers. All chemicals were of analytical grade. Growth was monitored by measuring the optical density of the cultures at 660 nm (OD₆₆₀).

Generation of in-frame ΔpprR deletion mutation.

DNA was isolated according to standard protocols (56). Generation of the in-frame ΔpprR deletion mutation was carried out as previously described for the etpR and pcrSR genes of A. aromaticum EbN1^T (29, 30) and using the primers listed in Table 1. The ΔpprR mutant retained the start and stop codons of pprR flanking a BamHI restriction site. The final knockout vector, pk19 ΩbenKebA5316, for in-frame deletion of pprR (ebA5314) was constructed in an E. coli S17-1 background by following a two-step strategy. First, a 1,250-bp fragment (benK) representing the 3′-flanking region of pprR including the stop codon of benK, and a BamHI restriction site was amplified from genomic DNA of A. aromaticum EbN1^T using a Phusion high-fidelity DNA polymerase (ThermoFischer Scientific, Waltham, MA, USA) and cloned into the vector using its PstI and BamHI restriction sites, giving rise to pk19 ΩbenK. Second, a 1,500-bp fragment (ebA5316) was amplified as described above and cloned into pk19 ΩbenK using the In-Fusion HD cloning plus kit (TaKaRa Bio Inc., Kusatsu, Japan) according to the manufacturer's instructions. This fragment represented the 5′-flanking region of pprR, including the start codon of ebA5316 and a BamHI restriction site, yielding pk19 ΩbenKebA5316. Conjugational plasmid transfer to A. aromaticum EbN1^T was conducted as described previously (56). PCR-based screening of single-crossover clones (kan^R) involved the ΔpprR primer pair, yielding 316-bp and 1,207-bp amplicons, respectively (see Fig. S5 in the supplemental material). The second crossover event was achieved by transferring the single-crossover mutant into kanamycin-free liquid medium containing a mixture of benzoate (4 mM), acetate (5 mM), and pyruvate (5 mM). Plating on solid medium containing the same tripartite substrate mixture and 5% (wt/vol) sucrose was used for selection of the ΔpprR genotype. For identification, colonies were screened as described above, yielding only one amplicon of 316 bp (Fig. S5).

TABLE 1.

Oligonucleotide primers used for mutant construction

Primera	Target gene	Nucleotide sequence (5′→3′)	Product length (bp)
Gene-specific primer pair
pprR_240_F	pprR	GCTGAGTGACAGCCAACG	235
pprR_474_R	CGACGCGAAGTGGTGATA
Generation of ΔpprR deletion mutation
benK_PstI_F	benK	GCCGCTGCAGGCTGGCA	1,250
benK_BamHI_R	benK	AAGGATCCTAGCGGGCGCACCCCG
ebA5316_BamHI_F	ebA5316	GGTGCGCCCGCTAGGATCCCATCCGGAATCCTGCCGC	1,500
ebA5316_BamHI_R	ebA5316	GCTCGGTACCCGGGGATCCGTGCCTCGATGTTCACGACC
Identification of ΔpprR genotype
ΔpprR_F	pprR	GCTGAGTGACAGCCAACG	292
ΔpprR_R	CGACGCGAAGTGGTGATA
Generation of in trans-complementation of pprR
pprR_EcoRI_F	pprR	CGATCGCGGAATTCGCGGTGTC	1,434
pprR_ApaI_R	ACGCACCGGGCCCGG

^aF, forward primer; R, reverse primer.

Complementation of pprR in trans into ΔpprR deletion mutant.

In trans-complementation was performed as described in two previous studies (29, 30) and using the primers listed in Table 1. Essentially, a 1,434-bp fragment containing the pprR gene together with its ribosomal binding site was amplified from genomic DNA of A. aromaticum EbN1^T as described above. The fragment was cloned into the broad-host-range vector pBBR1MCS-2 using its ApaI and EcoRI restriction sites. The final construct was verified by sequencing. The vector was transferred from the E. coli S17-1 background to the ΔpprR mutant of A. aromaticum EbN1^T via conjugation, yielding the pprR-complemented mutant (genotype of ΔpprR pBBR1MCS-2 ΩpprR). The conjugational transfer via agar plate mating and the verification of positive clones was performed as previously described (56).

Sequence validation by Sanger sequencing.

For sequence validation of the obtained ΔpprR mutant, a 2,393-bp region of genomic DNA spanning across the entire up- and downstream regions, including the deletion site, was analyzed via Sanger sequencing (57), as previously described (29). Additionally, a 1,892-bp region of the complementation vector (pBBR1MCS-2ΩpprR) spanning the entire insert was sequenced. Used primer pairs are compiled in Table S11.

Growth experiments. (i) Threshold determination.

The responsiveness of A. aromaticum EbN1^T to various concentrations of 3-phenylpropanoate, 3-(4-hydroxyphenyl)propanoate, cinnamate, and p-coumarate, respectively, was assessed via an experimental setup essentially as described in previous studies (28, 29). In short, benzoate-adapted cells of A. aromaticum EbN1^T were exposed to a single pulse of each tested 3-phenylpropanoid, yielding extracellular concentrations ranging from 100 µM down to 0.1 nM. Subsequently, samples for monitoring growth and time-resolved transcript profiling were retrieved from the cultures as described previously (28, 29): 5 min prior as well as 5, 15, 30, 60, and 120 min after addition of either 3-phenylpropanoid (effector). The respective detailed growth curves are provided in Fig. S1 to S4.

(ii) Mutant characterization.

To assess the effect of the ΔpprR deletion mutation (see “Generation of in-frame ΔpprR deletion mutation,” above), the wild type, the ΔpprR mutant, and the pprR-complemented mutant of A. aromaticum EbN1^T were grown with either succinate (8 mM), benzoate (4 mM), or p-coumarate (2 mM). Since cultures were inoculated with benzoate-adapted cells (described above), in the case of succinate and p-coumarate, substrate adaptation had to be conducted over five passages at first. Cultures of the pprR-complemented mutant contained kanamycin (50 µg ml⁻¹). For each substrate condition, three replicate cultures were conducted. These cultures were sampled in short intervals to monitor growth (OD₆₆₀) and substrate consumption. Additional cultures served to provide samples (harvested at ½OD_max) for targeted transcript and proteomic analyses (see below).

(iii) Binary substrate mixture experiments.

To test a possible involvement of catabolite repression-like processes in the regulation of ebA5316-21 transcription, benzoate-adapted cells of A. aromaticum EbN1^T (wild type and ΔpprR mutant) were grown with a binary mixture of p-coumarate (1 mM) and either benzoate (1 mM) or succinate (3 mM).

(iv) (Aromatic) acyl-CoA ester analyses.

For the quantification of intracellular (aromatic) acyl-CoA ester pool sizes, A. aromaticum EbN1^T (wild type and ΔpprR mutant) was adapted to anaerobic growth with benzoate, succinate, or p-coumarate. Corresponding cells were anaerobically grown with the respective substrate under nitrate-reducing conditions as described above. Substrate concentrations were 1 mM each for benzoate and p-coumarate and 3 mM for succinate. For each substrate condition, cultures were harvested at ½OD_max for analyses of (aromatic) acyl-CoA esters.

(v) Differential proteomics.

To investigate transport systems possibly involved in the uptake of 3-phenylpropanoids, A. aromaticum EbN1^T was grown with acetate (8 mM), benzoate (4 mM), or 3-(4-hydroxyphenyl)propanoate (2 mM). Again, cultures were adapted to each substrate over five passages and harvested at ½OD_max.

Cell harvesting. (i) Transcript analyses.

Samples were retrieved and processed as described previously (28, 29). Essentially, 5 ml of culture broth were withdrawn and immediately added to RNAprotect bacterial reagent (Qiagen, Hilden, Germany), mixed thoroughly, incubated for 5 min at ambient temperature, and centrifuged. Pellets were washed with 0.5 ml RNAprotect bacterial reagent (Qiagen), shock frozen in liquid N₂, and stored at −80°C until further analysis.

(ii) (Aromatic) acyl-CoA ester analyses.

Substrate-adapted cultures were transferred into N₂-flushed centrifuge beakers while maintaining a constant flow of N₂. After centrifugation (14,334 × g, 15 min,4°C), cell pellets were each resuspended in 1 ml methanol (ULC/MS; Biosolve, Valkenswaard, The Netherlands) while flushing with N₂ and transferred into sterile screw-cap 2-ml polypropylene tubes (Sarstedt, Germany) containing 0.6 g of glass beads (0.1-mm diameter) and 0.4 g zirconia beads (0.7-mm diameter) (Carl Roth, Karlsruhe, Germany). The samples were shock-frozen in liquid N₂ and stored at −80°C until further analysis.

(iii) Proteomic analysis.

Samples were retrieved and processed as previously described (58). In short, the entire 400-ml cultures were centrifuged (14,334 × g, 30 min, 4°C), and the pellets were washed in 250 ml washing buffer (100 mM Tris-HCl, 5 mM MgCl₂·6H₂O, pH 7.5) and resuspended in 0.8 ml of the same washing buffer. Following centrifugation (20,000 × g, 10 min, 4°C), pellets were shock frozen in liquid N₂ and stored at −80°C until further analysis.

Quantification of growth substrates by (micro)HPLC.

Quantitative depletion profiling of benzoate and p-coumarate was achieved using a micro-high-performance liquid chromatography (microHPLC) system (UltiMate 3000; ThermoFisher Scientific, Germering, Bavaria, Germany) as previously described (29). Essentially, compounds were separated via reverse-phase chromatography (Hypersil Gold column; ThermoFisher Scientific), applying an acetonitrile gradient and detected by means of a diode array detector. Benzoate was detected at 229 nm, with a retention time of 9.32 min and a dynamic range from 5 nM to 50 µM. p-Coumarate was detected at 309 nm, with a retention time of 8.84 min and a dynamic range from 25 nM to 50 µM.

Quantification of succinate was achieved with an HPLC system (UltiMate 3000; ThermoFischer) equipped with an ion exchange column (Eurokat H; 300 by 8 mm, 10-µm inner diameter; Knauer, Berlin, Germany) and an RI detector (RI-101; Shodex, Munich, Germany). The system was operated at 75°C with H₂SO₄ (5 mM) as the eluent provided, at a flow rate of 1.2 ml min⁻¹. Succinate was detected at a retention time of 6.51 min with a dynamic range from 25 µM to 10 mM.

Preparation of total RNA.

Preparation of total RNA was performed according to standard methods (50, 59) and as previously described (29). (i) Essentially, stored cell pellets were treated twice with hot acidic phenol (Roti Aqua-Phenol; CarlRoth, Karlsruhe, Germany), and the resultant aqueous phase was transferred into a 2-ml 5PRIME phase-lock gel tube (Quantabio, Beverly, MA, USA). After addition of one volume of phenol-chloroform-isoamyl alcohol (25:24:1), the tube was gently inverted for 5 min and centrifuged. (ii) Nucleic acids were precipitated and washed with ice-cold ethanol. (iii) The pellet was dried and resuspended in RNase-free water. Digestion with DNase I (RNase-free; Qiagen) was verified via PCR. The RNA quality of all 792 samples was assessed with an Experion automated electrophoresis station (Bio-Rad, Hercules, CA, USA). RNA concentration was determined using a TrayCell (Hellma Analytics, Müllheim, Germany) operated in a spectrophotometer (UV-1800; Shimadzu, Duisburg, Germany). RNA samples were stored at −80°C until further analysis. All chemicals used for RNA preparation were of molecular biology grade.

Transcript profiling by qRT-PCR.

For the selected target genes, specific primers (Table 2) were designed using the Primer3 software package (version 0.4.0; www.primer3.org). Synthesis of cDNA and real-time PCR was performed with 150 ng of total RNA using the Brilliant III ultrafast SYBR green quantitative reverse transcription-PCR (qRT-PCR) master mix (Agilent, Santa Clara, CA, USA) and analyzed in a CFX96 real-time system (Bio-Rad). PCR settings comprised initiation for 3 min at 95°C, 40 cycles of 10 s of denaturation at 95°C, 30 s of annealing at 60°C, and 30 s of extension at 60°C, followed by real-time detection for 5 s. For validation of specific product formation, melting-curve analysis was performed with intervals of 0.5°C, ranging from 60°C to 90°C. For time-resolved transcript profiling, all samples retrieved after effector addition represented test states, while a sample retrieved 5 min prior to effector addition served as the reference. For assessing the effect of the ΔpprR deletion mutation, relative transcript levels of each target gene were normalized against a housekeeping gene (gyrB; encoding DNA gyrase subunit B). Three biological replicates per sample were retrieved, for each of which 3 technical replicate measurements were performed. Relative transcript abundances were calculated, as previously described (28), according to the following equation:

$$\text{ratio}=E^{\Delta C_T\left(\text{reference}-\text{test}\right)}$$ (2)

TABLE 2.

Oligonucleotide primers used for targeted transcript profiling for 3-phenylpropanoid catabolic genes

Primera	Target gene	Nucleotide sequence (5′→3′)	Product length (bp)	PCR efficiencyb
ebA5316_F	ebA5316	GACGCCTACGGTGAAGACTG	266	1.89
ebA5316_R	ACCTTCAGGAAAGCCTGGTT
ebA5317_F	ebA5317	CCGAAAGCGGTCATCAATAC	266	1.97
ebA5317_R	GGCACGCTGAAATAGATCGT
ebA5319_F	ebA5319	GAGCTTCGATGCGTATTTCC	276	2.06
ebA5319_R	GTGTCCTTCGTCGCTTCC
ebA5321_F	ebA5321	GCCAGTATGGGAGATGTGGT	246	1.89
ebA5321_R	CCGAAGAGGCGAAAACTTC

^aF, forward primer; R, reverse primer.

^bMean value from all performed qRT-PCR experiments.

Primer-specific PCR efficiencies (E) were determined as described previously (60).

Identification and quantification of (aromatic) acyl-CoAs.

Cell breakage was achieved by vigorous shaking at 10 m s⁻¹ in a bead beater (MP Biomedicals, Schwerte, Germany) for 40 s, followed by cooling on ice for 2 min. This step was repeated twice. Then, the solution was centrifuged at 14,000 × g and 4°C for 10 min, and the supernatant was transferred to a 4-ml opaque vial. The remaining pellet was resuspended in 0.5 ml methanol, homogenized, and centrifuged, and the supernatant was transferred to the same 4-ml vial. This step was repeated once with 0.5 ml methanol. The solution was dried down under a nitrogen stream, and the residue was dissolved in 200 µl ultrapure water. The extract was transferred to a centrifuge filter (0.22-µm cellulose acetate filter, 2.0-ml polypropylene tube; Corning, Glendale, AZ, USA) and centrifuged under the same conditions as those described above.

The filtered solutions were analyzed by liquid chromatography on a Vanquish UHPLC system (ThermoFisher Scientific, Waltham, MA, USA) coupled with an Orbitrap Fusion mass spectrometer (ThermoFisher Scientific). The UHPLC was equipped with a Gemini C₁₈ column (150 by 2 mm, 3-µm pore size; Phenomenex, Torrance, CA, USA) and operated at 35°C with an eluent flow rate of 400 µl min⁻¹. The amount of injection was 5 µl. The mobile phase was a gradient composed of 10 mM ammonium formate at pH 8.1 (A) and acetonitrile (ULC-MS; Biosolve) (B). The gradient steps were the following: 0 to 2 min, 100% A; 2 to 20 min, 100% to 80% A; 20 to 23 min, 80% to 0% A; 23 to 27 min, 0% A; 27 to 29 min, 0% to 100% A; and 29 to 34 min, 100% A. The settings of the H-ESI-II source of the mass spectrometer were 4,000 V spray voltage, 350°C vaporizer temperature, 320°C ion transfer tube temperature, 40 arbitrary units sheath gas, and 10 arbitrary units auxiliary gas. The Orbitrap instrument was operated in full scan mode with a resolution of 60,000, a mass range of 750 to 950 Da, a 60% RF lens, an AGC target of 4.0e⁵, and a maximum injection time of 50 ms.

p-Coumaroyl-CoA, 4-hydroxybenzoyl-CoA, and benzoyl-CoA were identified by comparison with reference standards, which were synthesized according to the method of Kawaguchi et al. (61). The other coenzyme A esters were tentatively identified based on relative retention times and exact masses of the molecular ions. The individual coenzyme A thioesters were quantified by external calibration using synthetic benzoyl-CoA.

Proteomic analysis. (i) Whole-cell shotgun analysis.

Analysis of the soluble protein fraction was performed as previously described (29). In short, cells were disrupted by means of bead beating, and the soluble protein fraction was obtained via ultracentrifugation. After quantification using the Bradford assay (62), proteins were subjected to tryptic digestion. The peptide mixtures were decomplexed using a nanoLC system run in a trap column mode (UltiMate3000 nanoRSLC; ThermoFischer Scientific) applying an acetonitrile gradient. Continuous analysis proceeded via an online-coupled ion-trap mass spectrometer (amaZon speed ETD; Bruker Daltonik GmbH, Bremen, Germany) equipped with a captive spray ion source (Bruker Daltonik GmbH). Two different lengths of linear gradients were applied: 180 min, samples for mutant characterization (Fig. S6); 280 min, samples for comprehensive pathway analysis (Fig. 3B). Proteins were identified using the ProteinScape platform (version 3.1; Bruker Daltonik GmbH) on an in-house Mascot server (version 2.3; Matrix Science Ltd., London, UK) based on the complete genome sequence of A. aromaticum EbN1^T (27), employing a target-decoy strategy (63).

(ii) Membrane protein-enriched fraction.

Analysis of the membrane protein-enriched fraction was conducted essentially as described in a previous study (58). Briefly, cell disruption was achieved using a French press (French pressure cell press SLM Aminco; BioSurplus Inc., San Diego, CA, USA), and the resultant extract was treated with Na₂CO₃ (100 mM) prior to ultracentrifugation. Proteins were solubilized from the membrane fraction using SDS (1% [wt/vol]), followed by ultracentrifugation. The supernatant was shock-frozen in liquid N₂ and stored at −80°C until further analysis. Proteins were separated by SDS-PAGE (10 µg load). Then, sample lanes were cut into 8 slices, which were subjected to in-gel digestion. Subsequently, nanoLC electrospray ionization-tandem mass spectrometry analyses were applied using a 130-min linear gradient. Protein identification of the membrane protein-enriched fraction was performed as described above.

(iii) Estimation of EbA5316 (SBP) copy numbers per cell.

As a basis for the estimation of EbA5316 copy numbers, we selected six ribosomal proteins (RplADFIJ and RpmC) of the 50S subunit of A. aromaticum EbN1^T (see Fig. S8 in the supplemental material), assuming comparable copy numbers of ribosomal proteins in proteobacterial A. aromaticum EbN1^T and E. coli during active growth. Accordingly, experimentally determined peptide counts of the six selected ribosomal proteins of A. aromaticum EbN1^T were set in relation to the quantified copy numbers of the respective E. coli homologues previously reported by Lu et al. (35). This relation was calculated as

$$M_i\cdot {\left(\frac{c_i}{f_i}\right)}_{\text{EbN1}}={\left(N_i\right)}_{E.\mathit{coli}}$$ (3)

where (N_i)_{E. coli} is the copy number (APEX) of protein i (running over RplADFIJ and RpmC) per cell determined for E. coli (35), c_i denotes the peptide counts of the protein i, and f_i represents the number of detected tryptic peptides for that protein. We assume that the detection probability of a tryptically digested protein i is proportional to the number, N_i, of tryptic peptides (≥5 amino acids). This procedure (equation 3) yields six protein-specific multipliers, M_i, from which we take the average value, $\overline{M}$, for converting measured peptide counts into estimated protein copy numbers of EbA5316 in A. aromaticum EbN1^T:

$$\overline{M}\cdot {\left(\frac{c_i}{f_i}\right)}_{\text{EbN1}}={\left(N_i\right)}_{\text{EbN}1}$$ (4)