Simultaneous carbon catabolite repression governs sugar and aromatic co-utilization in Pseudomonas putida M2

Strain M2 grows on both hexose (glucose) and pentose (xylose, arabinose) sugars and biomass-derived aromatic monomers [p-coumarate (pCA), 4-hydroxybenzoate (4-HBA), ferulate (FA), and vanillate (VA)] (Fig. S1), consistent with metabolic pathways identified in its genome (10). Previous studies have shown that Pseudomonads, such as KT2440, prefer sugars over aromatic compounds (14, 17, 18), which complicates the development of an industrially relevant strain capable of efficiently utilizing both sugar and aromatic compound. We, therefore, investigated the growth phenotypes of M2 in a mixture of glucose/xylose+aromatic compound (Fig. 1; Fig. S2; Tables S1 to S5).

Both glucose and aromatic compound (pCA, 4-HBA, FA, or VA) were simultaneously metabolized (Fig. 1; Fig. S3) in M2; however, both compounds were not consumed completely after 48h. In general, glucose between 52.1% and 76.9% was depleted, while it was observed that aromatic compound between 58.6% and 81.7% was consumed over a period of 60h. Also, M2 exhibited a longer lag phase in glucose and aromatic compound mixtures compared to growth in single aromatic compound conditions (Fig. 1; Fig. S1 and S3). M2 exhibited a 24-h lag time when grown in glucose+pCA after which simultaneous consumption of both substrates commenced. It should be noted that M2 also showed long lag phase and incomplete substrate consumption when grown in glucose alone. In contrast, growth on xylose and aromatic compounds showed no lag, and the aromatic compounds were consumed faster than the xylose. After 48h, xylose consumption was not complete, as was observed for glucose. For example, 100% of the pCA was depleted within 9h in growth experiments with xylose+pCA, at which time approximately 29% of xylose was consumed. During growth on both glucose and xylose, aromatic intermediates 4-HBA (in pCA) and VA (in FA), accumulated during the cultivation. In the case of glucose, the aromatic intermediates persisted as the uptake of pCA and FA ceased (Fig. 1b and c). For xylose growth, transient accumulation of 4-HBA and VA was observed (Fig. 1e and f). The co-consumption of glucose and aromatic compounds in strain M2 was compared to co-consumption by KT2440 under the same conditions. Both glucose and aromatic compounds were consumed simultaneously in KT2440 with aromatic compounds consumed at a faster rate; unlike M2, both substrates were completely metabolized (Fig. 1h and i). Furthermore, the metabolic intermediates 4-HBA and VA accumulated transiently in KT2440 during the growth consistent with previous studies (14, 17).

Phenotypic growth data of M2 and KT2440 on sugars and aromatic compounds indicated simultaneous CCR was occurring. To study these effects on protein levels, the proteome of M2 and KT2440 grown in dual carbon conditions (sugar+aromatic compound, Table S6) were analyzed and compared with those grown in a minimal medium containing sugar only or aromatic compound only (control). We assessed the differentially abundant proteins using a criterion of P ≤ 0.05 and log₂ fold change (FC) ≥0.5 value (see Supplementary Information Tables S6 to S26 for a complete list of the differentially abundant proteins).

For M2, the proteins involved in the catabolism of pCA and FA to 4-HBA and VA, respectively, such as feruloyl-CoA synthase (Fcs), feruloyl-CoA hydratase-lyase (Ech), and vanillin dehydrogenase (Vdh) were present at lower abundance (−1.3 to −3.1 log₂ FC) in glucose+pCA and glucose+FA conditions compared to the control cells grown in pCA or FA only (Fig. 2a and b; Tables S7 and S9). The abundances of 4-hydroxybenzoate hydroxylase (PobA) and vanillate O-demethylase monooxygenase subunit (VanA) responsible for converting 4-HBA and FA, respectively, to protocatechuate (PCA), were also significantly lower (−1.9 log₂ FC for PobA and −2.3 log₂ FC for VanA). These results are in agreement with the accumulation of intermediates 4-HBA and VA observed during the growth in glucose+pCA or FA mixture (Fig. 1). Many of the proteins in the PCA branch of the β-ketoadipate pathway (PcaHG, PcaB, PcaC, and PcaD) and β-ketoadipate conversion to the tricarboxylic acid (TCA) intermediates (PcaIJ and PcaF) were also found at significantly lower abundances in dual substrate conditions compared to the control even though PCA was not observed to accumulate during growth. Similar results were observed for M2 cells grown on glucose+4HBA or VA where PobA (−1.7 log₂ FC), Van A (−2.8 log₂ FC), and most of the β-ketoadipate pathway proteins (Fig. S4; Tables S8 and S10) were at significantly lower abundance in dual substrate growth conditions. For KT2440, the aromatic catabolic proteins (−1.0 to −2.5 log₂ FC) and most of the β-ketoadipate (−0.6 to −2.4 log₂ FC) proteins, with the notable exception of Ech, which is involved in the catabolism of both pCA and ferulate, were also present at significantly lower abundance (Fig. S5; Tables S23 to S26). The lower abundances observed during glucose+aromatic compound growth in KT2440 were consistent with previous proteomic studies of substrate co-utilization (14, 17).

Comparing the proteome of M2 cells grown in glucose +aromatic compound with glucose as the sole carbon source identified that 2-keto-3-deoxy-6-phosphogluconate (KDPG) aldolase (Eda), which converts KDPG into glyceraldehyde-3-phosphate and pyruvate, was present at significantly lower abundance (−0.8 log₂FC, Fig. 2d; Table S11), consistent with incomplete glucose consumption. A subset of TCA and glyoxylate cycle proteins [isocitrate dehydrogenase (Icd), aconitate hydratase (Acn), malate synthase (GlcB), and isocitrate lyase (AceA)] were also observed in significantly lower abundance compared to the single carbon conditions (Fig. 2e). Additionally, some amino acid biosynthesis proteins (2-isopropylmalate synthase, glutamine synthetase, O-succinylhomoserine sulfhydrylase), catabolic proteins (ornithine decarboxylase and ornithine carbamoyltransferase), and transporters were also significantly at lower abundance (Tables S7 to S14).

When grown in minimal media containing xylose and pCA or FA, only some of the aromatic catabolic and β-ketoadipate proteins were present at lower abundance, unlike what was observed in the presence of glucose (Table S15 to S18; Fig. 2a and b). Interestingly, pyruvate carboxylase subunits A and B proteins, which catalyze the first step of gluconeogenesis and replenish TCA cycle intermediates, were also found at significantly lower abundances (−0.5 to −1.9 log₂ FC for subunit A and −1.0 to −2.7 log₂ FC for subunit B) (Fig. 2e). For xylose metabolism, co-feeding pCA lowered the relative abundance of 2-keto-3-deoxyxylonate dehydratase (XylX) (−0.7 log₂FC, Fig. 2d; Table S19), a key xylose catabolic protein, which dehydrates α-keto-3-deoxy-xylonate to α-ketoglutarate semialdehyde, thus explaining the incomplete consumption of xylose. Amino acid biosynthesis (threonine synthase, 2-isopropylmalate synthase, acetylornithine/N-succinyldiaminopimelate aminotransferase, glutamate N-acetyltransferase/amino-acid N-acetyltransferase, tryptophan synthase alpha chain) and transport proteins, as well as ribosomal proteins, were present at significantly lower abundance during xylose +aromatic compound growth compared to when only single carbon was present (Tables S15 to S22).

As Crc has been identified as the main regulatory protein governing CCR in Pseudomonads (12), the crc gene expression was reduced using a previously developed dual-inducible CRISPR/dCas9-based gene repression system to determine if the aromatic and sugar metabolism in M2 are subject to Crc regulation (24, 25). The functionality of the arabinose-inducible pRGPspdCas9bad-edd plasmid (the backbone plasmid used to design CRISPRi strains) was confirmed to verify sgRNA expression (Fig. S6). The pRGPspdCas9bad-edd strain showed delayed growth on glucose compared to the control due to sgRNA-targeted repression of the edd gene, an essential gene for glucose metabolism.

We compared the growth phenotype (Fig. 3; Fig. S7 and S8; Tables S2 to S5) and proteomes (Table S27) of the CRISPRi strains to the control strain containing empty plasmid (pRGPspdCas9bad, Table 1) in a minimal medium containing glucose +pCA or FA and xylose +pCA. The CRISPRi strains (crc-1, crc-2, and crc-3) grew faster than the control strain (maximal growth rate of 0.17–0.23 h⁻¹ vs 0.11-0.16 h⁻¹, Table S2) upon induction when fed with a 10:1 mixture of glucose +pCA or FA. The CRISPRi strains grew within 9–12 h, while the control strain exhibited a much longer lag phase (32–36 h, Fig. 3; Fig. S7). However, the growth profiles were similar between the CRISPRi and control strains in the presence of xylose +pCA. The substrate consumption results also showed that both glucose and aromatic compound (pCA or FA) were consumed faster by the CRISPRi strains than the control (Tables S4 and S5 ), while xylose and pCA consumption profiles were identical in all strains, thus corroborating the OD₆₀₀ measurements. However, both glucose and pCA/FA were still not completely consumed in both the CRISPRi strains as observed in the control experiment. It should be noted that the expression of the sgRNA targeting crc gene was under the control of the arabinose inducible P_araC/bad promoter system. Even though M2 can use arabinose as a carbon source for its growth, we confirmed that arabinose was not consumed in the presence of a high concentration of glucose (10 g L⁻¹, Fig. S9) and therefore the faster growth observed in CRISPRi strains compared to the control can be attributed to decrease in crc expression during glucose +pCA or FA conditions and not due to the presence of additional carbon source.

The MS-based proteomics showed that Eda was significantly upregulated (0.8 log₂FC, Fig. 3g; Table S27) in crc-1 which might explain faster glucose consumption than the control strain. The proteins involved in amino acid biosynthesis (acetolactate synthase, glutamine synthetase, 3-isopropylmalate dehydratase, 3-isopropylmalate dehydrogenase, aspartate kinase) and transport were also significantly upregulated in the CRISPRi strains possibly explaining faster growth compared to the control (Table S27). However, there was no significant difference in the aromatic catabolic and the β-ketoadipate proteins between the CRISPRi strains (except for upregulation of PcaJ in the crc-1 strain) and the control grown in media containing glucose+pCA (Fig. 3h).

Bacterial sRNAs are non-coding RNAs, ranging from 50 to 500 nt, that are found to be involved in gene regulation (26). In Pseudomonads, the Crc protein interacts with Hfq, RNA chaperon protein, creating a stable Hfq/Crc/mRNA complex (20). The strength of CCR is inversely proportional to the abundance of one or more sRNAs as they antagonize Crc by sequestering Hfq (22). Their abundance is low under conditions that elicit strong CCR and high in the absence or low catabolic repression (27–29). Moreno et al. (27) showed that the levels of sRNAs, CrcY and CrcZ, in P. putida PBA1 were lower in cells growing exponentially in lysogeny broth (LB, strong CCR conditions) than in minimal media containing citrate or succinate (non-repression conditions). Under non-repressing conditions, CrcY and CrcZ sequestered Crc and impeded its binding to target mRNAs.

In the current study, the abundance of CrcY and CrcZ in KT2440 correlated with catabolite repression observed in the phenotypic study as well as with the proteomics results. The CrcY and CrcZ levels were lowest in the dual carbon condition compared to that in the single substrate condition (Fig. 4), showing potential catabolite repression. A search of M2 sRNA sequences identified homologous sequences with 94% and 97% similarities to KT2440 CrcY and CrcZ, respectively. The expression of these homologues in M2 was lower under the glucose +pCA condition (strong CCR) than when only pCA or glucose (non-repressing condition) was present, possibly allowing the Crc protein to block the translation of the target mRNAs. On the other hand, when pCA or glucose was provided as the sole carbon source, the CrcY and CrcZ levels were higher compared to glucose +pCA condition, sequestering Crc by binding to it with high affinity as catabolite repression is not needed. Furthermore, CrcZ and CrcY in both KT2440 (six and four repeated motifs in CrcZ and CrcY, respectively) and M2 (six repeated motifs in both sRNA homologues) contain several repeated AANAANAA boxes, which have been described as Crc binding site (20, 30). These preliminary results suggest that the levels of sRNAs might have a role in controlling the amount of free Crc protein and, therefore, the strength of catabolite repression in KT2440 and M2 as already demonstrated in other P. putida strains.

All microbial cultivation experiments were conducted in triplicates at 30°C using 10mL 0.2µm filter sterilized modified M9 mineral medium (final pH adjusted to 7) with the following composition: (NH₄)₂SO₄ (1.0 g L⁻¹), KH₂PO₄ (1.5 g L⁻¹), Na₂HPO₄ (3.54 g L⁻¹), ammonium ferric citrate (0.06 g L⁻¹), MgSO₄·7H₂O (0.2 g L⁻¹), CaCl₂·2H₂O (0.01 g L⁻¹), and trace elements H₃BO₃ (0.3 mg L⁻¹), CoCl₂·6H₂O (0.2 mg L⁻¹), ZnSO₄·7H₂O (0.1 mg L⁻¹), MnCl₂·4H₂O (0.03 mg L⁻¹), NaMoO₄·2H₂O (0.03 mg L⁻¹), NiCl₂6H₂O (0.02 mg L⁻¹), and CuSO₄·5H₂O (0.01 mg L⁻¹). The M2 and KT2440 strains were inoculated from glycerol stocks onto an LB agar plate and a single colony from the plate was grown in 5mL of LB medium (first preculture) overnight with continuous shaking at 200rpm. The first preculture (100 µL) was inoculated into 5mL freshly prepared M9 minimal medium with a corresponding carbon source to prepare the second preculture and incubated overnight. Finally, the growth curve and growth kinetics experiments were started the following day from the second preculture at a desired starting optical density (0.025–0.05) in 50mL glass tubes and 48-well plates with 250µL cell culture in each well, respectively. The minimal media was supplemented with a mixture of sugar (glucose or xylose) and aromatic compound (pCA, 4-HBA, FA, or VA) as model lignocellulose-derived substrates at a concentration of~10 g L⁻¹ (sugars) and~1 g L⁻¹ (aromatic compounds). For the growth curve experiments, the cell culture samples were taken at different time points to determine growth spectrophotometrically (SpectraMax M2, Molecular Devices, San Jose, CA) by measuring absorbance at 600nm (OD₆₀₀). Growth kinetics was done using a Synergy plate reader (BioTek Instruments, Winooski, VT, USA).

Quantification of sugars was performed using high performance liquid chromatography (HPLC) using Agilent 1260 Infinity system (Santa Clara, CA, USA) equipped with an Aminex HPX-87H (300 mm × 7.8 mm) column (Bio-Rad, Hercules, CA) and a refractive index detector heated at 35°C. An aqueous solution of 4 mM sulfuric acid was used as the mobile phase (0.4 mL min⁻¹, column temperature of 25°C). The amount of monomeric aromatic compounds was analyzed using Agilent 1200 system with Eclipse Plus Phenyl-Hexyl (250 mm length, 4.6 mm diameter, 5 µm particle size) column kept at 50°C. The mobile phase consisted of binary system phases with 10 mM ammonium acetate in 0.07% formic acid (Solvent A) and 10 mM ammonium acetate in 90% acetonitrile and 0.07% formic acid (Solvent B). The mobile phase gradient method was as follows: 30% B (0 min; 0.5 mL min⁻¹), 80% B (12 min; 0.5 mL min⁻¹), 100% B (12.1 min; 0.5 mL min⁻¹), 100% B (12.6 min; 1 mL min⁻¹), 30% B (12.8 min; 1 mL min⁻¹), and 30% B (15.6 min; 1 mL min⁻¹). The metabolite concentration was quantified from peak areas using calibration curves generated by each metabolite standard.

All strains were grown in three biological replicates at 30°C following a similar procedure as stated in the culturing conditions section above starting with an LB agar plate followed by the first and second preculture before growing them in 10mL M9 minimal media supplemented with corresponding carbon sources (Table S6) at 30°C. The cells were harvested by centrifugation at OD between 0.4 and 0.5 and stored at −80°C until sample preparation. Proteins from the samples were extracted using a previously described chloroform/methanol precipitation method (55). Extracted proteins were resuspended in 100mM ammonium bicarbonate buffer supplemented with 20% methanol, and protein concentration was determined by the DC assay (BioRad). Protein reduction was accomplished using 5mM tris 2-(carboxyethyl)phosphine (TCEP) for 30min at room temperature, and alkylation was performed with 10mM iodoacetamide (IAM; final concentration) for 30min at room temperature in the dark. Overnight digestion with trypsin was accomplished with a 1:50 trypsin:total protein ratio. The resulting peptide samples were analyzed on an Agilent 1290 UHPLC system coupled to a Thermo scientific Obitrap Exploris 480mass spectrometer for the discovery proteomics (56). Briefly, 20µg of tryptic peptides were loaded onto an Ascentis (Sigma–Aldrich) ES-C18 column (2.1 mm × 100mm, 2.7µm particle size, operated at 60°C) and were eluted from the column by using a 10-min gradient from 98% buffer A (0.1% FA in H2O) and 2% buffer B (0.1% FA in acetonitrile) to 65% buffer A and 35% buffer B. The eluting peptides were introduced to the mass spectrometer operating in positive-ion mode. Full MS survey scans were acquired in the range of 300–1200 m/z at 60,000 resolution. The automatic gain control (AGC) target was set at 3E06 and the maximum injection time was set to 60ms. The top 10 multiply charged precursor ions (2–5) were isolated for higher-energy collisional dissociation (HCD) MS/MS using a 1.6-m/z isolation window and were accumulated until they either reached an AGC target value of 1e5 or a maximum injection time of 50ms. MS/MS data were generated with a normalized collision energy (NCE) of 30, at a resolution of 15,000. Upon fragmentation, precursor ions were dynamically excluded for 10s after the first fragmentation event. The acquired LCMS raw data were converted to mgf files and searched against the latest UniProt P. alloputida KT2440 protein database or JGI P. putida M2 protein database (IMG submission ID: 236130) with Mascot search engine version 2.3.02 (Matrix Science). The resulting search results were filtered and analyzed by Scaffold version 5.0 (Proteome Software Inc.). The normalized spectra counts of identified proteins were exported for relative quantitative analysis. The proteomics data were analyzed using a customized Python script to calculate log₂ FC values and identify proteins that were significantly downregulated in the dual substrate conditions compared to the single substrate condition. Only proteins with spectral counts above five were retained in the analysis. Proteins were identified to be significantly differentially expressed when they displayed a log₂ FC ≥ 0.5 across at least two out of three biological replicates with the P-value cut-off of 0.05.

Escherichia coli TOP10 cells were used as the primary host for gene cloning and plasmid isolation. Kanamycin (50 µg mL⁻¹) was used for the selection of plasmid-harboring cells. E. coli competent cell preparation and transformation were performed using standard molecular biology techniques (57). Electroporation (1 mm gap cuvette, with parameters set at the resistance of 200 Ω, voltage 1.8 KV, capacitance of 25 μFD for 4.0–5.0 ms) was used for the transformation of competent M2 cells with selected plasmids. Cells were incubated in LB medium for 1.5 h at 30°C and 200 rpm on a shaker incubator and plated on LB plates with kanamycin for selection. Q5 High-Fidelity 2X and OneTaq 2X Master Mix (New England Biolabs; Ipswich, MA, USA) were used according to the manufacturer’s instructions for PCR amplification during gene cloning and colony PCR to confirm gene cloning, respectively. NEBuilder HiFi DNA Assembly Master Mix (New England Biolabs) was used to assemble plasmids. QIAprep Spin miniprep kit and QIAquick gel extraction kit (Qiagen, Hilden, Germany) were used according to the manufacturer’s protocol to isolate plasmids and to separate PCR products from agarose gels, respectively. The sequences of all plasmids and DNA constructs were confirmed by Sanger sequencing (Genewiz, CA, USA).

The CRISPR interference system utilizing the type II dCas9 homologs from Streptococcus pyogenes (pRGPspdCas9bad, Table 1) previously developed for KT2440 and M2 (24, 25) was used as the backbone plasmid to develop a dual inducible CRISPR/dCas9 interference-based gene repression system in M2. Expression of spdCas9 was under the control of the P_tac [Isopropyl β-D-1-thiogalactopyranoside (IPTG)-inducible] promoter, while sgRNA expression was under the control of the P_araC/bad (arabinose-inducible) promoter. The wild-type M2 strain carrying an empty pRGPspdCas9bad plasmid served as the control. An open reading frame (Ga0436255_01_2263068_2263847) was found which had 95% sequence similarity to the KT2440 crc sequence in the M2 genome. Three sgRNAs (20 bp) targeting three different regions of the crc gene (Fig. S10), resulting in three recombinant strains (crc-1, crc-2, and crc-3) were designed and cloned into the plasmid by PCR amplifications using primers listed in Table 2. One sgRNA targeted a region close to the start codon at the 5′ end (crc-1) and two inside the ORF (crc-2 and crc-3) of the crc gene. The constructed plasmids were electroporated into M2 to generate the CRISPRi strains. The recombinant CRISPRi strains were grown in minimal media supplemented with corresponding carbon sources (glucose +pCA, glucose +FA, and xylose +pCA), and both inducers (~3.5 g L⁻¹ arabinose and 1 mM IPTG) for phenotypic characterization (following a similar protocol as stated in the culturing conditions section above) and proteome comparison (glucose, pCA, or glucose +pCA as carbon sources) with the control strain.

The oligonucleotides were ordered from Integrated DNA Technologies (IDT, San Diego, CA, USA). All the bacterial strains, plasmids, and primers used in this study can be found in Tables 1 and 2.

M2 and KT2440 cells were grown in minimal media containing either glucose, pCA, or glucose +pCA, and cell pellets were submitted to CD genomics (NY, USA) for total RNA extraction and sRNA sequencing. Briefly, total RNA was isolated using Qiagen miRNeasy mini kit (Hilden, Germany) following the manufacturer’s protocol, quantified by NanoDrop 2000 (Thermo Fisher Scientific, Waltham, MA, USA), and quality assessed with agarose gel electrophoresis. The preparation of the sRNA-enriched library was done with NEBNext Ultra II RNA kit (New England Biolabs, Ipswich, MA, USA) following the manufacturer’s recommendation which included cDNA synthesis, 3’ and 5’ adapter ligation, PCR enrichment of adapter-ligated RNA, and size fractionation (80–400 bp) with Pippin Prep to specifically enrich for sRNAs. The library quantity was measured by KAPA SYBR FAST qPCR, while the quality was evaluated by TapeStation D1000 ScreenTape (Agilent Technologies, CA, USA). Equimolar pooling of libraries was performed based on the QC values and sequenced on Illumina NovaSeq 6000 (Illumina, CA, USA) with a read length configuration of 150 paired-end for 10 million paired-end reads per sample.

Illumina adapter sequences and small RNA library adapter sequences were removed from raw sequencing reads using Fastp and Cutadapt, respectively. The coding mRNA sequences were removed by aligning the quality-filtered RNA-seq reads with the reference genome. We ran BLASTX against sequences similar to CrcY and CrcZ, the sRNAs found in KT2440 (27), and determined their abundances in both M2 and KT2440 under different carbon conditions.