Metabolic engineering of Pseudomonas bharatica CSV86T to degrade Carbaryl (1-naphthyl-N-methylcarbamate) via the salicylate-catechol route

ABSTRACT Pseudomonas bharatica CSV86T displays the unique property of preferential utilization of aromatic compounds over simple carbon sources like glucose and glycerol and their co-metabolism with organic acids. Well-characterized growth conditions, aromatic compound metabolic pathways and their regulation, genome sequence, and advantageous eco-physiological traits (indole acetic acid production, alginate production, fusaric acid resistance, organic sulfur utilization, and siderophore production) make it an ideal host for metabolic engineering. Strain CSV86T was engineered for Carbaryl (1-naphthyl-N-methylcarbamate) degradation via salicylate-catechol route by expression of a Carbaryl hydrolase (CH) and a 1-naphthol 2-hydroxylase (1NH). Additionally, the engineered strain exhibited faster growth on Carbaryl upon expression of the McbT protein (encoded by the mcbT gene, a part of Carbaryl degradation upper operon of Pseudomonas sp. C5pp). Bioinformatic analyses predict McbT to be an outer membrane protein, and Carbaryl-dependent expression suggests its probable role in Carbaryl uptake. Enzyme activity and protein analyses suggested periplasmic localization of CH (carrying transmembrane domain plus signal peptide sequence at the N-terminus) and 1NH, enabling compartmentalization of the pathway. Enzyme activity, whole-cell oxygen uptake, spent media analyses, and qPCR results suggest that the engineered strain preferentially utilizes Carbaryl over glucose. The plasmid-encoded degradation property was stable for 75–90 generations even in the absence of selection pressure (kanamycin or Carbaryl). These results indicate the utility of P. bharatica CSV86T as a potential host for engineering various aromatic compound degradation pathways. IMPORTANCE The current study describes engineering of Carbaryl metabolic pathway in Pseudomonas bharatica CSV86T. Carbaryl, a naphthalene-derived carbamate pesticide, is known to act as an endocrine disruptor, mutagen, cytotoxin, and carcinogen. Removal of xenobiotics from the environment using bioremediation faces challenges, such as slow degradation rates, instability of the degradation phenotype, and presence of simple carbon sources in the environment. The engineered CSV86-MEC2 overcomes these disadvantages as Carbaryl was degraded preferentially over glucose. Furthermore, the plasmid-borne degradation phenotype is stable, and presence of glucose and organic acids does not repress Carbaryl metabolism in the strain. The study suggests the role of outer membrane protein McbT in Carbaryl transport. This work highlights the suitability of P. bharatica CSV86T as an ideal host for engineering aromatic pollutant degradation pathways.

Bioremediation using microbes has been proposed to be a desirable alternative to abiotic removal of xenobiotics (11)(12)(13)(14)(15).However, application of natural isolates poses limitations, such as inefficient degradation, poor survival due to abiotic stress, limited metabolic diversity, repression of xenobiotic metabolism in the presence of simple carbon sources, among others (15)(16)(17)(18).These limitations can be resolved by applica tion of recombinant DNA techniques to construct "patchwork assembly" of various pathways into a single organism/host, referred to as metabolic engineering.An ideal host for such assembly should exhibit various traits, such as well-characterized genome sequence and growth conditions, stress tolerance as well as availability of host-spe cific genetic manipulation tools, among others (19).The heterologous expression of degradative enzymes can either be plasmid or chromosome based.For example, the pSEVA plasmids (shuttle vectors) are widely used expression vectors for metabolic engineering of Pseudomonas spp.(20).Whereas, chromosomal integration can be carried out using homologous recombination, transposons, or CRISPR/Cas9-based vectors for phenotypic stability (19)."Patchwork assembly" approaches have been employed to engineer strains for degradation of aromatic pollutants through diverse pathways.For example, Pseudomonas spp.have been engineered for chloro-and methyl-aromatic compound catabolism (21), P. putida KT2440 for carbofuran hydrolysis (22), Azoarcus communis SWub3 for anaerobic benzoate degradation (23), Escherichia coli for 4-fluorophenol mineralization (24), P. putida KTU for 1,2-dichloroethane degradation (25), among others.These approaches have also been combined with directed evolution of degrada tive enzymes and proteins to enhance the substrate range and degradation efficiency (26).
Pseudomonas bharatica CSV86 T , an Indian soil bacterium isolated from petroleum product-contaminated soil, degrades a wide range of mono-and poly-cyclic aromatic hydrocarbons efficiently (12-14 h) at high concentrations (up to 1% wt/vol) through diverse metabolic routes (27,28).Strain CSV86 T exhibits the novel property to metab olize aromatic compounds preferentially over simple carbon sources like glucose and glycerol, and co-metabolize them with organic acids (28)(29)(30).This trait makes CSV86 T unique among various reported Pseudomonas spp., which exhibit carbon catabolite repression of aromatic compound utilization in the presence of both organic acids and glucose (31).The carbon catabolite repression in Pseudomonas spp. is primarily orchestrated by CbrA/CbrB-Hfq/Crc-crcZ system, wherein the Crc-Hfq complex binds the mRNA of secondary carbon source utilization genes, thereby repressing translation in the presence of preferred carbon sources.The crcZ (or its homologs crcY or crcX) small RNA sequesters the Crc-Hfq complex to relieve repression.The expression of these small RNAs is controlled by the CbrA/CbrB two-component regulatory system, although the underlying signals are unknown (32).However, the specific regulatory mechanisms responsible for the unique preferential utilization phenotype of strain CSV86 T remain elusive.
The metabolism of naphthalene in P. bharatica CSV86 T is initiated by the concer ted action of naphthalene dioxygenase and cis-dihydrodiol dehydrogenase to form 1,2-dihydroxynaphthalene, which is subsequently metabolized to salicylate.Further more, salicylate-1-hydroxylase converts salicylate to catechol, which is ring cleaved by the action of catechol-2,3-dioxygenase (C23DO) to yield aliphatic intermediate that enters the central carbon pathway (33,37; Fig. 1).Genes encoding naphthalene degradation enzymes are present as two operons (nah and sal operon; 13 kb apart) on an integrative and conjugative element ICEnahCSV86 in strain CSV86 T (34; Fig. S1A).The nah operon (nahAaAbAcAdBFCED) consists of genes coding for enzymes involved in the metabolism of naphthalene to salicylate.Whereas, the sal operon (nahGTHINLOMKJX) consists of genes encoding enzymes involved in the conversion of salicylate to central carbon intermediates.The transcription of nah operon is driven by the promoter Pnah while that of sal operon is driven by Psal (34,38; Fig. S1A).
The bacterial degradation of Carbaryl is initiated by the action of Carbaryl hydrolase (CH) to generate 1-naphthol, a highly toxic metabolic intermediate (39).1-Naphthol is further hydroxylated by 1-naphthol-2-hydroxylase (1NH) to yield 1,2-dihydroxynaph thalene.This intermediate is metabolized to salicylate through a series of enzymatic steps and funneled into central carbon pathway either through catechol or gentisate route (4).Based on biochemical evidence, it is proposed that Carbaryl degradation in Pseudomonas sp.C5pp proceeds via formation of 1-naphthol, which is converted to 1,2-dihydroxynaphthalene and further, salicylate.The action of salicylate-5-hydroxylase generates gentisate, which is ring cleaved to aliphatic intermediates by the action of gentisate dioxygenase (GDO) (7; Fig. 1).Genes encoding enzymes for the conversion of Carbaryl via 1,2-dihydroxynaphthalene to salicylate are arranged as Carbaryl degradation "upper operon" (40, 41; Fig. S1B).The presence of common metabolic steps involved in Carbaryl and naphthalene degradation (Fig. 1) can be exploited to increase the metabolic diversity of strain CSV86 T by pathway engineering.
In the present study, strain CSV86 T was engineered to metabolize Carbaryl via salicylate-catechol route by expressing CH, 1NH, and the putative Carbaryl transporter, McbT from Pseudomonas sp.C5pp (Fig. 1).The engineered strain was characterized at biochemical and molecular level for Carbaryl metabolism, phenotypic stability, and preferential utilization property.

Co-expression and localization of Carbaryl hydrolase and 1-naphthol 2-hydroxylase
The metabolic steps from 1,2-dihydroxynaphthalene to salicylate are common for naphthalene degradation in strain CSV86 T and Carbaryl degradation in strain C5pp (Fig. 1; Table 1).The promoter Pnah from CSV86 T was found to be leaky and induced to high levels using naphthalene or salicylate, indicating it to be a strong promoter (using construct pSEVAPnah-1NH derived from pSEVA234; Table 2) (38).This promoter was used to engineer CSV86 T for Carbaryl degradation.In Carbaryl-degrading soil isolate Pseudomonas sp.C5pp, the presence of 96-amino-acid-long N-terminus transmembrane domain plus signal peptide (Tmd + Sp) region as a part of Carbaryl hydrolase enzyme (encoded by mcbA) aids in localizing it (~72 kDa, mature CH) to the periplasmic space (42).We constructed pMEC1 (Pnah→CH-1NH), which harbors full-length mcbA (encoding CH carrying Tmd + Sp at N-terminus to localize mature CH to periplasm and 6× His tag at C-terminus) and mcbC (encoding 1NH) under the Pnah promoter (see Materials and Methods for details; Table 2; Fig. S2A).
To assess the functionality of pMEC1 in strain CSV86 T (transformant referred as CSV86-MEC1), expression and localization of CH and 1NH were monitored by measuring enzyme activity, SDS-PAGE, and Western blot analysis.The activities of CH, 1NH, and ring-cleaving enzyme (catechol-2,3-dioxygenase) were monitored from the periplasmic, cytoplasmic, and cell-free extract (CFE; which contains both cytoplasmic and periplasmic proteins) fractions prepared from CSV86-MEC1 cells grown on naphthalene.Naphtha lene was chosen as the growth substrate for CSV86-MEC1 for this analysis as it is the native inducer of the Pnah promoter (38).Subsequently, naphthalene-grown cells were used as inoculum for carrying out growth, metabolic, and biochemical analyses of the engineered strain.The strain CSV86 T naturally degrades naphthalene, and various global-level adaptive responses occur while growing on naphthalene which might be advantageous for Carbaryl degradation by the engineered strain.Enzyme activities from naphthalene-grown wild-type CSV86 T (which lacks CH and 1NH enzymes) and Carbarylgrown Pseudomonas sp.C5pp were taken as negative and positive control, respectively.
The CH activity was determined from various fractions to assess the functionality of the intrinsic N-terminus signal sequence, Tmd + Sp, to translocate CH to periplasmic space in strain CSV86-MEC1.The activity of 1NH (which lacks a signal sequence) was examined in various fractions to determine its localization, as certain aromatic hydroxy lases of Pseudomonas spp., which lack a signal sequence, have been reported to be localized to the periplasm (44,45).The activity of ring-cleaving dioxygenase C23DO [encoded by nahH gene of the naphthalene degradation lower operon in CSV86 T (sal operon)], which is reported to be cytoplasmic enzyme (38), was taken as a control to ensure proper fractionation of CSV86-MEC1 cells, i.e., no contamination of cytoplasmic proteins in the periplasmic fraction.Similarly, the activity of the cytoplasmic enzyme gentisate dioxygenase (encoded by mcbO gene of Carbaryl degradation lower operon in C5pp) was taken as control to ensure efficient fractionation of C5pp cells (42).
The SDS-PAGE analysis of these fractions from CSV86-MEC1 failed to show any detectable protein band corresponding to CH (~72 kDa; Fig. 2A).Western blot anal ysis using anti-His tag antibodies revealed the presence of immunoreactive protein a Km R represents kanamycin resistance.In the nucleotide sequence, restriction sites (in parentheses) are indicated by underlining, ribosome-binding sites are indicated in italics, and 6× His tTag is marked in bold.
band (~72 kDa, mature CH) in the periplasmic fraction as well as in the CFE (Fig. 2B).This indicates the cleavage of 96-amino-acid-long N-terminus Tmd + Sp region during translocation of CH to the periplasm in strain CSV86-MEC1 (Fig. 2B).These results corroborate the role of Tmd + Sp in translocation of CH to the periplasm as proposed earlier in strain C5pp (42).The immunoreactive band was not detectable in the cytoplasmic fraction, probably due to low amount of protein, as reflected by low CH activity (see enzyme activity in Table 3).Regarding 1NH activity from CSV86-MEC1, intriguingly the periplasmic fraction (1,659 nmol•min −1 •mg −1 ) displayed an ~2.3-fold increase as compared to cytoplasmic fraction (715 nmol•min −1 •mg −1 ) and CFE (731 nmol•min −1 •mg −1 ; Table 3).Carbaryl-grown C5pp cells showed ~3.2-fold higher activity of 1NH in the periplasm (1,667 nmol•min −1 •mg −1 ) as compared to cytoplasmic fraction (516 nmol•min −1 •mg −1 ) and CFE (500 nmol•min −1 •mg −1 ; Table 3).Wild-type strain CSV86 T (negative control) displayed negligible activity of 1NH in all fractions, indicating lack of enzymes responsible for hydroxylation of 1-naphthol.SDS-PAGE analysis showed a distinct protein band (~66 kDa), corresponding to the  molecular weight of 1NH in all fractions of CSV86-MEC1.The absence of this protein band in wild-type CSV86 T suggests it to be 1NH (Fig. 2A).
The activity of C23DO from CSV86-MEC1 and wild-type CSV86 T as well as GDO from strain C5pp was found to be present only in the cytoplasmic fraction and CFE, as compared to the periplasmic fraction (Table 3).These results indicate that the periplas mic fractionation of the cells by cold-osmotic shock treatment was efficient, suggesting no leakage/contamination of cytoplasmic enzymes due to rupture of inner membrane.
These observations indicate that 1NH, previously reported to be cytoplasmic in strain C5pp (42), was also found to be localized in the periplasm, despite absence of a signal peptide.Aromatic hydroxylases (para-cresol methyl hydroxylase from P. putida NCIMB 9869 and 4-ethylphenol methylene hydroxylase from P. putida JD19), which lack a conventional signal peptide, have been reported to be localized in the periplasmic space (44,45).The periplasmic localization of 1NH is intriguing as the enzyme requires NAD(P)H cofactor to function, and the periplasm is an oxidizing environment.However, certain NADH-dependent enzymes such as NADH peroxidase (46), sulfite reductase (47), and glyceraldehyde-3-phosphate dehydrogenase (48) are also found to occur in the periplasmic space.The presence of periplasmic 1NH might play a role in scavenging 1-naphthol from the periplasmic space, mitigating its toxicity.
These results suggest the successful expression of CH and 1NH under Pnah in CSV86-MEC1, indicating the functionality of pMEC1.CH and 1NH were found to be present in the periplasm of CSV86-MEC1 and C5pp, and expression levels were similar in both strains.

Identification of an ORF encoding a putative outer membrane protein, McbT
An open reading frame of 897 bp (rc13993-14889 in Supercontig-A in Pseudomonas sp.strain C5pp [41]), referred to as mcbT, was identified upon reannotation and found to be located in between mcbA (encodes CH) and mcbB (encodes 1,2-dihydroxynaphtha lene dioxygenase) (Fig. 3Ai).In silico promoter prediction of upper operon indicated presence of multiple promoter elements (as indicated by linear discriminant function or LDF scores) (Fig. 3Ai).To validate in silico prediction, co-transcription analysis was performed using cDNA prepared from Carbaryl-grown C5pp cells as template and intergenic primers for two-, three-, or four-gene pairs (Table 2).The results are depicted in Fig. 3Aii and iii.The presence of amplicons of expected length indicates that the upper pathway genes (including mcbT) are co-transcribed as a polycistron.
The mcbT (897 bp) was translated in silico to 298-amino-acid-long polypeptide (referred as McbT) and showed sequence homology with characterized as well as uncharacterized outer membrane proteins involved in transport.BLAST 51)], respectively.N-terminus of McbT harbors a signal peptide (located from 1 to 24 amino acid) with prediction probability of 0.97 and a cleavage site between A24 and T25 position (Fig. S4A).The mature McbT was expected to be 274 amino acid long with a theoretical molecular weight of 29.3 kDa.The secondary structure analysis predicted it to possess 12 contiguous β-strands, a characteristic feature of bacterial outer membrane proteins (52).The three-dimensional structure model constructed for McbT using AlphaFold2.0 has 14 stranded β-barrel structure with N-terminal loop occluding the barrel lumen from the periplasmic side (Fig. S4B and S4C), a characteristic feature of bacterial outer membrane proteins involved in the uptake of hydrophobic compounds (51).The SDS-PAGE analysis of outer membrane fractions prepared from Carbaryl-grown cells of strain C5pp showed a distinct protein band (~30 kDa), corresponding to the molecular weight of mature McbT (upon cleavage of the N-terminus signal peptide).The absence of this protein band in cells grown on benzoate or glucose strongly indicates that it is likely to be McbT, induced by Carbaryl (Fig. 3B).
The lipid asymmetry of the outer membrane and hydrophilic channels of porins in gram-negative bacteria prevent the entry of hydrophobic compounds like Carbaryl.However, some uncharged aromatic compounds can still passively diffuse through the membrane (53).Specific outer membrane proteins facilitate the uptake of certain hydrophobic compounds, such as CymD for cymene, TodX for toluene, SphA for sphingosine, and TcpY for trichlorophenol.The genes encoding these outer membrane proteins were also found to be present within their respective degradation operons (49,50,54).In strain C5pp, mcbT is present in the upper Carbaryl pathway operon and is co-transcribed with neighboring genes as mcbFEDCBTA to encode outer membrane protein McbT only in the presence of Carbaryl.Thus, based on in silico, co-transcription, and expression analyses, we propose McbT to be involved in the transport of Carbaryl across the outer membrane in strain C5pp.In order to assess the impact of McbT on growth on Carbaryl, mcbT was cloned into pMEC1 (Pnah→CH-1NH), to generate construct pMEC2 (Pnah→CH-McbT-1NH; see Materials and Methods for details; Table 2; Fig. S2B).

Growth kinetics and metabolic analysis of engineered P. bharatica CSV86 T
The expression of CH and 1NH under Pnah (pMEC1) in strain CSV86 T (CSV86-MEC1) rendered it capable of utilizing Carbaryl as the sole source of carbon and energy (Fig. 4).In order to assess the impact of McbT on growth, growth kinetics of CSV86-MEC1 and CSV86-MEC2 on Carbaryl were compared.Pseudomonas sp.C5pp and wild-type CSV86 T were used as controls.The specific growth rate (μ, h −1 ) of CSV86-MEC1 on Carbaryl was 0.08 h −1 with a long lag phase (6-8 h).However, the growth was slower than strain C5pp on Carbaryl (μ = 0.27 h −1 ).Whereas, CSV86-MEC2 (CSV86 T carrying pMEC2), which encodes McbT, CH, and 1NH under Pnah, exhibited a growth profile with a shorter lag phase (4 h) and higher specific growth rate of 0.12 h −1 as compared to CSV86-MEC1 (Fig. 4).This can be attributed to the enhanced transport of Carbaryl in CSV86-MEC2, facilitated by the McbT transporter.P. bharatica CSV86 T (negative control) failed to utilize and grow on Carbaryl (tested till 48 h) as it lacks CH and 1NH (Table 3; Fig. 4).Pseudo monas sp.C5pp displayed faster growth (μ = 0.27 h −1 ) on Carbaryl than engineered strains of CSV86 T .The total biomass obtained for strain C5pp on Carbaryl (0.1% [wt/vol]) was comparable to CSV86-MEC1 and CSV86-MEC2 (1.6-1.7 g wet weight•L −1 ).The lower specific growth rate of CSV86-MEC1 or CSV86-MEC2 could probably be due to the absence of global adaptive responses, such as oxidative stress response, general stress response, changes in membrane-fatty acid composition, and overall energy regulation (55) required for the efficient metabolism of Carbaryl.The strain C5pp isolated from soil with the ability to degrade Carbaryl would have gone through evolution and adaptation steps/process to optimize the specific growth rate (7,41).CSV86-MEC2 exhibited good growth (OD 540 = 0.37 at 12 h) on 1-naphthol (0.01% [wt/vol]) in the presence of yeast extract (0.025% [wt/vol]) with characteristic olive-green color.Whereas, wild-type CSV86 T (negative control) failed to grow on 1-naphthol even in the presence of yeast extract, as it lacks 1NH.
The activities of CH and 1NH as well as C23DO (encoded by nahH gene of the sal operon from strain CSV86 T ; transcribed by Psal promoter) were monitored from Carbaryl-and glucose-grown cells of CSV86-MEC1 and CSV86-MEC2.These analyses were conducted to assess the induction of degradative pathway enzymes by Carbaryl and its metabolism through the salicylate-catechol route.Naphthalene-grown strain CSV86 T or  5).Thus, the upregulation/induction of CH and 1NH (expressed under Pnah) confirmed the metabolism of Carbaryl to 1,2-dihydroxynaphthalene via 1-naph thol, which is subsequently metabolized to salicylate.The Pnah promoter is reported to be activated by salicylate upon binding to the NahR transcriptional activator (38).
The induction of the naphthalene pathway (nah and sal operon) by Carbaryl was analyzed by measuring the cell respiration rates (in vivo) on various pathway intermedi ates using whole cells grown on single substrate, i.e., Carbaryl, naphthalene, or glucose as the sole carbon source (Table 4).Carbaryl-grown cells of CSV86-MEC2 showed oxygen consumption (9 nmol•min −1 •mg −1 ) on Carbaryl which was similar to Carbaryl-grown C5pp (~7 nmol•min −1 •mg −1 ).Wild-type CSV86 T showed negligible respiration on Carbaryl as it lacks CH and 1NH (Table 4).Interestingly, naphthalene-grown wild-type CSV86 T cells displayed oxygen uptake on 1-naphthol (~28 nmol•min −1 •mg −1 ), which could be probably due to naphthalene dioxygenase, reported to have a broad substrate specificity (15,33,37).Compared to naphthalene-grown wild-type CSV86 T or CSV86-MEC2, the oxygen consumption by Carbaryl-grown CSV86-MEC2 cells on naphthalene or catechol was low (~0.5-to0.8-fold, Table 4).However, the oxygen uptake rates on salicylate were comparable.The oxygen uptake rates of Carbaryl-grown CSV86-MEC2 on naphthalene, salicylate, and catechol were ~2.5-, 3.9-, and 3.7-fold higher, respectively, than glucosegrown cells of CSV86-MEC2 (Table 4).Pseudomonas sp.C5pp was the only strain which exhibited oxygen consumption on gentisate, as it serves as a metabolic intermediate in the Carbaryl degradation pathway of the strain.Furthermore, the strain C5pp exhibits the ability to utilize gentisate as sole carbon source (7).Whereas, strain CSV86 T (wild type or engineered) does not grow or show oxygen consumption on gentisate as it lacks gentisate degradation enzymes.Thus, Carbaryl could induce the host naphthalene pathway as the oxygen uptake rates on various metabolic intermediates were increased in Carbaryl-grown cells as compared to glucose-grown cells.Carbaryl-grown CSV86-MEC1 exhibited oxygen uptake rates similar to CSV86-MEC2 (Table 4).Thus, the expression of McbT, in addition to CH and 1NH, did not have an impact on cell respiration of the engineered strains.
The whole-cell biotransformation of Carbaryl by CSV86-MEC2 cells was carried out to identify the metabolic intermediates by thin layer chromatography (TLC).The spent medium analysis at various time points during biotransformation showed spots corresponding to 1-naphthol (R f = 0.75; reddish brown quench), salicylate (R f = 0.36; sky blue fluorescence), catechol (R f = 0.39; deep purple quench), and Carbaryl (R f = 0.60; dark blue quench) with maximum intensity at 6 h, which decreased majorly by 9 h (Fig. S5).The presence of 1-naphthol, salicylate, and catechol in the spent medium indicated that Carbaryl is metabolized through the salicylate-catechol route in CSV86-MEC2.
These metabolic studies (enzyme activity, whole-cell respiration, and biotransforma tion) indicate that the engineered CSV86 T metabolizes Carbaryl to 1,2-dihydroxynaph thalene via 1-naphthol (by CH and 1NH), which is further converted to catechol via salicylate.Generated catechol is funneled through meta ring-cleavage route to the central carbon pathway.

Carbaryl degradation phenotype is stable in P. bharatica CSV86 T
The Carbaryl degradation phenotype of CSV86-MEC2 (encoded by pMEC2) was found to be stable (96%-100%) in the absence of selection pressure (kanamycin or Carbaryl) for cells grown on lysogeny broth (LB) for 96 h or repeatedly sub-cultured for 15 transfers (equivalent to 75-90 generations).Furthermore, randomly selected colonies showed CH (~250-350 nmol•min −1 •mg −1 ) and 1NH (~800-1,000 nmol•min −1 •mg −1 ) activity, corroborat ing the stability data.Similar results were obtained for pMEC1.The loss of plasmid in the absence of selection pressure poses a major disadvantage for bioremediation and metabolic engineering applications.However, pMEC1 or pMEC2 encoding Carbaryl degradation enzymes was stable in strain CSV86 T in the absence of selection pressure, indicating suitability for bioremediation applications.

Engineered CSV86-MEC2 prefers Carbaryl over glucose
Wild-type strain CSV86 T metabolizes aromatic compounds preferentially over glucose and co-metabolizes them with organic acids, a unique property that makes strain extremely desirable for bioremediation (28)(29)(30).To assess the preferential utilization phenotype of CSV86-MEC2, growth kinetics and metabolic studies were performed on mixed carbon source, Carbaryl plus glucose or naphthalene plus glucose.
CSV86-MEC2 on naphthalene plus glucose (using naphthalene inoculum) displayed diauxic growth profile with utilization of naphthalene in the first exponential phase (Fig. S6).The observed growth profile and culture properties of CSV86-MEC2 were similar to that of wild-type CSV86 T (29), and the presence of the plasmid did not impact the preferential utilization phenotype.
On Carbaryl plus glucose, CSV86-MEC2 exhibited a growth profile with first lag phase of 6-8 h and second lag phase (small deflection, not so distinct) at 17-18 h (Fig. 6A).The observed growth rates were 0.20 h −1 for the first and 0.16 h −1 for the second log phase.The extracellular concentration of glucose was steady till 12 h, after which it decreased linearly till 22 h indicating glucose utilization, which coincides with the second log phase.On Carbaryl alone, CSV86-MEC2 showed slower growth with a specific growth rate of 0.11 h −1 .The presence of glucose during Carbaryl metabolism might aid in overcoming the toxicity of Carbaryl and 1-naphthol generated, resulting in better growth.Glucose might play a role in providing tolerance to oxidative stress during Carbaryl metabolism in CSV86-MEC2.The glucose metabolic enzyme Zwf has been implicated to play a major role in oxidative stress tolerance in Pseudomonas spp.by generation of NAD(P)H (57).Glucose has been reported to mitigate the toxicity and enhance degradation of various xenobiotics such as p-nitrophenol (58), 4-chlorophenol (59), 2,4-dichlorophenol (60), and 1-naphthol (61) in bacterial isolates.
To assess the carbon utilization hierarchy, the activities of C23DO (indicator of Carbaryl utilization via catechol) and Zwf (indicator of glucose utilization) from the CFE as well as whole-cell oxygen uptake on catechol and glucose were monitored from CSV86-MEC2 growing on Carbaryl plus glucose.The activity of C23DO was found to be maximum at 6-10 h (~1,100 nmol•min −1 •mg −1 , in the first log-phase) of diauxic growth profile (Fig. 6B).The activity of Zwf was observed to increase gradually reaching maximum at 16-18 h (~150 nmol•min −1 •mg −1 , in the second log phase), which correlates with the linear decrease in the extracellular concentration of glucose (Fig. 6B).During the initial phase of growth, cells exhibited gradual increase in the oxygen consumption on catechol, reaching maximum at 10-12 h (~16 nmol•min −1 •mg −1 ; Carbaryl utilization phase; Fig. 6C).Whereas, cells displayed maximum respiration on glucose at 16-18 h (~11 nmol•min −1 •mg −1 ; glucose utilization phase; Fig. 6C).The growth, enzyme activity, and whole-cell oxygen uptake profiles suggest Carbaryl utilization during 4-14 h followed by glucose utilization.
To assess the induction of Carbaryl/glucose metabolic genes in CSV86-MEC2 on Carbaryl plus glucose, qPCR analysis was performed at various time points.The foldchange of genes nahAa (naphthalene dioxygenase reductase subunit; part of nah operon; transcribed by Pnah) and nahG (salicylate-1-hydroxylase; part of sal operon; transcribed by Psal) was measured as indicators of Carbaryl metabolism.While foldchange of zwfA (glucose-6-phosphate dehydrogenase) was assessed as indicator of glucose metabolism.The qPCR analysis indicated that the Carbaryl degradation genes were induced maximally (~4-to 5-fold for nahAa and approximately ~2-to 3-fold for nahG) at 2-8 h (first log phase), whereas zwfA showed maximum induction at 12-20 h (~5-fold increase; second log phase; Fig. 6D) which corroborates with glucose consump tion, enzyme activity, and oxygen uptake analyses of the diauxic growth profile.
Carbaryl degradation metabolites from the spent media of CSV86-MEC2 while growing on Carbaryl plus glucose were extracted and analyzed by TLC and high-per formance liquid chromatography (HPLC) to assess carbon source utilization.TLC analysis showed metabolite spot corresponding to salicylate (R f = 0.29; sky blue fluorescence) as early as 4 h (Fig. S7).HPLC analysis showed metabolite peaks corresponding to salicylate and 1-naphthol at 4 h (Fig. S8).The presence of 1-naphthol and salicylate during the early (at 4 h) phase of the growth suggests that the presence of glucose does not suppress Carbaryl utilization in the engineered strain CSV86-MEC2.
On Carbaryl plus succinate, CSV86-MEC2 exhibited a monophasic growth profile, indicating co-metabolism (Fig. S9).Thus, succinate does not repress Carbaryl metabo lism in CSV86-MEC2.This is advantageous for bioremediation since in strain C5pp, the utilization of Carbaryl is strongly repressed by the presence of glucose or succinate (62).The observed carbon utilization hierarchy is similar to P. bharatica CSV86 T (29,30).

Conclusion
The co-expression of CH and 1NH resulted in funneling of Carbaryl into the native naphthalene pathway of P. bharatica CSV86 T (Fig. 7).Furthermore, both enzymes were found to be localized in the periplasm of the engineered strain (pathway compart mentalization), thus preventing the interaction of toxic 1-naphthol with cytoplasmic macromolecules.In addition to CH and 1NH, the expression of McbT (Fig. 7), the putative outer membrane Carbaryl transporter, resulted in increased growth rate of the engi neered strain on Carbaryl.However, the growth of strain C5pp was faster as compared to the engineered strain CSV86-MEC2.Over the course of evolution, strain C5pp would have acquired global adaptive responses to the presence of Carbaryl and its metabolites.Adaptive laboratory evolution can further be employed to optimize the growth of the engineered strain CSV86-MEC2 by subculturing on Carbaryl as the sole carbon source.The engineered strain CSV86-MEC2 degraded Carbaryl preferentially over glucose and co-metabolized it with organic acid, offering a major advantage over strain C5pp (which prefers glucose or succinate over Carbaryl) in bioremediation.Overall, meta bolic diversity of CSV86 T was successfully broadened to degrade naphthalene-based carbamate pesticide Carbaryl, in addition to naphthalene and methylnaphthalenes.Based on these results and stability experiments, we propose that P. bharatica CSV86 T is an ideal host for the engineering of various aromatic compound metabolic routes for efficient bioremediation.This genetically modified organism (GMO) can be used for the clean-up of Carbaryl-and/or 1-naphthol-contaminated soil or wastewater.However, various ethical/legal issues such as ecological disruption, occurrence of unwanted genetic crossover events/mutations, and horizontal gene transfer of antibiotic selection markers limit the application of GMOs for clean-up of contaminated niches.Therefore, in a controlled environment, a bioreactor-based ex situ bioremediation approach that ensures the containment and destruction of the GMO before release of the treated samples into the environment is ideal.Furthermore, immobilized engineered strains can be applied for the bioremediation of contaminated wastewater using construc ted wetlands, which can operate in a subsurface flow (vertical flow) or surface flow (horizontal) mode.

Co-transcription of Carbaryl upper pathway operon
Total RNA was isolated from Pseudomonas sp.C5pp cells grown till mid-log phase (OD 540 = 0.6) on MSM supplemented with Carbaryl using protocol provided with RNeasy Mini Kit (Qiagen, Germany).RNA was made DNA free by treating it with Ambion Turbo RNase-free DNase (Thermo, USA) at 37°C for 60 min.cDNA was synthesized using DNA-free total RNA as template and random hexamers using SuperScript III first-strand synthesis system as per manufacturer's instructions (Invitrogen, USA).The cDNA was used as template to perform co-transcription analysis of "Carbaryl upper pathway operon" genes using primers listed in Table 2. RNA preparation treated with RNase-H was used as template for PCR used in co-transcription analysis as negative control to assess genomic DNA contamination.

Quantitative PCR
The quantification of transcripts of nahAa (naphthalene dioxygenase reductase subunit; part of nah operon), nahG (salicylate-1-hydroxylase; part of sal operon), and zwfA (glucose-6-phosphate dehydrogenase) was performed by qPCR for CSV86-MEC2 cells grown on Carbaryl plus glucose at various time points.DNA-free RNA and cDNA were prepared as described above.qPCR was performed using cDNA as a template, genespecific primers (Table 2), and Platinum SYBR green qPCR SuperMix-UDG with ROX (Invitrogen, Thermo, USA) using StepOnePlus Real-Time PCR system (Applied Biosystems, USA).Primer efficiency was found to be 90%-110% for all genes.Melt curve analysis showed a single peak, while a single band of specific amplicon was detected on agarose gel electrophoresis, indicating amplification was gene specific.RNA preparation treated with RNase-H was used as template for PCR as negative control to assess genomic DNA contamination.The ΔCt value was calculated using rpoD (housekeeping gene encoding σ 70 transcription factor) as an internal reference.The ΔΔCt values were calculated with respect to the basal expression of these genes at 0 h (time of inoculation).Fold change in the gene expression was determined by the mathematical equation 2 −ΔΔCt (74).

Protein expression, cell-free extract preparation, and Western blot analysis
A single colony of transformants (CSV86-MEC1 or CSV86-MEC2) was inoculated into LB (5 mL) supplemented with kanamycin (40 µg•mL −1 ) and incubated at 30°C on a rotary shaker (200 rpm) for 16 h.Cultures were grown till mid-log phase (OD 540 = 0.8-1) on MSM (150 mL, pH 7.5) supplemented with naphthalene (12 h) or glucose (16 h) as the sole carbon source in the presence of kanamycin (40 µg•mL −1 ) using LB-grown culture (1 mL) as inoculum.Alternatively, cultures were grown on Carbaryl (150 mL, pH 6.8) using naphthalene-grown culture (2 mL) as inoculum.Strain CSV86 T or C5pp was grown on MSM supplemented with naphthalene (8 h) or Carbaryl (12 h), respectively.The cells were harvested by centrifugation at 7,000 × g for 10 min (4°C), washed twice, and resuspended (1 g cells in 5 mL) in buffer A (potassium phosphate buffer, 50 mM, pH 7.5).Cells were lysed on ice by sonication (14 W/15 pulses/1 s on, 1 s off/4 cycles) and centrifuged at 30,000 × g for 30 min (4°C) to obtain a clear supernatant referred to as cell-free extract, which was used for monitoring various enzyme activities.
Protein concentration in various fractions was determined by the method of Bradford (75) using BSA as the standard.Proteins present in various fractions were analyzed by SDS-PAGE 12%) (76).

Periplasmic and cytoplasmic fraction preparation
The mid-log phase (OD 540 = 0.8-1) naphthalene-grown cultures of CSV86-MEC1, CSV86 T and Carbaryl-grown culture of strain C5pp were used for the preparation of periplasmic and cytoplasmic fraction using the "cold osmotic shock" method (78).Briefly, cells were harvested by centrifugation at 7,000 × g for 10 min (4°C) and washed twice with buffer B (potassium phosphate, 20 mM, pH 7.5).The cells were re-suspended in buffer B (1 g cells in 5 mL) containing MgCl 2 (0.2 M) and incubated at 35°C for 10 min on shaking water bath followed by incubation on ice for 10 min.Cycle was repeated twice (total three cycles).The periplasmic fraction (supernatant) was obtained by centrifuging the cell suspension at 20,000 × g for 20 min at 4°C.The obtained cell pellet was washed twice with ice-cold buffer A, resuspended, and lysed by sonication (as described above).The cell homogenate was centrifuged (30,000 × g, 30 min, 4°C), and the obtained supernatant was referred to as "cytoplasmic fraction." The activities of various enzymes were monitored from these fractions.

Enzyme assays
Enzyme activities were monitored spectrophotometrically (Lambda-35, PerkinElmer).The activity of Carbaryl hydrolase was monitored by measuring the rate of appearance of 1-naphthol at 322 nm in buffer A as described previously (7).The activity of 1-naphthol 2-hydroxylase was monitored by measuring the rate of oxidation of NADH at 340 nm in buffer A as described previously (79).The activity of catechol-2,3-dioxygenase was monitored by measuring the rate of appearance of 2-hydroxymuconic semialdehyde at 375 nm in buffer A as described previously (29).The activity of C12DO was monitored by measuring the rate of appearance of cis,cis-muconic acid at 260 nm in buffer A as described previously (29).The activity of gentisate-1,2-dioxygenase was monitored by measuring the rate of appearance of maleylpyruvate at 330 nm in buffer A as described previously (7).The activity of glucose-6-phosphate dehydrogenase (Zwf ) was monitored by measuring the rate of reduction of NADP + at 340 nm in Tris-Cl buffer (50 mM, pH 7.5) as described previously (80).The total amount of protein used per assay reaction mixture was 25-40 µg.All enzyme activities are expressed as the specific activity in nanomoles of substrate consumed/product formed per minute per milligram of total protein present in the respective fraction (nmol•min −1 •mg −1 ).
The time-dependent spectral scan for CH reaction to monitor the conversion of Carbaryl to 1-naphthol was recorded spectrophotometrically from 240 to 340 nm at every 1 min interval for 10 cycles.One milliliter reaction mixture contained enzyme (25 µg), Carbaryl (200 µM), and buffer A.

Outer membrane protein fractionation
Carbon source-dependent expression of McbT from strain C5pp cells was analyzed by preparing the outer membrane fraction as described by Nakajima et al. (81) with minor modifications.Briefly, cells grown on MSM supplemented with Carbaryl, benzoate, or glucose till late log phase were harvested, washed twice, resuspended (1 g cells in 5 mL) in Tris-Cl buffer (10 mM, pH 8.0, containing 10 mM EDTA), and lysed by sonication.The cell homogenate was centrifuged at 10,000 × g for 10 min (4°C) to remove cell debris, followed by ultracentrifugation of supernatant at 50,000 × g for 60 min (4°C).The pellet was suspended in N-laurylsarcosine (1% in Tris-Cl buffer) and vortexed for 30 min (to dissolve the inner membrane) followed by centrifugation at 50, 000 × g for 60 min (4°C).The resulting membrane pellet which consists of outer membrane was suspended in minimum amount of Tris-Cl buffer and analyzed by SDS-PAGE.

Spent media analysis
CSV86-MEC2 culture (growing on Carbaryl plus glucose) was centrifuged (7,000 × g, 10 min, 4°C) to remove cells, and the supernatant collected was referred to as the spent media.The glucose concentration from the spent medium was estimated by the method of Miller (82) using 3,5-dinitrosalicylic acid reagent.The Carbaryl degradation metabolites were identified after extraction from spent medium in ethyl acetate and resolved by TLC as described above.The metabolites were also resolved and identified by high-performance liquid chromatography with photodiode array detector (HPLC; Jasco, LC-4000) using octadecylsilane silica (C18) reverse-phase column (dimensions: 4.6 × 150 mm).The metabolites were detected at 280 nm (λ max for Carbaryl), 302 nm (λ max for salicylate), and 322 nm (λ max for 1-naphthol).The solvent system used was acetonitrile:water:ortho-phosphoric acid (88%) (400:600:2, vol/vol/vol) (83) at a flow rate of 1 mL•min −1 .

Whole-cell oxygen uptake
The whole-cell oxygen uptake was monitored by measuring the rate of oxygen consumption in the presence of various pathway intermediates using Oxygraph (Hansatech, UK) fitted with Clark's oxygen electrode (29).The late log phase cultures of wild-type or engineered strains grown on various carbon sources were harvested by centrifugation (7,000 × g, 10 min, 4°C), washed twice, and resuspended in buffer A (200 mg cells in 1 mL).The reaction mixture (2 mL) contained cells (4 mg), substrate (50 µM), and buffer A. The respiration rates were measured at 30°C and expressed as nanomoles of O 2 consumed per minute per milligram of cells (nmol•min −1 •mg −1 ).

Plasmid stability
The stability of pMEC2 construct in strain CSV86 T (CSV86-MEC2) was assessed.CSV86-MEC2 was grown onto LB (5 mL) in the presence of selection pressure (kanamycin; 40 µg•mL −1 ) for 16 h at 30°C referred to as the zeroth transfer.This was transferred (100 µL) onto fresh LB (5 mL) in the absence of selection pressure (Carbaryl and kanamycin) for 12 h at 30°C and subsequently sub-cultured till 15 transfers (~5-6 generations per transfer) on LB with no selection pressure.In another set of experiment, CSV86-MEC2 was grown on LB in the absence of any selection pressure for 96 h at 30°C.In both the experiments, the retention of kanamycin resistance and Carbaryl degradation phenotype was assessed by replica plate method as well as by measuring CH and 1NH activities from three randomly selected kanamycin-resistant colonies from 0th, 5th, 10th, and 15th transfers or 96 h culture.

FIG 3
FIG 3 In silico analysis, co-transcription studies, and carbon-source-dependent expression of McbT in Pseudomonas sp.C5pp.(A) Gene arrangement, promoters, and co-transcription analysis of the Carbaryl upper pathway operon in Pseudomonas sp.strain C5pp.(i) Genes (name and length [bp]) are depicted in the orange box arrows.The newly annotated gene mcbT is indicated by blue box arrow.The regulatory gene mcbG is indicated by gray box arrow.The numbers in the parentheses indicate the intergenic distance in base pairs.The small green boxes depict putative promoters (with −10 box, −35 box, and LDF scores), whereas blue circles indicate ribosome-binding sites.Co-transcription primers are depicted in red arrows.Note: figure is not to the scale.Co-transcription analysis: (Continued on next page)

FIG 5
FIG 5 Specific activity of (A) Carbaryl hydrolase and (B) 1-naphthol 2-hydroxylase in CSV86-MEC1, CSV86-MEC2, and strain C5pp grown on minimal salt medium supplemented with glucose (open bar) or Carbaryl (dashed bar).The mean values of at least three independent experiments performed in triplicates with standard deviation are depicted.

FIG 6
FIG 6 Growth profile and metabolic properties of CSV86-MEC2 on double carbon source: (A) Growth profile on Carbaryl (0.1%) (■), or Carbaryl (0.1%) plus glucose (0.25%) (◯).The extracellular concentration of glucose from the medium is indicated by ♦. (B) Growth-dependent specific activity profile of catechol-2,3-dioxygenase and glucose-6-phosphate dehydrogenase (Zwf ) on Carbaryl plus glucose.The C23DO activity is indicated by dashed bars while Zwf by solid bars.(C) Growth-dependent whole-cell oxygen uptake profile on Carbaryl plus glucose.The oxygen uptake on catechol at various time points is indicated by dashed bars while on glucose by solid bars.(D) Growth-dependent quantitative real-time PCR profile of the genes nahAa, nahG, and zwfA on Carbaryl plus glucose.The relative fold change (2 −ΔΔCt ) for nahAa is indicated by black solid bars, nahG by dashed bars, and zwfA by open bars.Values are corrected for rpoD as internal reference.The mean value of at least three independent experiments performed in triplicates with standard deviation is depicted.

FIG 7
FIG 7 The expression of Carbaryl hydrolase, 1-naphthol 2-hydroxylase, and the putative outer membrane protein McbT (indicated in green) renders CSV86-MEC2 to utilize Carbaryl as the carbon source via salicylate-catechol route.

TABLE 1
List of microbes and engineered strains used in the study

TABLE 2
List of plasmids and primers used in the

Organism (grown on a ) Whole-cell oxygen uptake (nmol O 2 consumed min −1 mg −1 of cells) on metabolic intermediates
a Cells were grown on minimal salt medium containing respective aromatic compound or glucose as the sole carbon source.b Tr, trace; whole-cell oxygen consumption less than 0.5 nmol min −1 mg −1 is reported as trace.