The Metabolic Redox Regime of Pseudomonas putida Tunes Its Evolvability toward Novel Xenobiotic Substrates

Abstract

During evolution of biodegradation pathways for xenobiotic compounds involving Rieske nonheme iron oxygenases, the transition toward novel substrates is frequently associated with faulty reactions. Such events release reactive oxygen species (ROS), which are endowed with high mutagenic potential. In this study, we evaluated how the operation of the background metabolic network by an environmental bacterium may either foster or curtail the still-evolving pathway for 2,4-dinitrotoluene (2,4-DNT) catabolism. To this end, the genetically tractable strain Pseudomonas putida EM173 was implanted with the whole genetic complement necessary for the complete biodegradation of 2,4-DNT (recruited from the environmental isolate Burkholderia sp. R34). By using reporter technology and direct measurements of ROS formation, we observed that the engineered P. putida strain experienced oxidative stress when catabolizing the nitroaromatic substrate. However, the formation of ROS was neither translated into significant activation of the SOS response to DNA damage nor did it result in a mutagenic regime (unlike what has been observed in Burkholderia sp. R34, the original host of the pathway). To inspect whether the tolerance of P. putida to oxidative challenges could be traced to its characteristic reductive redox regime, we artificially altered the NAD(P)H pool by means of a water-forming, NADH-specific oxidase. Under the resulting low-NAD(P)H status, catabolism of 2,4-DNT triggered a conspicuous mutagenic and genomic diversification scenario. These results indicate that the background biochemical network of environmental bacteria ultimately determines the evolvability of metabolic pathways. Moreover, the data explain the efficacy of some bacteria (e.g., pseudomonads) to host and evolve with new catabolic routes.IMPORTANCE Some environmental bacteria evolve with new capacities for the aerobic biodegradation of chemical pollutants by adapting preexisting redox reactions to novel compounds. The process typically starts by cooption of enzymes from an available route to act on the chemical structure of the substrate-to-be. The critical bottleneck is generally the first biochemical step, and most of the selective pressure operates on reshaping the initial reaction. The interim uncoupling of the novel substrate to preexisting Rieske nonheme iron oxygenases usually results in formation of highly mutagenic ROS. In this work, we demonstrate that the background metabolic regime of the bacterium that hosts an evolving catabolic pathway (e.g., biodegradation of the xenobiotic 2,4-DNT) determines whether the cells either adopt a genetic diversification regime or a robust ROS-tolerant status. Furthermore, our results offer new perspectives to the rational design of efficient whole-cell biocatalysts, which are pursued in contemporary metabolic engineering.

Keywords: NADPH oxidases; Pseudomonas putida; biodegradation; dinitrotoluene; evolution; oxidative stress; reactive oxygen species.

FIG 1

Construction and phenotypic testing of a 2,4-DNT–degrading P. putida strain. (a) 2,4-DNT degradation pathway in Burkholderia sp. R34. The catabolic route begins with the action of DntA, a 2,4-DNT dioxygenase belonging to the Rieske nonheme iron family that hydroxylates the aromatic ring in positions 4 and 5 to yield 4M5NC, releasing the first nitro substituent. The substituted catechol is subsequently mono-oxygenated by DntB, a 4M5NC hydroxylase that eliminates the remaining nitro group in the structure, yielding 2H5MQ. The rest of the reactions in the pathway (executed by DntCDGE) include a ring cleavage reaction and channeling of the products toward central carbon metabolism. (b) Assembly of the genes encoding proteins for the whole route for 2,4-DNT degradation, along with the dntR-encoded regulatory protein, in a synthetic Tn7 transposon. The resulting plasmid, pTn7⋅DNT, was delivered at the defined attTn7 chromosomal site of the target host for stable insertion of the dnt gene cluster. (c) Qualitative testing of the recombinant P. putida EM⋅DNT strain, in which the dnt genes have been stably inserted in the chromosome of strain EM173. The appearance of a reddish color in cultures added with 2,4-DNT at 0.5 mM indicates presence of 2H5MQ. (d) Quantification of kinetic parameters in cultures of P. putida KT2440 (wild-type strain), EM173 (a reduced-genome derivative of strain KT2440), and EM⋅DNT (expressing the dnt genes) grown in the presence of 2,4-DNT at 0.5 mM. The specific rates of 2,4-DNT consumption (q_S) and formation of 4M5NC and 2H5MQ (q_P) were calculated by measuring the concentrations of the substrate and the products in culture supernatants. Bars represent mean values ± SD (n = 4) obtained after 24 h of incubation. Concn., concentration; CDW, cell dry weight.

FIG 2

Phenotypic and transcriptional stress response of P. putida EM⋅DNT exposed to 2,4-DNT. (a and b) Flow cytometry-assisted determination of ROS formation. (a) Histograms of raw data from untreated cells (Nil, added with DMSO, the 2,4-DNT solvent carrier) and cells exposed to 2,4-DNT at 0.25 or 0.5 mM. Cell suspensions were treated with the ROS-sensitive probe H₂DCF-DA, and the resulting dichlorofluorescein (DCF) fluorescence levels were recorded in at least 15,000 individual cells. Gray rectangles indicate the maximum fluorescence in cells without addition of H₂DCF-DA. A representative experiment per condition is shown in the diagram. (b) Fold change of the geometric mean (x mean) of DCF fluorescence levels for each experimental condition in P. putida EM173 (parental strain) and EM⋅DNT (carries the dnt gene cluster). Bars represent the average of x means ± SD (n = 6), and the asterisk identifies significant differences at a P level of <0.05 as determined indicated by the Mann-Whitney U test (comparing DCF fluorescence levels in cells exposed to 2,4-DNT with that in nontreated cultures). The dashed gray line indicates the basal level of DCF fluorescence in the control experiment for each strain. (c) Transcriptional activities of stress-responsive promoters. Two different oxidative stress reporters were constructed by placing the corresponding promoter (P_ROS) in a pBBR1-based, kanamycin-resistant (Km^r) vector bearing the promoterless gene encoding the monomeric and superfolder GFP (msf⋅GFP). The P_ROS → msf⋅GFP construct was transcriptionally insulated by means of the T0 and T1 terminators. Elements are not drawn to scale. The x mean of the msf⋅GFP fluorescence was detected by flow cytometry in P. putida EM⋅DNT exposed to 2,4-DNT at 0.5 mM. The resulting msf⋅GFP fluorescence was compared to that in cells harvested from cultures that were not treated with 2,4-DNT (Ctrl., baseline indicated with a dashed gray line). Bars represent the mean values of the x means of the msf⋅GFP fluorescence ± SD (n = 4), and the asterisks identify significant differences at a P level of <0.05 as determined using Student’s t test.

FIG 3

Effect of 2,4-DNT degradation on DNA recombination and SOS response. (a) Construction of a reporter P. putida strain to study DNA recombination. A suicide, integrative plasmid, obtained as detailed in Text S1 in the supplemental material, was used to disrupt the pyrF gene of P. putida, which results in uracil auxotrophy. The frequency of excision of the plasmid from the chromosome of P. putida EM⋅DNT⋅U (i.e., reversion to prototrophy) was adopted as an indication of DNA recombination. (b) DNA recombination frequency upon exposure to 2,4-DNT or norfloxacin (NOR). Bars represent mean values ± SD (n = 4), and the asterisks identify significant differences at a P level of <0.05, determined using Student’s t test. (c) General mechanism of SOS response upon DNA damage. The RecA protein, stimulated by either damaged or single-stranded DNA, triggers the inactivation of LexA (a repressor of the SOS response genes), thereby inducing the response. The LexA degradation-dependent activation of the recA promoter of P. putida was used as a proxy of the SOS response by constructing an msf⋅GFP-based biosensor. Elements in the outline are not drawn to scale. (d) Testing of the SOS response biosensor in soft agar experiments. P. putida EM⋅DNT was transformed with the reporter plasmid, and two filter paper disks, soaked with either NOR or 2,4-DNT, were applied onto the bacterial lawn. The plates were photographed under blue light after 24 h of incubation. (e) Quantification of the SOS response biosensor activity in cultures of P. putida EM⋅DNT in the presence of the additives indicated. Bars represent the mean values of the fold change in msf⋅GFP fluorescence ± SD (n = 4), and the asterisks identify significant differences at a P level of <0.05 as determined using Student’s t test. The baseline in cultures with no additives (Nil) is indicated with a dashed gray line.

FIG 4

Perturbation of the redox metabolism of P. putida. (a) Simplified scheme of the upper carbon metabolism of P. putida. Note that redox balance is maintained through the action of the EDEMP cycle, indicated in blue shading in the diagram. Abbreviations: ED pathway, Entner-Doudoroff pathway; EMP pathway: Embden-Meyerhof-Parnas pathway; PP pathway, pentose phosphate pathway; G6P, glucose-6-P; F6P, fructose-6-P; FBP, fructose-1,6-P₂; DHAP, dihydroxyacetone-P; GA3P, glyceraldehyde-3-P; 6PG, 6-phosphogluconate; KDPG, 2-keto-3-deoxy-6-phosphogluconate. (b) Reaction catalyzed by the water-forming Nox NADH oxidase from Streptococcus pneumoniae. (c) Construction of a synthetic NADH-burning device for tightly regulated expression of nox. The gene encoding Nox was placed under the control of the P_chnB promoter, which responds to the cyclohexanone-activated ChnR regulator. Elements in the outline are not drawn to scale. (d) Nox activity and impact of endogenous redox imbalance on the overall physiology of P. putida EM⋅DNT. The specific (Sp) in vitro Nox activity was compared in P. putida EM⋅DNT carrying either the empty pSEVA2311 vector (Ctrl.) or plasmid pS2311⋅Nox, with (Ind.) or without (Nil) addition of cyclohexanone at 0.1 mM to induce the expression of nox. The specific growth rate (μ) was determined in the same cultures. Each bar represents the mean value of the corresponding parameter ± SD (n = 5), and the asterisks identify significant differences at a P level of <0.05 as determined using Student’s t test.

FIG 5

DNA mutagenesis in redox-challenged P. putida EM⋅DNT cells exposed to 2,4-DNT. (a) Frequency of spontaneous rifampin-resistant mutants (M) in cells exposed to 2,4-DNT at 0.5 mM, DMSO (the 2,4-DNT solvent carrier), cyclohexanone at 0.1 mM (inducer of the synthetic device driving the expression of the Nox NADH oxidase), or combinations thereof. The dimensionless frequencies of mutation were calculated using the Ma-Sandri-Sarkar maximum likelihood estimator. Bars represent the mean M values ± SD (n = 4), and the asterisks identify significant differences at a P level of <0.05 as determined with Student’s t test. (b) Model for metabolism-driven evolution. Faulty redox reactions on novel substrates trigger ROS, which may cause direct or indirect damage to DNA in a fashion dependent on the background metabolism, which provides the reducing power necessary to fuel detoxifying enzymes (typically dependent on reduced glutathione [GSH]).

FIG 6

Evolutionary hormesis driven by metabolic formation of reactive oxygen species. (a) Interplay between redox stress and genetic and functional diversification of bacteria. As explained in the text, the action of ROS as a direct or indirect DNA mutagenesis agent is checked (and kept at bay) by the endogenous levels of reductive currency [i.e., availability of NAD(P)H]. This situation defines a scenario in which the gain of diversity occurs at the expense of population survival. (b) Species-specific definition of a productive diversification space. We argue that the evolvability of given bacterial species and the potential to host novel catabolic pathways is framed by the ability of bacteria to endure ROS-driven stress in a fashion that depends on their particular redox metabolism. These independent elements demarcate a space of productive diversification, i.e., an optimum in the novelty supply rate of the system (110), which fosters the exploration of the evolutionary solution space and which could be unique for each bacterial species.