A Type VI Secretion System Facilitates Fitness, Homeostasis, and Competitive Advantages for Environmental Adaptability and Efficient Nicotine Biodegradation

Mixtures of various pollutants and the coexistence of numerous species of organisms are usually found in adverse environments. Concerning biodegradation of nitrogen-heterocyclic contaminants, the scientific community has commonly focused on screening functional enzymes that transform pollutants into intermediates of attenuated toxicity or for primary metabolism. Here, we identified dual roles of the T6SS effector TseN in Pseudomonas sp. strain JY-Q, which is capable of degrading nicotine.

ABSTRACT

Gram-negative bacteria employ secretion systems to translocate proteinaceous effectors from the cytoplasm to the extracellular milieu, thus interacting with the surrounding environment or microniche. It is known that bacteria can benefit from the type VI secretion system (T6SS) by transporting ions to combat reactive oxygen species (ROS). Here, we report that T6SS activities conferred tolerance to nicotine-induced oxidative stress in Pseudomonas sp. strain JY-Q, a highly active nicotine degradation strain isolated from tobacco waste extract. AA098_13375 was identified to encode a dual-functional effector with antimicrobial and anti-ROS activities. Wild-type strain JY-Q grew better than the AA098_13375 deletion mutant in nicotine-containing medium by antagonizing increased intracellular ROS levels. It was, therefore, tentatively designated TseN (type VI secretion system effector for nicotine tolerance), homologs of which were observed to be broadly ubiquitous in Pseudomonas species. TseN was identified as a Tse6-like bacteriostatic toxin via monitoring intracellular NAD⁺. TseN presented potential antagonism against ROS to fine tune the heavy traffic of nicotine metabolism in strain JY-Q. It is feasible that the dynamic tuning of NAD⁺ driven by TseN could satisfy demands from nicotine degradation with less cytotoxicity. In this scenario, T6SS involves a fascinating accommodation cascade that prompts constitutive biotransformation of N-heterocyclic aromatics by improving bacterial robustness/growth. In summary, the T6SS in JY-Q mediated resistance to oxidative stress and promoted bacterial fitness via a contact-independent growth competitive advantage, in addition to the well-studied T6SS-dependent antimicrobial activities.

IMPORTANCE Mixtures of various pollutants and the coexistence of numerous species of organisms are usually found in adverse environments. Concerning biodegradation of nitrogen-heterocyclic contaminants, the scientific community has commonly focused on screening functional enzymes that transform pollutants into intermediates of attenuated toxicity or for primary metabolism. Here, we identified dual roles of the T6SS effector TseN in Pseudomonas sp. strain JY-Q, which is capable of degrading nicotine. The T6SS in strain JY-Q is able to deliver TseN to kill competitors and provide a growth advantage by a contact-independent pattern. TseN could monitor the intracellular NAD⁺ level by its hydrolase activity, causing cytotoxicity in competitive rivals but metabolic homeostasis on JY-Q. Moreover, JY-Q could be protected from TseN toxicity by the immunity protein TsiN. In conclusion, we found that TseN with cytotoxicity to bacterial competitors facilitated the nicotine tolerance of JY-Q. We therefore reveal a working model between T6SS and nicotine metabolism. This finding indicates that multiple diversified weapons have been evolved by bacteria for their growth and robustness.

INTRODUCTION

Microbial bioremediation is widely applied for the removal of contaminants such as polycyclic aromatic hydrocarbons (PAHs) and N-heterocyclic aromatics. Bacterial genomes encode a variety of membrane transport channels for the export of ions, chemicals, and proteins, such as efflux pumps for the extrusion of toxic substrates, which include virtually all classes of clinically relevant antibiotics, and secretion systems for the transport of proteinaceous effectors (1–5). Additionally, type VI secretion systems (T6SS) for the transport of metal ions and to regulate the production of reactive oxygen species (ROS) have been characterized in Yersinia pseudotuberculosis strain YPIII (6, 7), Escherichia coli O157:H7 strain EDL933 (8), Pseudomonas aeruginosa strain PAO1 (9), and Burkholderia thailandensis strain E264 (10, 11). The T6SS can deliver a vast array of effector proteins into target cells. Besides canonical contact-dependent antibacterial and antieukaryotic effectors (12–14), T6SS effectors also play important roles in bacterial fitness and in shaping gut population dynamics (7). For example, T6SS effectors have been shown to participate in the antagonism of ROS and the transport of zinc, manganese, and iron ions (7, 10, 11). These observations clarified that T6SS facilitates bacterial survival and adaptability, in addition to its fascinating antimicrobial activities to outcompete rivals.

Pseudomonas sp. strain JY-Q, which has been isolated from tobacco waste extracts, can completely degrade 5 g/liter of nicotine over a 24-h period and, remarkably, can survive in culture containing 10 g/liter of nicotine, which is notably greater than the inhibitory dose in other bacteria (15). Two candidate genetic determinants, termed Nic (consisting of two homologous gene clusters for the pyrrolidine pathway, namely, Nic-1 and Nic-2) and NA (nicotinic acid pathway), account for the capability of JY-Q to efficiently degrade nicotine (Fig. 1). However, nicotine and its biodegradation could induce excessive intracellular ROS (16–18). Bacteria have evolved several weapons to sense and combat external stresses. The efflux pumps SrpABC and TtgABC of Pseudomonas putida B6-2 play important roles in intermediate cellular detoxification in response to the degradation of PAHs, but the details are not fully elucidated (19). There may still exist other mechanisms for the bacteria to deal with toxic substrates. Our group previously reported that T6SS-1 transcription was significantly associated with the degradation of nicotine by JY-Q (15). Since nicotine degradation can yield sufficient ROS levels to repress bacterial growth and proliferation, we hypothesize that a T6SS effector(s) might antagonize the ROS stress induced by nicotine degradation and benefit the growth and fitness of JY-Q in nicotine-rich tobacco waste.

FIG 1.

Schematic presentation of T6SS and nicotine degradation gene clusters found in Pseudomonas putida S16 and Pseudomonas sp. strain JY-Q. (A) Schema for the nicotine degradation pathway. Note that intermediate chemicals for the tricarboxylic acid cycle (TCA) are indicated by black rectangles. Both Nic-1 and Nic-2 gene clusters were found to account for the pyrrolidine pathway of nicotine transformation. 3-Succinoylpyridine, 6-hydroxy-3-succinoylpyridine, and 2,5-dihydroxypyridine were detected as intermetabolites in this pyrrolidine pathway. (B) Cooccurrence of the type VI secretion system (T6SS) and Nic was only found in P. putida S16 and Pseudomonas sp. JY-Q. T6SS and nicotine metabolism gene clusters are flagged with solid triangles, while hollow schematics refer to gene clusters only present in strain JY-Q. NA, nicotinic acid-degrading gene cluster; Nic-1 and Nic-2 (HZN6), two homologous gene clusters responsible for nicotine degradation; T6SS, type VI secretion system. The strain S16 genome encodes one Nic and one T6SS (similar to T6SS-1 in JY-Q). (C) Reaction model for NicA2 to transform nicotine. NicA2 or Nox catalyzes nicotine to N-methylmyosmine, and then to pseudooxynicotine via simultaneous hydrolysis. (D) Functional modules encoded by the gene cluster of Nic-1, Nic-2, and NA. Functional module, Nic1 (red), nicotine to 3-succinoylpyridine; Spm (blue), 3-succinoylpyridine to 6-hydroxy-3-succinoylpyridine; Nic2 (black), 6-hydroxy-3-succinoylpyridine to 2,5-dihydroxypyridine and succinic acid and then to fumaric acid. nicXDFE genes pertaining to nicotinic acid degradation (NA) are responsible for sequential conversion of 2,5-dihydroxypyridine into TCA as well.

Here, we sought to systemically determine the roles of T6SS and its effectors in balancing JY-Q survival/fitness and nicotine degradation. A T6SS candidate effector, TseN, was predicted using bioinformatics and proteomics. Moreover, deletion mutations of the T6SS/TseN genes resulted in lower growth capabilities during nicotine degradation, as T6SS/TseN contribute to bacterial resistance to ROS and cellular homeostasis. These results suggest that the T6SS plays a positive role in nicotine degradation by maintaining sufficient growth/fitness of strain JY-Q. The results of the present study expanded current knowledge of the versatility of the ubiquitous T6SS for the degradation of N-heterocyclic aromatic chemicals.

RESULTS

Genetic organization of T6SS in strain JY-Q.

Three T6SS gene clusters (T6SS-1, T6SS-2, and T6SS-3) were found in the genome of Pseudomonas sp. JY-Q by T6SS-HMMER-based homology searches, the hidden Markov model (HMM) profiles of which were curated from experimentally characterized T6SS core genes (Fig. 2A). A total of 13 core components, TssA through TssM, were found in these three T6SS clusters, including TssB, a core component of the T6SS contractile machinery. The gene organizations of T6SS-1 and T6SS-2/3 are significantly different. T6SS-1 harbors a variable region (the open reading frames [ORFs] of which are indicated by gray arrows), compared to compact T6SS-2/3.

FIG 2.

Genetic organization and phylogeny of the T6SS in Pseudomonas sp. JY-Q, which is capable of degrading nicotine. (A) Schematic diagram of T6SS found in JY-Q. Almost identical T6SS-2 and T6SS-3 were found in the JY-Q genome, and a distinct T6SS-1 carrying a variable region was observed. The linear and circular genomic schemes are drawn to scale. (B) T6SS phylogeny was predicted based on representative TssB proteins using a maximum-likelihood approach. Representative T6SS gene clusters are indicated. Imp, Agrobacterium tumefaciens C58; H1-H3 T6SS, P. aeruginosa PAO1; Vas, Vibrio cholerae V52; Evp, Edwardsiella tarda PPD130/91; FPI, Francisella pathogenicity island; Sci-1, E. coli 042 (enteroaggregative E. coli [EAEC]). T6SS phylogeny subclades i (i1, i2, i3, i4a, i4b, and i5), ii (FPI), and iii (Bacteroides) are available at db-mml.sjtu.edu.cn/SecReT6 (13).

T6SS-1 in JY-Q is phylogenetically related to type i4b T6SS gene clusters. Type i4 and i5 T6SS representatives shown in Fig. 2B were demonstrated to facilitate bacterial competition and fitness demands and include Imp from Agrobacterium tumefaciens C58 (20, 21), Evp from Edwardsiella tarda PPD130/91 (22, 23), and H3-T6SS from Pseudomonas aeruginosa PAO1 (24). In contrast, T6SS-2 and T6SS-3 in JY-Q were phylogenetically related to those of group i1 (Fig. 2B). T6SSs of clade i1, such as H2-T6SS in strain PAO1, were commonly verified as antibacterial agents (25–27).

Nicotine concentration-dependent transcription of the T6SS-1.

To investigate the physiological roles of T6SS-1 and its effectors, the transcriptional levels of selected tssB1 (AA098_13300) and neighboring AA098_13375, coding for a putative bacterial polymorphic toxin similar to Tse6 characterized in P. aeruginosa PAO1 (28), were measured in response to nicotine versus (NH₄)₂SO₄ plus glucose conditions.

Strain JY-Q was cultured with different concentrations of nicotine and harvested at an optical density at 600 nm (OD₆₀₀) value of 0.6. Quantitative real-time PCR (qRT-PCR) analyses showed that the mRNA levels of T6SS-1 gene tssB1 (AA098_13300) were upregulated by 15-fold at 2 g/liter of nicotine and by 3.7-fold at 5 g/liter of nicotine compared to cultures supplemented with glucose plus (NH₄)₂SO₄ (Fig. 3A). T6SS-1 transcription was inhibited by the addition of 5 g/liter of nicotine compared to that in medium containing 2 g/liter of nicotine, but was still higher than that in response to glucose plus (NH₄)₂SO₄.

FIG 3.

T6SS-1 is involved in JY-Q survival and nicotine metabolism. (A) Induced expression of tssB (T6SS contractile component-encoding gene) under different nicotine concentration. (B) Induced expression of AA098_13375 (later named tseN) under different nicotine concentration. Nic indicates basic salt medium (BSM) plus nicotine, whereas Glc stands for BSM plus glucose and (NH₄)₂SO₄. Differences between two groups are significant unless indicated by NS (i.e., not significantly different between two groups). (C) The proliferation of the wild-type, tssB deletion mutant, and complementation strains under different concentration of nicotine. Complementation with tssB could promote mutant growth. (D) Total (left) and relative (right; per cell density) reactive oxygen species (ROS) levels were measured for wild-type, tssB deletion mutant, and tssB complementation strains. **, P < 0.01 (t test).

Moreover, AA098_13375, located in the variable region of T6SS-1, was predicted to encode a Tse6-like polymorphic toxin characterized in P. aeruginosa PAO1. qRT-PCR analyses also showed that the mRNA levels of AA098_13375 were also significantly increased by nicotine (Fig. 3B), suggesting that the expression levels of T6SS-1 and candidate effectors might be upregulated in response to nicotine metabolism in strain JY-Q. However, the expression levels of T6SS-2 and T6SS-3, both pertaining to phylogenetic group i1, were not induced significantly by nicotine. Therefore, the function of T6SS-1 was subsequently determined.

T6SS-1 is associated with nicotine tolerance by JY-Q.

To address whether T6SS-1 is important in stress antagonism, we compared the growth of the wild-type (WT) strain and a T6SS core tssB deletion strain in the presence of nicotine. The T6SS ΔtssB mutant strain, but not the tssB-complemented strain, showed impaired growth versus that of the wild type strain in the presence of nicotine, and the growth retardation expanded with the increase of nicotine concentration. Bacterial growth could be restored if the ΔtssB mutant was complemented with plasmid-carried tssB in culture supplemented with higher concentrations of nicotine, suggesting that T6SS-1 is responsible for nicotine tolerance (Fig. 3C).

To discern how the T6SS antagonizes stress resulting from nicotine, the chemical properties of nicotine were carefully considered. Nicotine is an alkaline toxicant that might induce multiple stress responses along with various environmental ROS levels and pH conditions. We hypothesize that T6SS-1 plays an important role in decreasing the level of ROS under culture with nicotine. Thus, we measured the ROS levels with 2′,7′-dichlorodihydrofluorescein diacetate (DCFH-DA) fluorescent dye in these three strains. The total and relative (per cell density) ROS levels of strain JY-Q as well as those of the T6SS mutant were compared in the presence of 2 g/liter of nicotine. The results showed that ROS levels were significantly higher in the tssB deletion mutant than those in the WT or the ΔtssB mutant complemented with tssB (Fig. 3D). Together, these results suggest that the T6SS can alleviate ROS production and enhance nicotine tolerance in order to facilitate bacterial survival and growth. Since T6SS-related functions have been largely determined by the activities and properties of related exported effectors, we sought to identify the effectors transported by T6SS-1 of strain JY-Q.

Characterization of the T6SS-1-deployed effector TseN via LC-MS/MS proteomics.

Mass spectrometry (MS)-based proteomics have been successfully used as a high-throughput approach to identify exported proteins. Thus, differences in supernatant proteins were compared between the WT and T6SS mutant strains challenged with nicotine in order to identify the T6SS effectors of Pseudomonas sp. JY-Q and to generate a hypothesis regarding the role of these proteins in bacterial survival and competition. Homology searches were therefore performed against all UniProt repertories for identified peptides using a liquid chromatography-tandem mass spectrometry (LC-MS/MS)-based proteomics platform (Table 1; details shown in Fig. S1 and Table S1 in the supplemental material). The peptide fragments of TssB1 and TssC1 (comprising the T6SS sheath) in the T6SS contractile transmembrane machinery were only detected in the supernatant of the WT strain but not in that of the ΔT6SS mutant strain, confirming that the system worked well. AA098_13375 and AA098_13325 were two potential T6SS effector candidates, since they were only present in the supernatant of the wild-type strain but not in that of the T6SS deletion mutant. The AA098_13375 N terminus codes for NAD⁺-dependent aldehyde dehydrogenase (ALDH-SF), which accounts for oxidization of a wide range of endogenous and exogenous aldehydes and plays an important role in detoxification, and the C-terminal Ntox46 domain acts as NAD(P)⁺ glycohydrolase (Fig. 4A). AA098_13325, whose product carries an ADP-ribosylglycohydrolase-related domain, was also identified in the T6SS-1 main cluster. Therefore, AA098_13375 (here TseN, for type VI secretion system effector for nicotine tolerance) was subsequently characterized on the basis of the following. (i) AA098_13370 was proposed as a cognate immunity protein, previously annotated as a hypothetical protein, and (ii) the tseN transcriptional level was increased by 2.5-fold compared to that under nonnicotine conditions (Fig. 3B). Thus, AA098_13370 is here renamed tsiN. Based on sequence homology between 13370 and Tsi6 (Tse6 antagonism), AA098_13370 was predicted as the cognate immunity protein of AA098_13375.

TABLE 1.

T6SS candidate effectors identified by LC-MS/MS-based comparative proteomics between the WT strain and the ΔT6SS mutant^c

Gene	Coordinates (bp) (strand)	Product	Domaina	Coverage (%)b	Score	Note
AA098_13300	2929989–2930528 (−)	T6SS component	COG3516	29.58	20.37	TssB
AA098_13375	2938691–2939794 (−)	Hypothetical protein	Ntox46; ALDH-SF	11.65	25.09	TseN
AA098_13295	2928472–2929974 (−)	T6SS component	COG3517	55.14	56.04	TssC
AA098_08220	1834241–1834426 (+)	UPF0434	YbaR	40.16	360.00
AA098_20710	4593928–4594206 (−)	Hypothetical protein	38.04	23.22
AA098_13325	2932791–2933906 (+)	ADP-ribosylglycohydrolase	ADP_ribosyl_GH	8.12	37.54
AA098_13240	2911822–2912340 (−)	Hemolysin coregulated protein	COG3157	29.65	637.59	Hcp-1
AA098_01190	266619–267107 (−)	ABC transporter	SBP_bac_3	29.01	158.16

^aDomain identified by CDD (http://ncbi.nlm.nih.gov/cdd/) and Pfam database (http://pfam.xfam.org).

^bMatching length ratio of captured peptides for UniProt archived sequences (www.uniprot.org).

^cUnder 2 g/liter nicotine induction (details shown in Fig. S1 and Table S1 in the supplemental material).

FIG 4.

Sequence assays for protein domains and phylogeny of TseN. (A) Domain organization for Tre1, Tse6, and TseN. Protein domains were predicted by HMMER 3.1 against the Pfam database (pfam.xfam.org). (B) Phylogenetic trees for TseN (left) and TsiN (right) predicted with the maximum-likelihood approach with 1,000 bootstrap replicates. Gene locus tags and strain names corresponding to the protein sequences are indicated. TseN and TsiN homologs were identified against Tse6/Tsi6 and TseN/TsiN prototypes. TsiN (navy blue), Tsi6 (blue), TseN (red), and Tse6 (green) were respectively labeled.

The phylogeny of TseN and its cognate immunity protein TsiN greatly coincide with known taxonomic groups (Fig. 4B). TseN and its highly related homologs were mainly identified in a subgroup of biodegraders for a series of pollutants (sequence similarities for the closet neighbor and Tse6 were 98% and 24%, respectively). However, homologs of less similarity [e.g., Tse6 as the NAD(P)⁺ glycohydrolase (Ntox46)] were only found in the clades of pathogens (Fig. 4B, left). A similar distribution pattern was observed for TsiN and Tsi6 (sequence similarity of 44%; Fig. 4B, right, navy and blue triangles), which were considered immunity determinants for TseN and Tse6 (indicated by the red and green triangles), respectively. In summary, these results suggest coevolution and functional linkages between AA098_13375 and AA098_13370 (TseN and TsiN).

T6SS-1 effector TseN has antibacterial activities.

An engineered pET-derived vector carrying tseN, tsiN, or both was individually transferred into Escherichia coli BL21(DE3) cells. Both TseN and TsiN proteins could be expressed in E. coli BL21 by isopropyl-β-d-thiogalactopyranoside (IPTG) induction (see Fig. S2 in the supplemental material). Overexpression of TseN, but not that of TsiN, caused growth inhibition of BL21(DE3) cells. When TsiN and TseN were coexpressed from the pETDuet-tseN-tsiN vector, the growth inhibition could be partly alleviated (Fig. 5A), suggesting that coexpressed TsiN was able to neutralize TseN as a defense mechanism.

FIG 5.

Antibacterial function of TseN and direct interaction between TseN and TsiN. (A) Growth profiles of E. coli BL21 with ectopic expression of tseN, tsiN, or both were examined by measuring optical density at 600 nm (OD₆₀₀) values at the indicated time points. As expected, tsiN overexpression did not inhibit E. coli growth. Inhibition driven by TseN was neutralized by supplementation with the flanking immunity determinant tsiN. Growth curves were measured based on at least three independent replicates. Error bars represent the standard deviation (SD) of three independent replicates. (B) Intraspecies competitive growth outcome of the indicated parental strain JY-Q and its derivatives as the donor (x axis) against prey strain ΔtseN ΔtsiN at 37°C for 15 h. The prey CFU result was acquired by using three strains as candidate competitors (wild type, ΔtseN, and ΔtseN + tseN). Statistical analysis was conducted for three independent replicates. ***, P < 0.001 (significant difference). (C) Direct binding of tagged TseN to TsiN was assessed using the pulldown assay. After ectopic expression and washing with phosphate-buffered saline (PBS), the GST-TsiN and His₆-TseN proteins as inputs were both identified by SDS-PAGE and Western blotting. His₆-TseN was retained by agarose beads coated with glutathione S-transferase (GST)-TsiN and detected by subsequent Western blotting.

Since TseN can be delivered by the T6SS, an intraspecies competition assay was performed to investigate the antibacterial activity of TseN. The JY-Q WT strain or ΔtseN derivative was employed as a donor strain (attacker), while JY-Q ΔtseN ΔtsiN served as the recipient strain. The inhibitory capability of the JY-Q WT strain was obviously stronger than that of the ΔtseN mutant. The survival rate of the prey strain was lower when incubated with tseN complementor than when coincubated with the ΔtseN mutant (Fig. 5B), further suggesting TseN cytotoxicity.

Direct interaction exists between TseN and TsiN.

We next sought to provide direct evidence that TsiN is the cognate immunity protein of TseN. To reveal whether TseN interacts with TsiN, a glutathione S-transferase (GST) pulldown assay was performed by incubating GST•Bind beads coated with GST-TsiN or GST with cell lysates and supernatants. After washing with phosphate-buffered saline (PBS), proteins retained on the beads were resolved by SDS-PAGE and then visualized by Western blotting. The specific interaction between TseN and TsiN was confirmed by an in vitro GST pulldown assay and binding examination with purified His₆-TseN and GST-TsiN proteins. As shown in Fig. 5C, His₆-TseN interacts with GST-TsiN (original blots available in Fig. S3 in the supplemental material). In summary, pulldown examination verified direct interplay between TseN and TsiN.

In recombinant E. coli BL21, no transport/extracellular signal sequence was genetically engineered into ectopic tseN. It is feasible that the target of TseN in the recipients is the cytoplasm, consistent with the predicted polymorphic toxin activities of TseN C-terminal NAD⁺ hydrolase (Ntox46 protein family in Table 1). On the other hand, TsiN could inhibit TseN activity by physical binding. Based on the above results, AA098_13375 and upstream AA098_13370 products constitute a pair of effector (TseN) and immunity (TsiN) proteins for JY-Q T6SS-1.

TseN is involved in JY-Q survival under ROS stress induced by nicotine metabolism.

To investigate the roles of the T6SS and its effector TseN in nicotine tolerance, the growth and nicotine degradation performances of the tseN mutant strain were compared with those of WT strain JY-Q, challenged with increasing nicotine concentrations of 2, 5, 10, and 15 g/liter. A growth assay of the WT and mutant strains cultivated in basic salt medium (BSM) supplemented with serial concentrations of nicotine was performed (Fig. 6A). A defect of TseN was detrimental to long-term bacterial growth/fitness in nicotine-containing medium, implying that TseN might be responsible for nicotine tolerance. In contrast, nicotine degradation was barely affected by the deletion of tseN (Fig. 6B). These results suggest that TseN prefers to facilitate JY-Q growth while degrading nicotine, in contrast to the heavy metabolic load for the ΔtseN mutant.

FIG 6.

Intracellular ROS levels associated with attenuated growth/fitness of strain JY-Q were significantly elevated in the T6SS-1 and tseN mutants in response to the addition of nicotine. (A) TseN is required for optimal growth of JY-Q in response to nicotine. (B) The degradation efficiency of tseN mutant was compared with that of wild-type JY-Q. (C) Total and relative (per cell unit) ROS levels of the WT and the JY-Q ΔtssB derivative, calculated based on at least three independent replicates. (D) Total and relative (per cell unit) ROS levels of the WT and the JY-Q ΔtseN derivative. ROS levels were measured by 2′,7′-dichlorodihydrofluorescein diacetate (DCFH-DA) methods. Error bars represent the SD of three independent replicates.

In addition, both degradation efficiency and cell growth between the WT and the ΔtseN mutant are obviously different under 10 g/liter nicotine culture. This observation further indicates that degradation capability is largely associated with cell viability, especially under high concentrations of nicotine. Next, we tried to address the mechanism of how TseN could protect cells from nicotine stress. Nicotine is an alkaloid pollutant that generates deleterious ROS, including highly destructive hydroxyl radicals (HRs) and alkyl radicals (ARs). Hence, subsequent assays were conducted to determine whether intracellular nicotine-induced oxidative stress was higher if tssB (coding for the T6SS core component) or tseN was deleted.

Although the total amount of the fluorescent signals varied, mutants lacking the effector gene tseN or the essential component tssB of T6SS-1 contained a significantly higher ROS content per cell unit in assays using the DCFH-DA ROS reporter. Moreover, relative ROS levels were obviously elevated in these mutants along with increasing amounts of nicotine (Fig. 6C and D). Remarkably, the nicotine-induced ROS fluorescent signals were specific in BSM plus nicotine, as no difference was detected between the control WT and mutant samples cultured in liquid broth medium. Based on this scenario, tseN was proposed to code for an effector with double effects in strain JY-Q, for antibacterial activities against competitors and population viability/robustness in response to oxidative stress.

TseN monitors NAD⁺ content to facilitate stable and economic biodegradation.

The cytotoxic properties and functional variability of TseN were subsequently examined. Sequence analysis suggested possible NADase (Ntox46 family domain in the C terminus) functionality of TseN, whereas the N-terminal domain of TseN is ALDH-SF (Fig. 4).

After the introduction and subsequent induction of tseN in E. coli strain BL21, NAD⁺/NADH amounts were compared to those in the noninduced group or a negative control of pET-nicA2 (nicotine dehydrogenase NicA2 from JY-Q) to exclude the impacts of protein expression. A significantly smaller amount of NAD⁺/NADH was observed with the expression of TseN (P < 0.001; two-way analysis of variance [ANOVA] test). Nicotine oxidization could be conducted by NicA2 by using NAD⁺ as its cofactor; however, this reaction is very slow (29), so it is suitable that NicA2 expression was chosen as a control group for TseN expression. The results suggest that the Ntox46 domain of TseN can rapidly remove NAD⁺ (Fig. 7A).

FIG 7.

TseN impacts intracellular NAD⁺ levels related to bacterial growth and nicotine catabolism. (A) tseN expression in E. coli BL21(DE3) contributed to the attenuated growth and the decreased amount of intracellular NAD⁺, indicating TseN as a Tse6-like bacteriostatic toxin. BL21 bearing empty vector and pET28-nicA2 encoding a nicotine dehydrogenase cloned from the JY-Q genome were set as an internal control. Notably, NicA2 could slowly consume nicotine by using NAD⁺ as its low-efficiency cofactor. (B) tseN overexpression facilitated better growth of the JY-Q derivative compared to that of the WT isolate and JY-Q bearing pHG101-BAD-tseN without 0.5 mg/ml arabinose induction. (C) Growth comparison among the wild-type and ΔtseN and ΔtseN-C_ter mutant strains by culture with 10 g/liter nicotine as their sole carbon/nitrogen source. (D) Residual content of nicotine for the wild-type and ΔtseN and ΔtseN-C_ter mutant strains after cultures of 1 day and 2 days.

The high-performance liquid chromatography (HPLC) results showed that the remaining nicotine content in the culture supernatant of the ΔtseN strain was slightly greater than that of the parental strain. Complementation or hyperexpression of tseN facilitated the growth of strain JY-Q in BSM supplemented with 10 g/liter of nicotine (Fig. 7B). Notably, basal constitutive expression of pHG101-BAD-tseN was observed. Moreover, the impacts of complementary TseN on bacterial propagation were improved by the induction of 0.5 mg/ml arabinose.

To further determine whether TseN C-terminal NAD⁺ hydrolase-like domain correlates nicotine metabolism by properly hydrolyzing NAD⁺, we deleted the C terminus of tseN to generate a JY-Q ΔtseN-C_ter mutant strain. We then detected the growth status and residual substrate of the WT and ΔtseN and ΔtseN-C_ter mutants by using BSM containing 10 g/liter nicotine. In summary, the WT strain grew better than the other two mutants, whereas no difference was observed between the ΔtseN and ΔtseN-C_ter mutants (Fig. 7C and D). This finding highlights the important role of the TseN C terminus in stress antagonism. Although the exclusive cofactor was still obscure for NicA2 in JY-Q, NAD⁺ could be considered an auxiliary synergist for NicA2 while confronted with stresses from nicotine (29). Based on this scenario, TseN could alleviate the metabolic burden in strain JY-Q by monitoring NAD⁺ level to accommodate ROS resulting from serial redox reactions of nicotine degradation (Fig. 8), since NAD⁺ has been identified as a candidate cofactor of NicA2 to transform nicotine.

FIG 8.

Proposed working patterns of T6SS-1 contributing to ROS resistance and bacterial survival. (A) T6SS-1 secrets proteinaceous effectors to outcompete enemies, including TseN in this study. (B) Intracellular TseN alleviates ROS stress triggered by the addition of nicotine, alongside NicA2 as a nicotine dehydrogenase (the degradation pathway shown in Fig. 1). A dynamic competitive binding model of TsiN or NAD⁺ to TseN is proposed that gradually releases cytoplasmic stress induced by nicotine metabolism.

Collectively, a novel role of TseN in pollutant removal was clearly demonstrated. In brief, JY-Q seems to have developed an appropriate and rational mechanism to assimilate nicotine by using TseN as an ROS mitigation agent to reduce the oxidative stresses induced by intracellular nicotine. The rapid accumulation of oxidative radicals contributed to bacterial growth deficiency and inhibited nicotine biotransformation. Bacterial survival was improved by fine-tuning nicotine oxidization with the elaborate assistance of T6SS-transferred TseN. Taken together, these results indicate that JY-Q T6SS-1 machinery and the auxiliary effector TseN are crucial to maintain JY-Q homeostasis and to neutralize ROS accumulation in response to stresses driven by nicotine metabolism.

DISCUSSION

In this study, the roles of T6SS-1 and its candidate effectors in nicotine degradation were systemically evaluated. In strain JY-Q, the addition of nicotine as the sole carbon/nitrogen source enhanced T6SS-1 expression, which in turn was responsible for antagonizing ROS. T6SS, with flexible transcription changes, could facilitate sequential resistance to oxidative stress in Escherichia, Burkholderia, and Pseudomonas species (7, 8, 11). These results indicate that the T6SS of strain JY-Q is a putative machinery to combat nicotine-induced oxidative stress. The additional roles of efflux pumps or secretion systems for the degradation of PAHs and other pollutants have been reported; however, the underlying mechanism of this interconnection has not been completely elucidated (15, 19). In addition, extensive interplay was found between bacterial secretion systems/transporters and ROS production or detoxification (30–33). For example, type III secretion systems in Vibrio parahaemolyticus (31) and P. aeruginosa (32), and the repertoire of effectors, have been experimentally shown to inhibit NADPH oxidase, which subsequently downregulates the generation of ROS.

Interestingly, the transcriptional level of tseN located in T6SS-1 of strain JY-Q was altered in response to the addition of nicotine. In the present study, secretion proteomics identified TseN as a T6SS-dependent effector prior to characterization of its additional role. Interestingly, the T6SS effector TseN plays a role in the killing of competitor cells, implying its contact-dependent antibacterial activities (2, 13, 34–36). Remarkably, TseN could act as a Tse6-like bacteriostatic toxin to prohibit cell growth. Next, in-frame deletion of T6SS-1 and tseN partially attenuated the resistance to oxidative stress, indicating plasticity of the effector repertoire for this highly conserved translocation machinery. TseN encoded by the nonconserved genomic region in T6SS-1 was determined to be related to bacterial fitness/proliferation in response to nicotine-induced oxidative stress, and such arms races of adaptability are commonly attributed to horizontally acquired traits (37, 38). Nicotine is notorious for its recalcitrance to environmental degradation, and it is classified as a toxic alkaloid that leads to the accumulation of ROS and induces intracellular oxidative stress (18, 39, 40). Likewise, ROS are thought to be overproduced, and the toxic by-products from PAHs are pumped out by Pseudomonas putida B6-2 when confronted with polycyclic aromatic hydrocarbons (19). However, TseN performs a completely unexpected role in shaping the composition of microbial populations by directly killing competitors and reducing metabolic load to ensure self-growth/proliferation.

T6SS engages a dual-effect TseN to outcompete rivals and overcome adversity. However, the rare presence of T6SS genes was observed in the genomes of effective nicotine degraders. The genetic cooccurrence of T6SS and nicotine degradation gene clusters was only clarified in Pseudomonas sp. JY-Q and P. putida strain S16 (Fig. 1). Interestingly, P. putida S16 and Pseudomonas sp. JY-Q were reported to be the strongest resistant biodegraders against nicotine (respectively characterized as 6 g/liter and 10 g/liter) (15, 18), indicating that the T6SS might be related to bacterial robustness under nicotine exposure conditions. Therefore, it needs to be clarified that the T6SS prompts bacterial viability and fitness to ensure degradation efficiency. Although decreasing detrimental ROS stress by TseN requires further quantification, the oxidative stress largely induced by nicotine might be alleviated or reduced by other apparatuses of nicotine-degrading strains (17, 41–43). In summary, the T6SS secrets TseN to maintain a suitable microniche for bacterial fitness and proliferation. These findings allowed us to tentatively propose a potential noncanonical role of the T6SS in bacterial biodegradation under unfavorable conditions (Fig. 8).

From our analysis, TseN could be evolved as a fusion product of the C-terminal domain of Tse6 and ALDH-SF. TsiN sufficiently conferred protection against toxicity from the TseN C-terminal domain to E. coli BL21 cells. An additional role of TseN prompts us to clarify in future studies the arms race safeguarding against adversity for strain JY-Q, which could assist in the adaptability of this strain to tobacco waste effluent (44–46). Note that the characterization of other T6SS-delivered effectors is necessary to sufficiently interpret the coordinated actions of the T6SS. Remarkably, AA098_13325 (ADP-ribosylhydrolase) might reverse promiscuous ADP-ribosylation for a variety of proteins. This phenomenon was first observed for the T6SS effector Tre1, which was neutralized by direct inhibition of Tri1 and functional antagonism with an ARH domain-containing protein in Serratia proteamaculans (47). In addition, the orthologue of Tse6 in strain PA14, Tas1 [a PAAR progenitor fused with C-terminal (p)ppApp-synthetase for Tas1 in contrast to with NAD⁺ glycohydrolase of Tse6 (28, 48, 49)], could result in the death of target cells driven by the rapid synthesis of (p)ppApp via depletion of ATP and its dysregulation of fundamental metabolic pathways. Interestingly, a Tsi6 homolog and the Tas1 cognate immunity protein Tis1 were both encoded immediately downstream of tas1. However, Tsi6 could not prohibit RelA-SpoT-like activities of Tas1 (50). Furthermore, a high-throughput screening strategy must be developed to elucidate the full repertoire of diverse effectors extracted from the full genome, such as predicting immunity proteins to discern their cognate effectors.

The prominent nicotine tolerance of strain JY-Q might result from the gentle and constitutive metabolism of nicotine and intermediate derivatives, contributing to lower cytotoxicity and fine-tuned metabolic load. Recently, we found that additional NAD⁺ could facilitate nicotine transformation by NicA2 or Nox, which are both nicotine dehydrogenases from JY-Q. However, this cofactor is low efficiency for recombinant NicA2 or Nox purified from E. coli, in contrast to prominent capability of strain JY-Q for nicotine degradation. It is feasible that another unknown factor could accelerate this reaction (29). In this cascade, nicotine could be scavenged by JY-Q as a nutrient material via hybrid transformation pathways of pyrrolidine and nicotinic acid (Fig. 1). TseN transcription is variable while encountering different content of nicotine. NAD⁺ content varied if confronted with TseN to fine tune the heavy metabolic flux driven by nicotine degradation. We speculated that the TseN-NAD⁺ complex underwent depolymerization in strain JY-Q due to possible competitive binding of TsiN to TseN. Moreover, TsiN could neutralize the TseN activity of hydrolyzing NAD⁺ in JY-Q. Lastly, NAD⁺ homeostasis driven by TseN might monitor nicotine catabolism to alleviate accumulated cytotoxicity and ROS stress driven by nicotine and its intermetabolites.

Altogether, an interesting manner of bacterial nutrient acquisition has been developed by the biodegrader strain JY-Q to improve fitness advantages and strengthen metabolic flux. The specific roles of T6SS in response to nicotine-driven toxicity or stress might be composite, although the significance of the T6SS and its effector TseN in strain JY-Q for bacterial survival/growth has emerged. The findings of this study provide insights into a new perspective on how bacterial cells maintain synergistic cooperation between intracellular homeostasis and pollutant degradation.

MATERIALS AND METHODS

Chemicals and media.

Nicotine and NAD (NAD⁺) were purchased from Fluka Chemie GmbH (Buchs, Switzerland). All chemicals were of analytical grade or better. Each liter of basic salt medium (BSM) contained Na₂HPO₄ (5.57 g), K₂SO₄ (1.0 g), KH₂PO₄ (2.44 g), MgCl₂·6 H₂O (0.2 g), CaCl₂ (0.001 g), FeCl₃·6 H₂O (0.001 g), MnCl₂·4 H₂O (0.0004 g), and an appropriate amount of nicotine. Solid plates were prepared with 2.0% (wt/vol) agar in liquid medium. Appropriate antibiotics were added when required.

Bacterial strains, plasmids, and culture conditions.

Pseudomonas sp. JY-Q (China Center for Type Culture Collection [CCTCC] no. M2013236), capable of efficiently degrading nicotine as the sole carbon/nitrogen substrate, was previously isolated by our laboratory. Unless specially stated, strain JY-Q was cultured in BSM supplemented with 2 g/liter of nicotine at 37°C and 180 rpm. E. coli strains were grown in Luria-Bertani (LB) medium supplemented with corresponding antibiotics or chemicals at 37°C and 180 rpm. The bacterial strains and plasmids used in this study are listed in Table 2.

TABLE 2.

Bacterial strains and plasmids used in this study

Strain or plasmid	Descriptiona	Reference or source
Strains
E. coli
WM3064	Donor strain for conjugation; ΔdapA; pK18mobsacB and its derivates for gene deletion	57
BL21(DE3)	Host for expression vector pET-28a(+)	Novagen
	Host for expression vector pETDuet	Novagen
	Host for expression vector pGEX	Novagen
DH5α	Donor strain for conjugation; pK18mobsacB and its derivates for gene deletion	Novagen
Pseudomonas sp. strain JY-Q
JY-Q WT	Wild-type Pseudomonas sp. JY-Q	15
JY-Q ΔtssB1	tssB1 gene deleted in JY-Q	This study
JY-Q ΔtseN	tseN gene deleted in JY-Q	This study
JY-Q WT (vector)	JY-Q wild type containing pBBR1MCS2; Kmr	This study
JY-Q ΔtssB1 (vector)	JY-Q ΔtssB1 containing pBBR1MCS2; Kmr	This study
JY-Q ΔtssB1 (tssB1)	JY-Q ΔtssB1 containing pBBR1MCS2-tssB1; Kmr	This study
JY-Q ΔtseN (vector)	JY-Q ΔtseN containing pBBR1MCS2; Kmr	This study
JY-Q ΔtseN (tseN)	JY-Q ΔtseN containing pBBR1MCS5-tseN; Gmr	This study
JY-Q ΔtseN (pHG-tseN)	JY-Q ΔtseN containing pHG101-BAD-tseN; Kmr	This study
JY-Q ΔtseN-Cter	JY-Q encoding TseN without C-terminal domain	This study
JY-Q ΔtseNΔtsiN (vector)	JY-Q ΔtseNΔtsiN containing pBBR1MCS2; Kmr	This study
Plasmids
pBBR1MCS2	Shuttle vector; Kmr	Lab collection
pBBR1MCS2-tssB1	pBBR1MCS2 carrying tssB1 coding region; Kmr	This study
pBBR1MCS5-tseN	pBBR1MCS5 carrying tseN coding region; Gmr	This study
pET-28a(+)	Expression vector with N-terminal hexahistidine affinity tag; Kmr	Novagen
pET-28a(+)-tseN/pET-28a(+)-75	pET-28a(+) carrying tseN coding region; Kmr	This study
pET-28a(+)-tsiN/pET-28a(+)-70	pET-28a(+) carrying tsiN coding region; Kmr	This study
pET-28a(+)-nicA2	pET-28a(+) carrying nicA2 coding region; Kmr	This study
pGEX	Expression vector with N-terminal GST affinity tag; Ampr	Novagen
pGEX-tsiN	pGEX carrying tsiN coding region; Ampr	This study
pETDuet	Coexpression vector; Ampr	Lab collection
pETDuet-tseN-tsiN/pETDuet-75-70	Coexpression vector carrying tseN and tsiN; Ampr	This study
pK18mobsacB	Suicide vector, mob, sacB; Kmr	Lab collection
pK18mobsacB-ΔtssB1	Construct used for in-frame deletion of tssB1; Kmr	This study
pK18mobsacB-ΔtseN	Construct used for in-frame deletion of tseN; Kmr	This study
pK18mobsacB-ΔtseN-Cter	Construct used for in-frame deletion of tseN-Cter; Kmr	This study
pHG101-BAD	Shuttle vector with the induced expression region of pBAD; Kmr	Lab collection
pHG101-BAD-tseN	pHG101-BAD carrying tseN coding region; Kmr	This study

^aKm, kanamycin; Gm, gentamycin; Amp, ampicillin; ^r, resistance.

Cell growth was measured by recording the optical density at 600 nm (OD₆₀₀) using a UV1000 spectrophotometer. OD₆₀₀ was measured to examine the growth status at the indicated time points. Thus, calibration curves between OD and cell amount were prepared to calculate CFU.

Quantitative real-time PCR.

Strains and mutants were cultured in BSM as the control and sample groups, respectively, for the comparison of gene transcriptional levels. Cultures were grown at 37°C until the mid-exponential phase (OD₆₀₀ of ∼0.6) and then centrifuged. About 10⁸ cells were harvested for RNA extraction with the use of the RNAprep pure cell/bacteria kit [Tiangen Biotech (Beijing) Co., Ltd., Beijing, China]. Complementary DNA (cDNA) was synthesized using EasyScript one-step genomic DNA (gDNA) removal and cDNA Synthesis SuperMix (TransGen Biotech Co.). Quantitative real-time PCR (qRT-PCR) was conducted using a CFX96 real-time PCR detection system (Bio-Rad Laboratories, Hercules, CA) with SYBR green real-time PCR mastermix (Tiangen Biotech Co.) and (diluted) cDNA as the amplification templates. The reaction systems were prepared with the corresponding primers, listed in Table 3. All qRT-PCRs were performed in triplicate, and relative expression levels were calculated using the threshold cycle (2^−ΔΔCT) method as described previously (51). The 16S rRNA gene was chosen as an internal reference. The amplification efficiency for all genes was about 100% (range, 95% to 105%).

TABLE 3.

Oligonucleotides used in this study

Primers	Sequence (5′–3′)a	Product length (bp)	Notes
For in-frame gene deletion
tssb-up-F	AATTCGAGCTCGGTACCCGGGGATCCCATCTCGGTGAACGAGTCCC	679	Knockout of tssB1
tssb-up-R	GCTCGCACTGGAGAACACG GCAGGAGCCCGAACATGAC
tssb-down-F	GTCATGTTCGGGCTCCTGCCGTGTTCTCCAGTGCGAGC	717	Knockout of tssB1
tssb-down-R	TAAAACGACGGCCAGTGCCAAGCTTGCAAACCCTGTTTGGCGAAT
13375-up-F	AATTCGAGCTCGGTACCCGGGGATCCTTTGCCGCGTTCAACAAGAG	516	Knockout of tseN
13375-up-R	AGGGCGATCAGTGTGGTGGCATTATGAAACCCACAACCCC
13375-down-F	GGGGTTGTGGGTTTCATAATGC CACCACACTGATCGCCCT	518	Knockout of tseN
13375-down-R	TAAAACGACGGCCAGTGCCAAGCTTATGCCTGCATCCAGGTTGAC
tseN-C-up-F	CAGGTCGACTCTAGAGGATCCCTACACTCAACTTCATGGCGGC	503	Knockout of segment coding for TseN C terminus
tseN-C-up-R	AGTTTCCCGCTGATAACACCAGGTGCATTATGAA
tseN-C-down-F	GGTGTTATCAGCGGGAAACTTCCGTTAAGCC	500	Knockout of segment coding for TseN C terminus
tseN-C-down-R	TATGACCATGATTACGAATTCTCCGCCATCGCCGAGTAT
13370-up-F	GAGCTCGGTACCCGGGGATCCCTTGCCTCTGGCTACGACAATG	539	Knockout of tsiN
13370-up-R	AGGTGCATTTTGCCGACGTTCGATCAGG
13370-down-F	AACGTCGGCAAAATGCACCTGGTGTTATCAGTTTG	474	Knockout of tsiN
13370-down-R	ACGACGGCCAGTGCCAAGCTTGAAGTTCGCTCGAACGGGTA
General PCR for cloning
pet-75-F	CAGCAAATGGGTCGCGGATCCCACAGGGCGATCAGTGTGGT	1,119	Expression of TseN
pet-75-R	CTCGAGTGCGGCCGCAAGCTTTCATAATGCACCTGGTGTTATCAGT
pet-70-F	CAGCAAATGGGTCGCGGATCCCCCACAGGGCGATCAGTGT	297	Expression of TsiN
pet-70-R	CTCGAGTGCGGCCGCAAGCTTTTCAATCGGGCAGTTGAATTTTG
duet-75-F	TCATCACCACAGCCAGGATCCGCACAGGGCGATCAGTGTGGT	1,119	Coexpression of TseN and TsiN
duet-75-R	TAAGCATTATGCGGCCGCAAGCTTTCATAATGCACCTGGTGTTATCAGT
duet-70-F	GGAGATATACATATGGCAGATCTCATGAAACCCACAACCCCC	297	Coexpression of TseN and TsiN
duet-70-R	CTTTACCAGACTCGAGGGTACCTCAATCGGGCAGTTGAATTTTG
pet-nicA2-F	CAGCAAATGGGTCGCGGATCCATGTATAACGACGGAAGCGT	1,491	Expression of NicA2 in BL21
pet-nicA2-R	CTCGAGTGCGGCCGCAAGCTTCTAGCTTAAGAGCTGCTTAACCTCC
pHG-75-F	CGATGGGGATCCGAGCTCGAGGTGGTGGCCGGGGCGTTC	1,146	Expression of TseN in JY-Q
pHG-75-R	TCCGCCAAAACAGCCAAGCTTTCATAATGCACCTGGTGTT
Primers for qRT-PCR
16S-F	CACACTGGAACTGAGACACG	124	Target: 16S rRNA
16S-R	TGCTTTACAATCCGAAGACC
q13375-F	AAATGGGGTAAGCGGGACAG	124	Target: tseN
q13375-R	ACGCGAGCTCTTCAAACTGA
q13300-F	TCGGCGTACTGGGTGATTTC	112	Target: tssB1
q13300-R	ATGCCCTTAAGCACACCGTT

^aUnderlined nucleotides indicate restriction cleavage sites.

DNA manipulation and mutant construction.

The suicide vector pK18mobsacB was prepared for gene knockout. Upstream and downstream (200 to 700 bp) fragments of the target gene were amplified using fusion/overlapping PCR with primers listed in Table 3. The resulting PCR fragments were inserted into the plasmid pK18mobsacB, which was then transferred from E. coli WM3064 cells (2,6-diaminopimelic acid [2,6-DAP] auxotroph) to strain JY-Q by conjugative transfer (52).

The donor strain E. coli W3064 and recipients were cultured overnight (mixed ratio, 3:1), centrifuged, washed with saline solution, and plated onto LB solid medium plus DAP. After 48 h, the mixture was suspended in saline solution and then transferred onto LB solid medium supplemented with kanamycin. The available transconjugants were examined by PCR, and the positive transformants were subsequently cultured in LB medium. It is of note that double-crossover in-frame deletions were counterselected on LB plates supplemented with 10% to 15% sucrose and 2.0 g/liter of nicotine. Finally, the cells were cultured in LB-sucrose agar medium for 2 days. Colonies of the correct gene knockout strains were confirmed by successive PCR amplifications.

For gene complementation, pBBR1MCS2 carrying the corresponding gene in E. coli DH5α was transferred to the mutant strain by triparental conjugation. DNA fragments of the tssB1 and tseN genes were amplified from total genomic DNA and respectively cloned onto the polylinker of pBBR1 (Table 3). Transformation by electroporation was conducted according to the given conditions, as follows: DNA-cell mixture (0.5 to 1 μg of DNA, 100 μl of electrocompetent cells), followed by electroporation at 12 kV · cm⁻¹, 200 Ω, and 25 μF with a Gene-Pulser Xcell (Bio-Rad Laboratories). For tseN ectopic expression in strain JY-Q, pHG101-BAD was designated a shuttle vehicle for the target gene by restriction cleavage and ligation (53).

T6SS effector genes and related immunity determinants were amplified from DNA templates with corresponding primers and ligated into the expression plasmid pET-28a(+), which contained BamHI and HindIII restriction sites. The resulting plasmid was confirmed by sequencing and then transformed into E. coli BL21(DE3) cells. It is of note that, for single protein expression of TseN and TsiN, the genes were amplified and ligated into plasmid pET-28a(+) with BamHI and HindIII restriction sites.

Moreover, antagonism of the T6SS effector and related immunity genes was assessed by coexpression on the plasmid pETDuet. pET-28a(+)-tseN, pET-28a(+)-tsiN, and pETDuet-tseN-tsiN were respectively electrotransporated into E. coli BL21(DE3) cells and then the recombinants were examined by PCR. These isolates were cultured in LB medium supplemented with corresponding antibiotics at 37°C. The primers used in this study are listed in Table 3. Then, growth was compared for these recombinants if the expression of TseN and/or TsiN was induced by IPTG.

Identification of T6SS effectors by liquid chromatography-tandem mass spectrometry.

Secretion assays for TseN (AA098_13375) and Hcp-1 were performed in accordance with previously described LC-MS/MS-based proteomics methods (8) with little modification, as follows. After appropriate growth to an OD₆₀₀ of 0.8 on BSM supplemented with 2 g/liter of nicotine by overnight inoculums of strain JY-Q and its derived tssB mutants, the culture was collected and subsequently diluted to 1:100 in 1 liter of BSM. After centrifugation at 20,000 × g for 15 min twice, the supernatant was harvested and passed through a 0.2-μm filter.

The filtered supernatant was then condensed 100-fold using the VivaFlow 50 system (Sartorius AG, Göttingen, Germany), centrifuged at 25,000 × g for 15 min, and ultrafiltered using the Amicon Pro purification system (cutoff, 10 kDa; EMD Millipore Corporation, Billerica, MA). Then, 100 μl of the resulting samples was mixed with 40 μl of 5× sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) sample buffer and boiled for 5 min. Last, the examined proteins were digested with 1 μg/μl trypsin at 37°C.

The protein content of the prepared samples was separated by SDS-PAGE and then subjected to LC-MS/MS analysis using a Q Exactive Ultra-High Mass Range Hybrid Quadrupole-Orbitrap mass spectrometer (Thermo Fisher Scientific, Waltham, MA). For comparative proteomics of the JY-Q and ΔtssB1 mutant supernatants, all survey spectra and database scans were processed using Mascot software v2.3.2 (Matrix Science, Ltd., London, England) against the self-customized JY-Q information with careful curation. The thresholds for further study were a false discovery rate of 0.1% and a minimum of two peptides per protein.

Protein expression and purification.

At an OD₆₀₀ of ∼0.4, the culture was cooled to 22°C, and then 1 mM isopropyl-β-d-thiogalactopyranoside (IPTG) was added to induce protein expression. After 16 h, the bacterial cells were centrifuged, washed, disrupted by sonication in Tris-HCl buffer (pH 7.5, 20 mM), and then centrifuged at 12,000 × g for 30 min in order to remove debris.

The acquired supernatant was filtered and loaded onto a Ni-nitrilotriacetic acid (NTA) His•Bind resin, which had been pre-equilibrated with binding buffer (20 mM Tris-base, 0.5 M NaCl, and 10 mM imidazole). The protein elution was prepared by adding elution buffer (20 mM Tris-base, 0.5 M NaCl, and 0.5 M imidazole). The eluted His₆-tagged proteins were desalted and then concentrated by ultrafiltration. The protein concentration resulting from this protocol was subsequently determined by the Bradford method. Finally, protein expression and purification were determined by SDS-PAGE. Likewise, protein expression of TseN using pHG101-BAD was induced by arabinose.

Pulldown examination.

The glutathione S-transferase (GST) pulldown assay was performed as previously described with minor modifications (11). A total of 150 μl of prewashed glutathione beads was mixed to prepare the reaction system. Briefly, 0.5 mg of harvested GST-TsiN fusion protein (negative control: pure GST tag) were mixed with cell lysates for 3 h at 4°C. Likewise, His₆-TseN/GST-TsiN complexes were captured by GST beads.

Putative protein binding was expected to occur within 2 h. Afterward, the beads were washed with TEN buffer (100 mM Tris-Cl [pH 8.0], 10 mM ethylenediaminetetraacetic acid, and 500 mM NaCl). Last, the acquired protein complexes were subjected to Western blotting analysis. After a 3-h incubation period, the beads were washed with phosphate-buffered saline (PBS). In brief, the reaction mixture was separated on a 15% nonreducing gel, stained with Coomassie blue, and subjected to Western blotting with antibodies against His and GST (EMD Millipore Corporation).

Western blotting.

Western blotting of His₆-tagged proteins was performed as previously described. In brief, protein samples were resolved by SDS-PAGE and then transferred onto polyvinylidene fluoride membranes (EMD Millipore Corporation), which were blocked with 5% (wt/vol) nonfat milk powder for 4 h. It is of note that this reaction system was prepared with antibodies against His₆ (dilution, 1:1,000; EMD Millipore Corporation) and GST (dilution, 1:1,000; Beijing Zhongshan Golden Bridge Biotechnology Co., Ltd., Beijing, China) at 4°C overnight. Finally, the immunoblots were visualized using an ECL Plus kit (GE Healthcare Bio-Sciences Corp., Piscataway, NJ) in compliance with the manufacturer’s protocol.

Intraspecies competition test.

Attacker and prey strains were both grown overnight, washed with double-distilled water (ddH₂O), and then coincubated at a 5:1 ratio supplemented with ddH₂O. Subsequently, the mixture was spotted onto a sterile 0.22-μm filter (EMD Millipore Corporation) on LB solid medium. Bacterial competition was conducted for 15 h at 37°C. For selective culture, the prey ΔtseN ΔtsiN strain contained a plasmid-borne kanamycin resistance gene. Each group of bacterial colonies was scraped off and harvested in 1 ml of liquid LB medium. After serial dilution and overnight culture, the CFU of the prey strains were counted on selective LB agar plates. Lastly, CFU changes to the prey were characterized. Growth status (mean ± standard deviation [SD]) was compared on at least three independent biological replicates. Statistical significance between different groups was inferred using GraphPad Prism 8.0.2 software.

Residual nicotine analytical methods.

To assess the nicotine degradation capability and tolerance of strain JY-Q and its mutants, the strains were inoculated onto BSM plus nicotine. The initial nicotine concentrations were 2, 5, 10, and 15 g/liter. The bacterial strains were grown in BSM supplemented with nicotine until the stationary phase. Next, cultures in 100-ml flasks were harvested and washed two times with 50 mM PBS (pH 7.0). Then, the pellets were resuspended in 20 ml PBS, incubated for 0.5 h, and centrifuged at 1,500 × g for 10 min.

The optical density/concentration of the growing cells (OD₆₀₀) and residual nicotine were determined every 2 to 4 h using a microplate reader and high-performance liquid chromatography (1200 series gradient system; Agilent Technologies, Inc., Santa Clara, CA), respectively, as reported previously (45, 52). In this study, three independent cultures were assessed in triplicate.

Intracellular ROS detection.

Fluorescent dye-based intracellular ROS detection was performed using 2′,7′-dichlorodihydrofluorescein diacetate (DCFH-DA; Invitrogen Corporation, Carlsbad, CA) as previously reported (8). Briefly, 1 ml of each sample was collected, resuspended in 1 ml of PBS containing 10 μM DCFH-DA, and subsequently incubated in the dark for 30 min. After removing the supernatant by centrifugation, the cells were pelleted and then resuspended in 1 ml of filter-sterilized PBS. Afterward, 200 μl of the resultant cell suspension were transferred to the wells of a prepared 96-well plate.

Relative fluorescence intensity was measured using a SpectraMax M2 microplate reader (Molecular Devices, Sunnyvale, CA) with predefined excitation/emission wavelengths of 488/525 nm (8). Each assay was performed at least in triplicate, and the results are presented as the mean ± standard deviation (SD). Statistical analysis was performed using Student’s t test.

Determination of the intracellular relative NAD⁺/NADH level.

E. coli BL21(DE3) cells bearing expression plasmids for TseN or NicA2 were grown in LB medium at 37°C and 180 rpm to the mid-log phase prior to induction of protein expression with 0.5 mM IPTG. At an OD₆₀₀ of 0.6, 200 μl of cells was harvested by microcentrifugation.

Samples were then subjected to NAD⁺/NADH analysis using a commercial kit (WST-8-based colorimetric approaches; Beyotime Biotechnology, China). A detection wavelength of 450 nm was used to quantitatively measure the NAD⁺/NADH pool. The residual amount of intracellular NAD⁺ was determined using a previously prepared standard curve.

Statistical analyses.

Growth/survival, intracellular ROS content, and gene transcriptional levels of at least three independent experimental replicates were assessed using a paired Student’s t test with Prism software (GraphPad Software, Inc., San Diego, CA). Statistical significance is indicated as follows: **, P < 0.01; ***, P < 0.001; NS, not significant.

Bioinformatics and phylogeny.

TssB proteins for phylogenetic tree calculation in Fig. 2 were retrieved from the SecReT6 (T6SS) database (13). Three T6SS gene clusters of strain JY-Q were predicted with the “T6SS-HMMER” subprogram in SecReT6. Adjacent 5-kb regions were examined for candidate effector-immunity genes using the Basic Local Alignment Search Tool (BLAST).

Identification was conducted using BLAST for proteins at an identity of 70% and coverage of 70% as cutoffs to exhaustively acquire Tse6/Tsi6 homologs. Lastly, candidate homologues of effector/immunity proteins were manually curated against 9,434 bacterial and archaeal genomes archived in the NCBI RefSeq database as of April 2018 (54).

Multiple sequence alignments of TssB, TseN, and TsiN protein sequences were generated with MUltiple Sequence Comparison by Log-Expectation (MUSCLE) software (55). Molecular Evolutionary Genetics Analysis version 7.0 software was used to construct maximum-likelihood phylogenetic trees of the TssB, TseN, and TsiN homologs with bootstrapping of 1,000 replicates and the Whelan and Goldman substitution model (56).

Data availability.

All supporting data involved in this study are included in Materials and Methods, and the genome sequence of Pseudomonas sp. strain JY-Q has been deposited in NCBI GenBank under accession number CP011525.