Biological functions of nirS in Pseudomonas aeruginosa ATCC 9027 under aerobic conditions

Abstract

Gang Zhou and Hong Peng contributed equally to this paper.

Electronic supplementary material

The online version of this article (10.1007/s10295-019-02232-z) contains supplementary material, which is available to authorized users.

Introduction

The opportunistic pathogen Pseudomonas aeruginosa is a gram-negative bacterium that is responsible for several acute and chronic infections [23, 24, 38]. The intrinsic high resistances of the organism to various antimicrobials or disinfectants make it difficult to control [12, 24, 26, 28], a trait that is attributed by several factors such as the low permeability of its cell wall, inhibitor inactivation, target site mutations, activation of multidrug efflux system, and biofilm formation [4, 19, 28]. In general, a bacterial biofilm is a surface-associated community that is distinct from their planktonic cells with respect to their phenotypical and biochemical properties [27]. The development of antimicrobial resistance plays a significant role in the infections of the urinary and pulmonary tract that are caused by the biofilms of P. aeruginosa [10]. Therefore, it has become exceedingly important to search for competent methods to eradicate the organism and its biofilms.

Denitrification is a sequential pathway including several reactions to reduce nitrate ion to dinitrogen gas under anaerobic conditions [48]. Among the process of denitrification, nitrite reductase (Nir) catalyzes the reduction of nitrite (NO₂⁻) to nitric oxide (NO) [1]. There are two distinct classes of Nir containing either copper (CuNir) or haem (cd  ₁Nir) as a cofactor in bacteria [32]. The nirS gene for cytochrome cd  ₁ nitrite reductase was firstly isolated from P. aeruginosa in 1989 with oligonucleotide probes and analysis of its protein sequence exhibited that this enzyme is fully water soluble and not membrane bound [37]. In recent years, it has been shown that anaerobic metabolism can occur inside P. aeruginosa biofilms grown under aerobic conditions [2]. Meanwhile, a P. aeruginosa mutant lacking the nitrite reductase (nirS) shows a thick biofilm without major dispersal events or cell death compared to the wild-type strain [2]. In addition, when cultured anaerobically, the mutant ΔnirS of P. aeruginosa PAO1 produced much lower catalase activity of KatA, which is the major catalase of this strain to detoxify H₂O₂ [40]. Up to now, most of the limited studies concerning nirS molecular functions were nearly performed under anaerobic conditions. However, it has been reported that denitrification can also be activated under aerobic conditions [15, 38]. As an important reductase in the denitrification, nirS may also play a vital role in the presence of oxygen.

Isothiazolones, such as benzisothiazolinone (BIT), are heterocyclic chemical compounds that are used to combat bacteria, fungi, viruses and algae in several industrial, agricultural and medical applications [29, 42]. It has been reported that the isothiazolone-resistant P. aeruginosa strains are frequently found in industrial spoilage [6]. Our previous study has demonstrated that the nirS gene from nitrogen metabolism pathway contributed to the resistance development of P. aeruginosa to isothiazolones [46]. However, the underlying molecular mechanisms remained elusive. In this study, the nirS gene of P. aeruginosa ATCC 9027 was deleted from the genome through homologous recombination, and the resulting phenotypical and transcriptional changes of the mutants were analyzed. We observed that the disruption of nirS resulted in an increase in the resistance of P. aeruginosa ATCC 9027 to BIT and a decrease of biofilm formation. Further data from RNA-seq revealed that the deletion of nirS gene led to several changes related to carbon metabolism pathways.

Materials and methods

Bacterial strains and chemicals

A wild-type strain of P. aeruginosa ATCC 9027 (WT) was purchased from the American Type Culture Collection (ATCC, Rockville, MD) and was routinely grown in Luria–Bertani (LB) liquid medium at 30 °C with 120 revolutions per min (rpm) in a shaking incubator. Escherichia coli S17-1, a kind gift from professor Hai-hong Wang of College of Life Sciences, South China Agricultural University, Guangzhou, China, was also cultured in LB medium but at 37 °C. All chemicals used in this study were of analytical grade and were purchased from Sigma Chemical Co. (St. Louis, MO, USA) unless otherwise stated.

Construction of knockout and complemented strains of nirS

The gene knockout and complementation of nirS in P. aeruginosa ATCC 9027 genome were conducted according to the previously reported methods with slight modifications [25]. Briefly, the total genomic DNA was extracted from P. aeruginosa ATCC 9027 using the EasyPure Genomic DNA Kit (Transgen Biotech, Beijing, China) according to the manufacturer’s instructions. To construct marker-less in-frame chromosomal deletions in P. aeruginosa ATCC 9027, the DNA fragments flanking nirS were amplified with the primer pairs nirS-up-F/R and nirS-down-F/R (Table 1) using the extracted genome as a template and were subsequently cloned into pK18-GM digested with the restriction enzymes BamHI and HindIII. The resulting plasmid of pK18-GM-nirS was used to transform E. coli S17-1, following which it was mobilized into P. aeruginosa ATCC 9027 via biparental mating. Exconjugants were selected on an LB medium containing ampicillin and gentamicin. Next, the candidate deletion mutants were recovered on LB medium containing 5% sucrose and were confirmed by PCR using the primers of nirS-QJ-F/R (Table 1). To construct the complemented strain, the full-length nirS gene with its upstream sequence was amplified from P. aeruginosa ATCC 9027 genome with primers nirS-Com-F/R (Table 1). The PCR product was cloned into the shuttle vector pSRK-GM between the NdeI and BamHI sites. The recombinant plasmid pSRK-GM-Com was transferred into E. coli S17-1 and subsequently mobilized into P. aeruginosa ATCC 9027 deletion strain of ΔnirS. Finally, the complemented strain was confirmed by PCR with the primers of nirS-BS-F/R and was designated as ΔnirS-com.

Bacterial growth assay

The strains of P. aeruginosa ATCC 9027 WT, ΔnirS and ΔnirS-com were pre-cultivated in LB medium overnight at 30 °C for their respective inoculums. The growth assay was conducted in a total volume of 200 μl in a Bioscreen C microtiter plate (Labsystems, Finland) using an initial bacterial density of approximately OD₆₀₀ = 0.05. The growths of each strain were monitored every 30 min by measuring the optical density at 600 nm, while they were incubated with continuous shaking for 24 h at 30 °C in a Bioscreen C analyzer (Labsystems) according to the manual instructions. The above assay was repeated at least three times with five replicates on different days.

Antimicrobial assays

The susceptibility test of WT, ΔnirS and ΔnirS-com planktonic cells and initial biofilms to BIT (Guangdong Dimei Biology Technology Co. Ltd., Guangzhou, China) was conducted in 96-well microplates (Corning Incorporated, Corning, NY, USA) according to the methods described previously [5, 36]. Briefly, twofold serial dilutions of the BIT suspension were transferred into 96-well plates at volumes of 100 μl/well, such that the final concentrations ranged from 4.0 to 64 mg/l. Following this, 100 μl each of WT, ΔnirS and ΔnirS-com suspensions at a final concentration of 2 × 10⁷ colony-forming units (CFU)/ml were added to the wells containing gradiently diluted BIT. After incubation at 30 °C for 24 h under static conditions, the bacterial growth of the planktonic cells was determined using a Multiskan GO reader (Thermo Scientific, Waltham, Massachusetts, USA) at an optical density of 600 nm (OD₆₀₀). The biofilms attached to the inner walls of the microplates were semi-quantified using 0.1% crystal violet (Chemical Reagents Co. Ltd., Shanghai, China) according to the previously established methods [39, 44]. The assay was conducted with eight replicates and repeated at least three times and the MIC was defined as the minimum amount of BIT that resulted in no visible growth after 24-h cultivation at 30 °C in a static condition.

Motility assay

The swimming motility assay was performed as described previously with minor modifications [14, 31]. Briefly, swimming plates containing 10 g/l tryptone, 5 g/l NaCl and 0.3% agar were inoculated in the center with 2-μl bacterial suspensions (OD₆₀₀ = 0.05) taken from an overnight culture grown in LB liquid medium at 30 °C. After 24-h incubation, the diameters of the colonies in each plate were measured with a ruler and the plates were additionally photographed using a digital camera (Nikon D7200, AF-S DX 18–140 mm; Nikon, Tokyo, Japan).

RNA-seq and analysis of differently expressed genes

Cultures of WT and ΔnirS grown in LB liquid media at 30 °C for about 24 h with constant shaking at 120 rpm were harvested at 4 °C. The total RNAs of the samples were isolated using the RNeasy Kit (Qiagen) and consecutively, the rRNAs were removed using the Ribo-Zero rRNA Removal Kit for Gram-negative bacteria (EPICENTRE Biotechnologies). The remaining RNA samples of WT and ΔnirS with A₂₆₀/A₂₈₀ ratio of 1.8–2.0 and A₂₆₀/A₂₃₀ ratio above 1.5, which were quantified with a Nanadrop ND-1000 spectrophotometer (Wilmington, DE, USA), were selected for subsequent analyses. A total of 20 μg of RNAs of both WT and ΔnirS were pooled and sent to Beijing Biomarker Technologies Co. Ltd. (Beijing, China) for cDNA library construction, followed by Illumina sequencing (HiSeqTM 2000, Illumina Inc., San Diego, CA, USA). The experiment was done with three biological replicates. The raw sequence reads were cleaned by removing adaptor sequences, empty and low-quality reads. And then, the obtained clean reads were mapped using TopHat 1.2.0 to the P. aeruginosa PAO1 reference genome (Accession No.: NC_002516.2), which can be downloaded from the Pseudomonas Genome Database (http://v2.pseudomonas.com). The differential expression analysis was carried out using the Bioconductor package EdgeR [33]. Differences with FDR ≤ 0.05 and log₂FC absolute value ≥ 1 were set as the threshold for significant differences in gene expression. Meanwhile, the Blast2GO program and the Kyoto Encyclopedia of Genes and Genomes pathway database (http://www.genome.jp/kegg/) were used for GO annotations (http://www.geneontology.org/) and pathway assignments of differently expressed genes (DEGs), respectively.

Quantitative real-time PCR (qRT-PCR) analysis

Although it has been reported that mexEF-oprN does not play any significant role in increasing the resistance in P. aeruginosa biofilms [9], we found an up-regulation of this type of multidrug efflux pump in ΔnirS, based on the RNA-seq results. Therefore, three differently expressed genes namely mexE, mexF and oprN in WT and ΔnirS were chosen for the determination of their expression levels in the absence or presence of BIT through qRT-PCR method, by adhering to previously described procedures [47]. Briefly, WT and ΔnirS were cultured in glass tubes with either LB medium or LB medium supplemented with 16 mg/l BIT at 30 °C with shaking at 120 rpm. After growing for 24 h, the bacterial suspensions from both conditions were harvested independently. Their total RNAs were isolated using RNAprep Pure Cell/Bacteria Kit (TIANGEN Biotechnology Co. Ltd., Beijing, China) and were treated with DNase I (Promega) to remove any residual DNA. Next, cDNA was synthesized using a Goldenstar™ RT6 cDNA Synthesis Kit (TSINGKE Biological Technology Co. Ltd., Beijing, China) according to the manufacturer’s instructions. The mRNA levels were measured using a 2-step RT-PCR with T5 Fast qPCR Mix SYBR Green I (TSINGKE) in QuantStudio 6 Flex Real-Time PCR System (Applied Biosystems), by following the manufacturer’s instructions. The relative mRNA levels and expression ratios of the selected genes were normalized to the expression of the 16S rRNA gene, and FCs were calculated using 2^−ΔΔCt method [22]. The primers used for all qRT-PCR analyses are listed in Table 1.

Detection of c-di-GMP

The method for detection of c-di-GMP from WT and ΔnirS was adapted from the previously reported procedures with minor modifications [34, 35]. Briefly, WT and ΔnirS were cultured at 30 °C for 24 h, harvested by centrifugation at 16,000g for 2 min, following which the cell pellets were washed with 1 ml of ice-cold PBS. Next, ice-cold ethanol was added to the cell pellet boiled in PBS. The samples were centrifuged and the supernatant containing the extracted c-di-GMP was retained in a new microfuge tube. The Waters Alliance e2695 (Waters, Milford, MA, US) was optimized and used to detect c-di-GMP. The separation was carried out using a reverse-phase C18 Targa column (4.6 × 250 mm; 5 μm) with a flow rate of 5 ml/min. Solvents A and B contained ammonium acetate (10 mM) dissolved in H₂O and methanol, respectively. The following gradient was used to elute c-di-GMP: 0–9 min, 95% A (= 95% solvent A and 5% solvent B); 9–14 min, 85% A; 14–19 min, 75% A; 19–26 min, 10% A; 26–40 min, 95% A. This gradient resulted in the elution of c-di-GMP in approximately 4–5 min. A standard curve was constructed using synthetic c-di-GMP purchased from Biolog (San Diego, CA). The c-di-GMP levels were normalized to total cellular protein levels by applying the calculation: Total c-di-GMP (pmol)/total protein (mg).

Statistical analysis

All the data obtained in this study were recorded as the mean ± standard deviation (SD) and were subjected to one-way ANOVA followed by a comparison of multiple treatment levels with the control using Fisher’s LSD test. All the statistical calculations were performed using a data processing system (DPS) software [41] and a p value < 0.05 was considered as significant.

Availability of data and material

Raw RNA-Seq data of both WT and ΔnirS were submitted and deposited in the NCBI Sequence Read Archive (SRA) under BioProject ID PRJNA482690. The other datasets supporting the conclusions of this article are included within the article and its additional files.

Results

Effect of gene knockout on the growth curve

According to the principles of homologous recombination, there were two rounds of gene exchange of nirS in the genome of P. aeruginosa ATCC 9027. When the mutant strains were identified through PCR with primers of nirS-QJ-F/R, there were one (2300 bp), two (2300 bp and 680 bp) and one (680 bp) bands from the strains of WT, single recombination strain, and ΔnirS, respectively (Fig. 1a). Sequencing the amplicons confirmed that they were identical to the expected fragments (data not shown). Meanwhile, the complementation strain of ΔnirS-com was constructed with the help of the plasmid pSRK-GM using ΔnirS as the receptive strain. Following this, the growth curves of P. aeruginosa ATCC 9027 WT, ΔnirS and ΔnirS-com under aerobic conditions were assayed using the Bioscreen C analyzer. The results illustrated that the deletion of nirS gene led to a significant increase in cell density and growth rates as displayed by ΔnirS when compared with the control strain of WT and ΔnirS-com under aerobic conditions (Fig. 1b).

Fig. 1

Identification of the gene knockout strain of ΔnirS by PCR (a) and the growth curves of P. aeruginosa WT, ΔnirS and ΔnirS-com across 24 h (b). According to the genome and the designed primers, the expected band numbers from WT, single recombination strain and ΔnirS following the PCR were one, two and one, respectively. The deletion strain of ΔnirS was selected to detect its phenotypic changes. The strains of WT, ΔnirS and ΔnirS-com were cultured in LB medium and the growth curves were measured using a Multiskan GO reader at 30 °C. All the experiments were conducted with eight replicates and repeated for at least three times

Open in new tabDownload slide

Deletion of nirS influences planktonic cell growth and biofilm formation

It was observed that with an increase in the concentration of BIT (especially at doses of 32 and 64 mg/l), the planktonic growths of WT, ΔnirS and ΔnirS-com were gradually inhibited in a dose-dependent manner under aerobic conditions, which also demonstrated that there are the same MICs (64 mg/l) for these three strains (Fig. 2). Meanwhile, there was a decrease in the formation of biofilms by ΔnirS at most of the tested concentrations of BIT, when compared with WT or ΔnirS-com (Fig. 2a–c). Taken together, these results suggest that both planktonic growth and formation of biofilms by WT, ΔnirS and ΔnirS-com could be affected by the increased concentrations of BIT, and the knockout of nirS gene further enhanced these inhibition effects.

Fig. 2

Bacterial growth (OD₆₀₀) and biofilm formation (OD₅₉₅) of WT (a), ΔnirS (b) and ΔnirS-com (c) in LB medium supplemented with different concentrations of BIT under aerobic conditions. All assays were performed in triplicates, and mean values and standard deviations (SD) are indicated. Values having different letters are significantly different from each other in each figure according to Fisher’s least significant difference test (p < 0.05)

Open in new tabDownload slide

nirS is required for swimming motility under the stress of BIT

Irrespective of WT, ΔnirS and ΔnirS-com, no differences were observed in the swimming motility of the cells under aerobic conditions in the absence of BIT (Table 2). However, in the presence of 16 mg/l BIT, ΔnirS displayed impairment in its swimming motility when compared with WT or ΔnirS-com under aerobic conditions (Table 2).

Absence of nirS affected the gene expression

To analyze the changes in gene expressions in the absence of nirS, the transcriptomes of ΔnirS and WT under aerobic conditions were sequenced using RNA-seq techniques, where WT served as a control. The results revealed that a total of 694 genes were differentially regulated in ΔnirS, of which 192 genes were activated, while 502 genes were repressed when compared to WT (FDR < 0.05 and |log2FC| > 1, Table S1).

Functional and KEGG analysis of DEGs based on RNA-seq data

The GO Ontology analysis that was inclusive of biological processes, molecular functions and cellular components, was used to classify the functions of the DEGs between WT and ΔnirS. Based on the sequence homology, the dominant groups for up-regulated DEGs in these three categories were revealed to be metabolic process, catalytic activity and cell, respectively (Fig. 3). For down-regulated DEGs, on the other hand, metabolic processes, catalytic activity, and membranes sub-groups were revealed in each of the three main categories, respectively (Fig. 3). In addition, DEGs were mapped to the terms in the KEGG database to elucidate the enriched signal transduction pathways. All the DEGs were divided into a total of 90 KEGG pathways and the top 20 from pathway enrichment are illustrated in Fig. 4. Notably, DEGs significantly enriched carbon metabolism together with glyoxylate and dicarboxylate metabolisms (p < 0.05, Q < 0.05; Fig. 4). The three central carbon metabolism cycles in P. aeruginosa are comprised of glycolysis, pentose phosphate pathway and tricarboxylic acid cycle. It was found that a majority of enzymes involved in TCA cycle (including PA0794, PA0854, PA1581, PA1582, PA1583 and PA5445; Table 3) were repressed in ΔnirS. On the contrary, several of the enzymes contributing to the pentose phosphate pathway (including PA2261, PA2263, PA2323, PA3181, PA3182, PA3183, and PA3194) and glycolysis (PA2323, PA3193, PA3195, PA5015, and PA5016) were found activated in ΔnirS (Table 3).

Fig. 3

Histogram representation of GO classification of differently expressed genes between P. aeruginosa WT and ΔnirS. The 694 differently expressed genes were categorized into three groups: Biological process, Molecular function and Cellular component. The x axis indicates the number of genes in each category. The three categories include 10, 9 and 7 functional groups, respectively. The predominant genes with up-regulation belonged to the functional groups of metabolic process, catalytic activity and cell for the categories of biological process, molecular function, and cellular component, respectively. Similarly, the predominantly down-regulated sub-groups in the three main categories are metabolic process, catalytic activity and membrane, respectively

Open in new tabDownload slide

 

Fig. 4

Classification of enriched KEGG pathways classification of differently expressed genes between P. aeruginosa ATCC 9027 WT and ΔnirS. Q value is the p value after multiple hypothesis tests. With value ranges from 0 to 1, a number closer to zero indicates significant enrichment. The graph is drawn with the top 20 pathways of Q value ranging from small to large. The rich factor represents the ratio of the number of enriched genes to the number of all background genes in a corresponding pathway. Higher the rich factor, greater is the degree of enrichment

Open in new tabDownload slide

 

Open in new tab

List of the altered genes of glycolysis pathway, Kreb’s Cycle and pentose phosphate pathway revealed by RNA-seq

Locus tag .	Log2 (FC) .	Gene name .	Description .
Glycolysis
PA0887	− 1.33	acsA	Acetyl-CoA synthetase
PA2323	2.51	–	Glyceraldehyde-3-phosphate dehydrogenase
PA3001	− 1.21	–	Glyceraldehyde-3-phosphate dehydrogenase
PA3193	1.27	glk	Glucokinase
PA3195	2.60	gapA	Glyceraldehyde 3-phosphate dehydrogenase
PA5015	1.28	aceE	Pyruvate dehydrogenase
PA5016	1.34	aceF	Dihydrolipoamide acetyltransferase
Kreb’s Cycle
PA0794	− 1.30	–	Aconitate hydratase
PA0854	− 1.05	fumC2	Fumarate hydratase
PA1581	− 1.13	sdhC	Succinate dehydrogenase subunit C
PA1582	− 1.39	sdhD	Succinate dehydrogenase subunit D
PA1583	− 1.05	sdhA	Succinate dehydrogenase flavoprotein subunit
PA5445	− 1.34	–	Coenzyme A transferase
PA5015	1.28	aceE	Pyruvate dehydrogenase
PA5016	1.34	aceF	Dihydrolipoamide acetyltransferase
Pentose phosphate pathway
PA1950	− 1.09	rbsK	Ribokinase
PA2261	1.99	–	2-Ketogluconate kinase
PA2263	1.97	–	Bifunctional glyoxylate/hydroxypyruvate reductase B
PA2323	2.51	–	Glyceraldehyde-3-phosphate dehydrogenase
PA3181	2.66	–	Keto-hydroxyglutarate-aldolase/keto-deoxy-phosphogluconate aldolase
PA3182	2.45	pgl	6-Phosphogluconolactonase
PA3183	1.74	zwf	Glucose-6-phosphate 1-dehydrogenase
PA3194	1.69	edd	Phosphogluconate dehydratase

Locus tag .	Log2 (FC) .	Gene name .	Description .
Glycolysis
PA0887	− 1.33	acsA	Acetyl-CoA synthetase
PA2323	2.51	–	Glyceraldehyde-3-phosphate dehydrogenase
PA3001	− 1.21	–	Glyceraldehyde-3-phosphate dehydrogenase
PA3193	1.27	glk	Glucokinase
PA3195	2.60	gapA	Glyceraldehyde 3-phosphate dehydrogenase
PA5015	1.28	aceE	Pyruvate dehydrogenase
PA5016	1.34	aceF	Dihydrolipoamide acetyltransferase
Kreb’s Cycle
PA0794	− 1.30	–	Aconitate hydratase
PA0854	− 1.05	fumC2	Fumarate hydratase
PA1581	− 1.13	sdhC	Succinate dehydrogenase subunit C
PA1582	− 1.39	sdhD	Succinate dehydrogenase subunit D
PA1583	− 1.05	sdhA	Succinate dehydrogenase flavoprotein subunit
PA5445	− 1.34	–	Coenzyme A transferase
PA5015	1.28	aceE	Pyruvate dehydrogenase
PA5016	1.34	aceF	Dihydrolipoamide acetyltransferase
Pentose phosphate pathway
PA1950	− 1.09	rbsK	Ribokinase
PA2261	1.99	–	2-Ketogluconate kinase
PA2263	1.97	–	Bifunctional glyoxylate/hydroxypyruvate reductase B
PA2323	2.51	–	Glyceraldehyde-3-phosphate dehydrogenase
PA3181	2.66	–	Keto-hydroxyglutarate-aldolase/keto-deoxy-phosphogluconate aldolase
PA3182	2.45	pgl	6-Phosphogluconolactonase
PA3183	1.74	zwf	Glucose-6-phosphate 1-dehydrogenase
PA3194	1.69	edd	Phosphogluconate dehydratase

Open in new tab

List of the altered genes of glycolysis pathway, Kreb’s Cycle and pentose phosphate pathway revealed by RNA-seq

Locus tag .	Log2 (FC) .	Gene name .	Description .
Glycolysis
PA0887	− 1.33	acsA	Acetyl-CoA synthetase
PA2323	2.51	–	Glyceraldehyde-3-phosphate dehydrogenase
PA3001	− 1.21	–	Glyceraldehyde-3-phosphate dehydrogenase
PA3193	1.27	glk	Glucokinase
PA3195	2.60	gapA	Glyceraldehyde 3-phosphate dehydrogenase
PA5015	1.28	aceE	Pyruvate dehydrogenase
PA5016	1.34	aceF	Dihydrolipoamide acetyltransferase
Kreb’s Cycle
PA0794	− 1.30	–	Aconitate hydratase
PA0854	− 1.05	fumC2	Fumarate hydratase
PA1581	− 1.13	sdhC	Succinate dehydrogenase subunit C
PA1582	− 1.39	sdhD	Succinate dehydrogenase subunit D
PA1583	− 1.05	sdhA	Succinate dehydrogenase flavoprotein subunit
PA5445	− 1.34	–	Coenzyme A transferase
PA5015	1.28	aceE	Pyruvate dehydrogenase
PA5016	1.34	aceF	Dihydrolipoamide acetyltransferase
Pentose phosphate pathway
PA1950	− 1.09	rbsK	Ribokinase
PA2261	1.99	–	2-Ketogluconate kinase
PA2263	1.97	–	Bifunctional glyoxylate/hydroxypyruvate reductase B
PA2323	2.51	–	Glyceraldehyde-3-phosphate dehydrogenase
PA3181	2.66	–	Keto-hydroxyglutarate-aldolase/keto-deoxy-phosphogluconate aldolase
PA3182	2.45	pgl	6-Phosphogluconolactonase
PA3183	1.74	zwf	Glucose-6-phosphate 1-dehydrogenase
PA3194	1.69	edd	Phosphogluconate dehydratase

Locus tag .	Log2 (FC) .	Gene name .	Description .
Glycolysis
PA0887	− 1.33	acsA	Acetyl-CoA synthetase
PA2323	2.51	–	Glyceraldehyde-3-phosphate dehydrogenase
PA3001	− 1.21	–	Glyceraldehyde-3-phosphate dehydrogenase
PA3193	1.27	glk	Glucokinase
PA3195	2.60	gapA	Glyceraldehyde 3-phosphate dehydrogenase
PA5015	1.28	aceE	Pyruvate dehydrogenase
PA5016	1.34	aceF	Dihydrolipoamide acetyltransferase
Kreb’s Cycle
PA0794	− 1.30	–	Aconitate hydratase
PA0854	− 1.05	fumC2	Fumarate hydratase
PA1581	− 1.13	sdhC	Succinate dehydrogenase subunit C
PA1582	− 1.39	sdhD	Succinate dehydrogenase subunit D
PA1583	− 1.05	sdhA	Succinate dehydrogenase flavoprotein subunit
PA5445	− 1.34	–	Coenzyme A transferase
PA5015	1.28	aceE	Pyruvate dehydrogenase
PA5016	1.34	aceF	Dihydrolipoamide acetyltransferase
Pentose phosphate pathway
PA1950	− 1.09	rbsK	Ribokinase
PA2261	1.99	–	2-Ketogluconate kinase
PA2263	1.97	–	Bifunctional glyoxylate/hydroxypyruvate reductase B
PA2323	2.51	–	Glyceraldehyde-3-phosphate dehydrogenase
PA3181	2.66	–	Keto-hydroxyglutarate-aldolase/keto-deoxy-phosphogluconate aldolase
PA3182	2.45	pgl	6-Phosphogluconolactonase
PA3183	1.74	zwf	Glucose-6-phosphate 1-dehydrogenase
PA3194	1.69	edd	Phosphogluconate dehydratase

Induction of mexEF-oprN efflux operon genes by BIT

The RNA-Seq results revealed that genes of mexEF-oprN efflux operon-mexE, mexF, oprN were up-regulated in ΔnirS. To verify the results and construct a possible relationship between the three genes and BIT resistance, their relative expression levels under normal and BIT stressed conditions were determined using the technique of qRT-PCR. The results displayed an up-regulation in the expression of all three genes-mexE, mexF, and oprN of ΔnirS under the detected conditions, especially in the presence of 16 mg/l BIT (Fig. 5a, b). However, the transcription of these three genes can be inhibited by BIT in the WT and ΔnirS-com (Fig. 5b). The above observations not only verify the reliability of the RNA-Seq results in this study but also confirm that mexE, mexF, and oprN are the response sites of BIT in P. aeruginosa ATCC 9027 and the deletion of nirS induces the activation of the above three genes against BIT.

Fig. 5

The relative expression levels of selected genes in the strains of WT and ΔnirS under normal (a) and 16 mg/l BIT stressed conditions (b). The relative transcription levels of all genes in WT cells treated without any BIT were considered as 1 when calculated with the 2^−ΔΔCt method. Error bars represent the standard deviation of triplicate measurements

Open in new tabDownload slide

Deletion of nirS leads to declined production of c-di-GMP

HPLC results showed that the concentration of c-di-GMP in ΔnirS cells was lower than that in WT (p < 0.05), suggesting that the nirS gene in P. aeruginosa is involved in the production of c-di-GMP (Fig. 6). Moreover, the decrease in the level of c-di-GMP in both WT and ΔnirS in the presence of 16 mg/l BIT emphasizes the inhibitory effect of BIT on c-di-GMP production (Fig. 6).

Fig. 6

Change in c-di-GMP concentration in the planktonic cells of WT and ΔnirS and ΔnirS-com. Bacteria were grown in LB media overnight supplemented with or without 16 mg/l BIT. Each sample was prepared in biological duplicates and measured in two or three experiments. Error bars represent the standard deviation of triplicate measurements. Values having different letters are significantly different from each other in each figure according to Fisher’s least significant difference test (p < 0.05)

Open in new tabDownload slide

Discussion

Denitrification is a microbially facilitated process where nitrate is reduced for the ultimate production of molecular nitrogen through a series of intermediate gaseous nitrogen oxide products [1, 48]. Several studies have reported that denitrification can also be activated under aerobic conditions. Using quadrupole membrane-inlet mass spectrometry, it is was found that the concentrations of both N₂ and N₂O increased with the introduction of O₂ into the gas phase in the presence of P. aeruginosa, suggesting that aerobic denitrification had occurred [8]. Further, Pseudomonas stutzeri SU2 was found capable of rapidly reducing nitrate to nitrogen gas without the accumulation of nitrite, when incubated in a nitrate-supplemented basal medium in an airtight crimp-sealed serum bottle containing a mixture of 92% oxygen [15]. In this study, the phenotypic changes in ΔnirS arising due to the successful deletion of nirS from the genome of P. aeruginosa ATCC 9027 (Fig. 1a) were compared with the WT and ΔnirS-com under aerobic conditions.

Isothiazolones have been found to inhibit biofilms. On a laboratory-scale rotating biological contactors (RBCs), 15 mg/l isothiazolones (Kathon) was seen to inhibit all microbial activity and resulted in the death of the biofilms [20]. However, with respect to Pseudomonas putida HB45, the mixed cultures and P. aeruginosa, higher concentrations of isothiazolone (625 and 1250 mg/l) were needed to reduce the biofilm formations [13]. Although the isothiazolones of BIT stimulated the initial biofilm formation at concentrations of 8 and 16 mg/l, it displayed inhibitory effects on both biofilm formation and planktonic growth of Enterobacter cloacae BF-17 at the concentrations of more than 32 mg/l [45]. In this study, however, the diminished construction of biofilms by ΔnirS in all tested concentrations of BIT indicates that different bacterial species exhibited different responses to BIT and nirS gene is an important but not essential factor for biofilm formation of P. aeruginosa ATCC 9027 (Fig. 2).

A P. aeruginosa PAO1 mutant lacking the sole enzyme capable of generating metabolic NO, encoded by the gene nirS exhibited normal biofilm development with no major changes in dispersal events when compared with WT, despite cultivation for 2 or 6 days, suggesting that nirS gene was not related to either the formation or dispersal of biofilms under anaerobic conditions [2]. In contrast, we found a decrease in the capacity of biofilm formation by ΔnirS under aerobic conditions without any BIT when compared with WT and ΔnirS-com (Fig. 2). The deletion of nirS gene led to down-regulation of the gene pslI involved with biofilm formation (Table S1), which possibly explains the reduction of biofilm formation observed in ΔnirS. In addition, the repressed growth rates of ΔnirS (Fig. 1b) suggest that nirS is an important effect factor for the formation of biofilms and growth of P. aeruginosa. In recent years, it has been proved that the secondary messenger c-di-GMP is a key player in the regulation of biofilm formation [30]. While high concentrations of c-di-GMP promoted the sessile lifestyle and biofilm formation, lower concentrations led to biofilm dispersal and favored the planktonic lifestyle [3]. In this study, we found that the deletion of nirS gene and the addition of BIT inhibited the production of c-di-GMP in aerobic conditions (Fig. 6). The change in trends of WT and ΔnirS for c-di-GMP production under the indicated conditions was consistent with the biofilm formations of these two strains (Fig. 2), implying that BIT inhibited the biofilm formation by constraining the production of c-di-GMP.

Pseudomonas aeruginosa is known for its intrinsic resistance to a variety of antimicrobial agents and its ability to develop multidrug resistance (MDR) through mechanisms such as mexEF-oprN [16]. It has been reported that the mexEF-oprN efflux operon endows resistance or tolerance to organic solvents [11, 21], and biocides such as triclosan [7], fluoroquinolones [16] and trimethoprim [17]. In this study, RNA-seq results showed that the deletion of nirS gene led to an up-regulation of mexEF-oprN efflux operon (Table S1). In the presence of 16.0 and 32.0 mg/L BIT, the planktonic growth of ΔnirS was higher than WT (Fig. 2) and results from RT-PCR additionally demonstrated that the relative expression levels of this efflux operon were higher in the same conditions (Fig. 5), implying that the mexEF-oprN efflux operon contributed to the resistance of ΔnirS to BIT. However, we also find that mexEF-oprN is a target site of BIT in WT and ΔnirS-com (Fig. 5b), which demonstrated that this efflux operon only plays its role in combating with BIT in the absence of nirS with unknown mechanisms.

The results from RNA-seq in the present study revealed that most genes belonging to the glycolysis pathway and pentose phosphate pathway were up-regulated, while those of the TCA cycle were down-regulated in the knockout nirS strain (Table 3), suggesting the probable use of HMP to generate the ATP and redox currency. It is well established that central metabolism has a profound influence on the bacterial response to ROS. The pentose phosphate pathway (PPP) contributes bacterial tolerance to oxidative stress by generating the redox currency (NADPH), which serves as the substrate for other reducing agents [18]. The production of free radicals by the isothiazolone biocides is likely to be the critical event leading to the reaction of the molecules contributing to cell death [43]. Perhaps, the up-regulation of PPP in ΔnirS relieves the action of radicals induced by BIT and thus explains its resistance to this biocide (Fig. 2). However, the complete molecular mechanisms of nirS together with the sugar metabolisms in P. aeruginosa ATCC 9027 warrant further study.

The ability of P. aeruginosa to swim in liquids is mediated by flagella, while its movement on surfaces is moderated by type IV pili [33, 39]. The deletion of fliM gene resulted in a defect in the swimming motility of P. aeruginosa PAO1 [2]. The fliM mutant did not show any swimming motility in either condition [47]. The flhF gene is essential for bacteria with polar flagella for their normal swimming motility. It was found through assays that the PAK strain of P. aeruginosa consistently showed a larger swimming zone than ΔflhF. However, the residual swimming motility of ΔflhF was significantly greater than that of ΔfliC, which do not express flagellin [28]. In the present study, we found an indifference in the swimming abilities of P. aeruginosa WT, ΔnirS and ΔnirS-com in the absence of BIT (Table 2). Additionally, the RNA-seq results demonstrated that there were no differences in the expression levels of all flagellar genes between WT and ΔnirS (Table S1), implying that nirS is not necessary for the swimming motility of P. aeruginosa. However, with the addition of 16 mg/l BIT, the swimming diameters of all three strains declined, suggesting that BIT has an inhibitory effect on this trait of P. aeruginosa, albeit unknown mechanisms.

Taken together, we successfully constructed the nirS deletion strain of P. aeruginosa ATCC 9027 and the phenotypic changes of ΔnirS were analyzed with its parental strain of WT in this study. The obtained results demonstrate that nirS gene is correlated with less BIT resistance, more biofilm formation, more c-di-GMP production and differential changes in carbon metabolism when compared to the nirS deletion strain.

Acknowledgements

This work was funded by the GDAS’ Project of Science and Technology Development (No. 2019GDASYL-0104006), the National Natural Science Foundation of China (Nos. 31770091 and 31500036), and Natural Science Foundation of Guangdong Province (No. 2015A030313713). We are grateful to Prof. Hai-hong Wang of South China Agricultural University for generously providing us the plasmids of pK18-GM and pSRK-GM.

Compliance with ethical standards

Conflict of interest

The authors declare that they have no conflicts of interest.

References

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.