A Library of Promoter-gfp Fusion Reporters for Studying Systematic Expression Pattern of Cyclic-di-GMP Metabolism-Related Genes in Pseudomonas aeruginosa

ABSTRACT

The opportunistic pathogen Pseudomonas aeruginosa is an environmental microorganism and is a model organism for biofilm research. Cyclic dimeric GMP (c-di-GMP) is a bacterial second messenger that plays critical roles in biofilm formation. P. aeruginosa contains approximately 40 genes that encode enzymes that participate in the metabolism of c-di-GMP (biosynthesis or degradation), yet it lacks tools that aid investigation of the systematic expression pattern of those genes. In this study, we constructed a promoter-gfp fusion reporter library that consists of 41 reporter plasmids. Each plasmid contains a promoter of corresponding c-di-GMP metabolism-related (CMR) genes from P. aeruginosa reference strain PAO1; thus, each promoter-gfp fusion reporter can be used to detect the promoter activity as well as the transcription of corresponding gene. The promoter activity was tested in P. aeruginosa and Escherichia coli. Among the 41 genes, the promoters of 26 genes showed activity in both P. aeruginosa and E. coli. The library was applied to determine the influence of different temperatures, growth media, and subinhibitory concentrations of antibiotics on the transcriptional profile of the 41 CMR genes in P. aeruginosa. The results showed that different growth conditions did affect the transcription of different genes, while the promoter activity of a few genes was kept at the same level under several different growth conditions. In summary, we provide a promoter-gfp fusion reporter library for systematic monitoring or study of the regulation of CMR genes in P. aeruginosa. In addition, the functional promoters can also be used as a biobrick for synthetic biology studies.

IMPORTANCE The opportunistic pathogen P. aeruginosa can cause acute and chronic infections in humans, and it is one of the main pathogens in nosocomial infections. Biofilm formation is one of the most important causes for P. aeruginosa persistence in hosts and evasion of immune and antibiotic attacks. c-di-GMP is a critical second messenger to control biofilm formation. In P. aeruginosa reference strain PAO1, 41 genes are predicted to participate in the making and breaking of this dinucleotide. A major missing piece of information in this field is the systematic expression profile of those genes in response to changing environment. Toward this goal, we constructed a promoter-gfp transcriptional fusion reporter library that consists of 41 reporter plasmids, each of which contains a promoter of corresponding c-di-GMP metabolism-related genes in P. aeruginosa. This library provides a helpful tool to understand the complex regulation network related to c-di-GMP and to discover potential therapeutic targets.

INTRODUCTION

Biofilms are communities of microorganisms embedded in extracellular polymeric substances (EPS) (1). Biofilms can form on biotic and abiotic surfaces and exist as non-surface-attached aggregates (2–4), which protect microorganisms from harsh conditions and are a key feature of chronic and persistent infections (5). The second messenger bis-(3′-5′)-cyclic GMP (c-di-GMP) plays an important role in biofilm formation (6, 7). In general, a high level of c-di-GMP stimulates the biosynthesis of adhesins and exopolysaccharides to promote biofilm formation, whereas a low level of c-di-GMP is associated with an increase in motility and virulence (8, 9). One well-studied regulation mechanism of c-di-GMP that controls biofilm or motility is through c-di-GMP receptors. For example, when c-di-GMP is at a high level, it interacts with Alg44 and PelD, the c-di-GMP receptors in the exopolysaccharide biosynthesis apparatus of alginate and Pel, respectively, to enhance their synthesis in Pseudomonas aeruginosa (10, 11). In addition, c-di-GMP could also fine-tune swimming speed by binding to a molecular brake, YcgR (12). c-di-GMP is synthesized from two GTP molecules by diguanylate cyclases (DGC) that usually harbor a conserved GGDEF domain and is hydrolyzed by c-di-GMP-specific phosphodiesterases (PDE) with conserved EAL or HD-GYP domains, which degrade c-di-GMP to pGpG (8). In some bacteria, such as P. aeruginosa, there are multiple genes predicted to have a conserved GGDEF domain and/or EAL or HD-GYP domains (13). The systematic expression pattern of those genes remains largely unclear due to lack of available tools for investigation.

P. aeruginosa is a Gram-negative gammaproteobacterium that can grow in diverse habitats and acts as an opportunistic pathogen over a wide range of hosts, including humans, animals, and plants (14). The success of P. aeruginosa infections relies on both the production of acute virulence factors and the ability to form biofilms (15–20). This opportunistic pathogen has become a model bacterium for biofilm research. There are 41 genes in P. aeruginosa strains, such as PAO1 and PA14, whose encoded proteins were predicted to be involved in the synthesis and degradation of c-di-GMP, including 17 different proteins with a GGDEF domain, 8 with an EAL or HD-GYP domain, and 16 with both types of domains (13). In strain PAO1, 33 genes were functionally characterized to encode a DGC or PDE; exceptions were PA0290, PA0338, PA1851, PA2072, PA2572, PA3258, PA4396, and PA5442 (21–52). The enzymes with both GGDEF and EAL domains usually have only one catalytic activity, either PDE or DGC activity (13). However, enzymes encoded by PA0861 (rbdA), PA1727 (mucR), and PA4601 (morA) were found to conditionally switch between the two activities (43, 53–58). Intracellular c-di-GMP level in P. aeruginosa relies on the expression of those 41 c-di-GMP metabolism-related (CMR) genes. However, it remains a mystery how P. aeruginosa controls the expression of those c-di-GMP metabolism enzymes.

P. aeruginosa is a ubiquitous microorganism which is able to survive in a variety of environments. Its optimum growth temperature is 37°C. Growth occurs at temperatures as high as 42°C but not at 4°C (59, 60). It has been reported that temperatures regulate biofilm formation of P. aeruginosa and the temperature-dependent changes in biofilm formation might be mediated by c-di-GMP (61). Almblad et al. have discovered a temperature-regulated DGC that coordinates temperature-dependent biofilm formation, motility, and virulence factor expression in P. aeruginosa CF39S, a strain isolated from the sputum of patients with chronic pulmonary cystic fibrosis (CF) (62, 63). Many environmental factors (such as oxygen levels, nitric oxide, iron, and nutrients) and small chemical compounds have been reported to trigger biofilm dispersion (43, 64–66). Studies have shown that subinhibitory concentrations of antibiotics can induce biofilm formation by P. aeruginosa (47, 67, 68). A systematic tool will be a great help for a deep understanding of how P. aeruginosa regulates the 41 CMR genes in response to various environmental temperatures, signals, and antibiotics. A few reports have focused on the comprehensive and systematic features of those c-di-GMP-related genes or proteins in bacteria (23, 25, 69–71); those studies are mainly focused on the functions of genes or enzymes. Therefore, a systematic transcriptional profile of the 41 CMR genes in P. aeruginosa remains lacking.

In this study, we provide a promoter-gfp transcriptional fusion reporter library of the 41 genes that are related to c-di-GMP metabolism in P. aeruginosa for systematic transcriptional profile analysis. The plasmid pPROBE-AT′ was selected as the vector to construct promoter-gfp transcriptional fusions due to its broad host range, high stability without antibiotic selection, and low background levels of expression in multiple taxa (72). More importantly, this promoter-probe vector is able to detect weak or moderate promoters. We have tested this CMR gene promoter-gfp transcriptional fusion reporter library in P. aeruginosa with different temperatures, growth media, and antibiotics. Each condition did exhibit a distinctive expression profile of CMR genes. Meanwhile, our results also revealed that expression of some genes could be enhanced by a certain condition, which would provide novel targets for the investigation and therapy of P. aeruginosa in the future.

RESULTS

Construction of a promoter-gfp transcriptional fusion reporter library for monitoring the transcriptional profile of 41 CMR genes in P. aeruginosa.

The promoter regions of 41 genes that are (or are predicted to be) involved in c-di-GMP metabolism in P. aeruginosa PAO1 were cloned into the vector pPROBE-AT′, resulting in a library of promoter-gfp transcriptional fusion reporter plasmids (Fig. 1) which can be used to monitor the expression of the corresponding genes in P. aeruginosa.

FIG 1.

The information of 41 plasmids in the library of c-di-GMP metabolism-related (CMR) gene promoter-gfp transcriptional fusion reporters and corresponding genes. +1, the first base of the start codon of corresponding gene or its operon; ND, not determined; PA#, corresponding locus number in genome of P. aeruginosa PAO1; †, the function is not fully identified; FIG, fluorescence intensity of Gfp (the value is the average fluorescence intensity per OD₆₀₀ of the corresponding sample grown in Jensen’s medium at 37°C for 24 h); *, the gene within an operon and its promoter region cloned into the vector is the region upstream of the first gene of corresponding operon.

To evaluate whether the cloned promoter regions were functional, the corresponding promoter-gfp fusion plasmids were transformed into PAO1. Their fluorescence intensities were measured in Jensen’s medium at 37°C. The fluorescence intensities of these reporters vary from the hundreds to over 10,000, indicating that all promoters can initiate the transcription of gfp. Among the 41 genes, the promoter of PA2818 (arr) showed the highest fluorescence intensity (Fig. 2A). PA2818 encodes a PDE named Arr, which is responsible for aminoglycoside antibiotic-induced biofilm formation (47). Moreover, the fluorescence intensities of 12 genes, namely, PA0169 (siaD), PA0338, PA2771 (encoded protein was named as Dcsbis), PA2870, PA4843 (gcbA), PA5487 (dgcH), PA4781, PA0861 (rbdA), PA2072, PA3258, PA3311 (nbdA), and PA4959 (fimX), is also above 10,000. PA0169 encodes SiaD, a well-characterized DGC (21); PA5487 (dgcH) encodes a conserved DGC (22); PA4843 encodes the DGC GcbA (33); both PA4781 and PA3311 (nbdA) encode functional PDE (43, 50, 51); PA0861 (rbdA) has intrinsic GTP-stimulated PDE activity as well as DGC activity (42, 53); and PA2072 encodes an enzyme whose function has not been determined so far (40). The fluorescence intensity of stationary-phase cultures is higher than that of the exponential-phase cultures (Fig. 2A), which is consistent with the stability of Gfp. Previous publications (23, 73) indicated that the DGC SadC had stronger effects on intracellular c-di-GMP concentration than other DGC of P. aeruginosa. However, our results indicated the transcriptional level of sadC was relatively low among DGC (Fig. 1). We then selected six genes with different transcriptional levels, PA0285 (pipA), PA0169 (siaD), PA2818 (arr), PA4332 (sadC), PA4601 (morA), and PA5487 (dgcH), to detect their transcription by quantitative reverse transcription-PCR (qRT-PCR) and the translation of Gfp from corresponding promoter-gfp by Western blotting against anti-Gfp antibody. As shown in Fig. 2C and D, the transcriptional level of these genes is consistent with the results of promoter-gfp, indicating that the library can be used to monitor the expression pattern of the corresponding genes. Given that pPROBE-AT′ is a stable plasmid, we compared the fluorescence intensities of the 41 plasmids with and without the addition of antibiotics. Without antibiotics, the fluorescence intensities of reporters showed some decrease but it was not significant, except for that of a few genes (especially those with fluorescence intensity over 10,000), suggesting that the library could be used without antibiotics in growth media (Fig. 2A) or in vivo study.

FIG 2.

Fluorescence intensity of the CMR gene promoter-gfp fusion reporter library in P. aeruginosa or E. coli and transcription of selected genes detected by qRT-PCR. (A) Fluorescence intensity CMR promoter-gfp fusion reporter in P. aeruginosa cultures at exponential phase with carbenicillin and at stationary phase with and without carbenicillin. The fluorescence intensities were normalized by corresponding OD₆₀₀. (B) Fluorescence intensity of CMR promoter-gfp fusion reporters in E. coli. Shown are the fluorescence intensities of stationary-phase cultures normalized by corresponding OD₆₀₀. (C) Expression levels of six genes from PAO1 by qRT-PCR, fluorescence intensity, and Gfp signal from anti-Gfp Western blotting of the corresponding promoter-Gfp reporter. In qRT-PCR, the housekeeping gene rpsL was used as an endogenous control to normalize the quantification of the mRNA target. The results of Western blotting were quantified by ImageJ. All data were relative to the level of PA0285. **, P < 0.01; ***, P < 0.001. (D) Correlation analysis of qRT-PCR data with fluorescence intensity and Western blotting.

To test whether the 41 promoters could remain functional in Escherichia coli, the fluorescence intensities of promoter-gfp fusion plasmids in E. coli strains were measured under the same cultural conditions. As shown in Fig. 2B, the promoters of four genes, namely, PA4843 (gcbA), PA2818 (arr), PA0861 (rbdA), and PA3258, kept high activity in E. coli (Fig. 2B, marked by a red dot; the fluorescence intensity of Gfp [FIG] is over 500), while their fluorescence intensity was approximately 10-fold less than that in P. aeruginosa. The promoters of 15 genes (Fig. 2B, marked with numbers in red) showed a baseline level of fluorescence intensity, suggesting that these genes’ transcription might be Pseudomonas specific. The other 22 genes’ promoters had a low expression level (FIG of approximately 50 to 300) (Fig. 2B).

Effect of temperature on the transcriptional profile of CMR genes in P. aeruginosa.

P. aeruginosa can live in diverse ecological niches. Therefore, we investigated whether temperature affected the transcription of genes that were involved in c-di-GMP metabolism in P. aeruginosa PAO1. The 41 strains containing corresponding plasmids were cultured at 25°C or 30°C, and their fluorescence intensities were compared to those of their corresponding cultures grown at 37°C. As shown in Fig. 3A, PA0169 (siaD) and PA2567 exhibited higher expression at both 25°C and 30°C; these genes also have high levels of transcription at 25°C, suggesting that they might be important at lower temperatures. Overall, the transcription levels of most genes were relatively low at 25°C or 30°C compared to those at 37°C, while 11 genes, namely, PA0338, PA3177, PA3343 (hsbD), PA4843 (gcbA), PA2818 (arr), PA1433 (lapD), PA4367 (bifA), PA4601 (morA), PA4959 (fimX), PA3258, and PA5295 (proE), had similar transcriptional levels at three tested temperatures (Fig. 3, marked with red numbers). Among these 11 genes, PA3177 and PA3343 (hsbD) encode DGC (29, 30), PA2818 (arr), PA4367 (bifA), PA4959 (fimX), and PA5295 (proE) encode PDE (44–47), and PA4601 (morA) encodes a protein that functions as both a DGC and PDE. Furthermore, MorA is conserved among diverse Pseudomonas species (55, 56). Compared to the corresponding samples at 37°C, 6 genes exhibited lower-level expression at 25°C yet had a similar expression level at 30°C (Fig. 3A, marked with black numbers). Taken together, our results showed that lower temperature generally reduced the expression of genes involved in c-di-GMP metabolism with some exceptions, suggesting that the gene with higher-level transcription at lower temperature might be important for the survival of P. aeruginosa in environment.

FIG 3.

Effects of different temperatures on the promoter activity of CMR genes, biofilm biomass, c-di-GMP level, and swimming motility of P. aeruginosa PAO1 in Jensen’s medium. (A) Promoter activity of each CMR gene in PAO1 at different temperatures. The value was normalized to the fluorescence intensity per OD₆₀₀ of the corresponding strain grown in Jensen’s medium at 37°C for 24 h. Red numbers indicate the genes for which promoter activity did not change at these three temperatures; black numbers mark the genes for which promoter activity at 30°C was similar to that at 37°C. Small asterisk, significantly decreased; large asterisk, significantly increased. Errors were calculated from three independent experiments. (B) Effects of temperatures on phenotypes of PAO1. Shown are the biofilm biomass, c-di-GMP level, and swimming zone of PAO1 after 24 h of growth in Jensen’s medium at different temperatures. *, P < 0.05; **, P < 0.01.

We then tested the biofilm biomass, swimming motility, and c-di-GMP level of PAO1 at 25°C, 30°C, and 37°C in Jensen’s medium. Strikingly, 25°C promoted the highest biofilm biomass and c-di-GMP levels, both of which were reduced at a temperature increasing manner. In contrast, motilities were increased at higher temperature (Fig. 3B). Given that the transcription of only a few genes was upregulated at 25°C and 30°C, our data suggested that the phenotype could be determined by the expression of a few key genes.

Effects of different media on the transcriptional profile of CMR genes in P. aeruginosa.

To investigate whether nutrients affect the transcriptome of genes related to intracellular c-di-GMP metabolism, we examined expression of 41 genes in four media, M63 minimal medium, chemical defined Jensen’s medium, Luria-Bertani (LB) rich medium, and artificial sputum medium (ASM). The fluorescence intensity of each promoter-gfp fusion reporter was normalized to its corresponding value in Jensen’s medium grown at 37°C (Fig. 3).

The promoter activities (or expression levels of the corresponding genes) significantly decreased in the M63 minimal medium compared to those in Jensen’s medium, except for PA2818 (arr) and PA4396, which had the same level in M63 medium as in Jensen’s medium (Fig. 3, marked in red). In the LB medium, most genes exhibited either similar expression levels (11 out of 41 genes) or reduced levels (28 out of 41 genes) compared to that in Jensen’s medium (Fig. 4). However, in ASM, the expression levels of PA2818 (arr), PA2133 (fcsR), and PA4396 were increased (Fig. 4, marked by a red star); 15 out of 41 genes (PA0169, PA0338, PA1120, PA2870, PA3343, PA4332, PA5487, PA2200, PA3947, PA0285, PA1727, PA3258, PA4959, PA5295, and PA5442) had similar expression levels, while the expression levels of the other 23 genes were decreased.

FIG 4.

Effects of different growth media on the promoter activity or expression of CMR genes in the P. aeruginosa PAO1 background. The value was normalized to the fluorescence intensity per OD₆₀₀ of the corresponding strain grown in Jensen’s medium at 37°C for 24 h. Black asterisks, significantly reduced; red asterisks, significantly elevated. Genes highlighted in red were significantly elevated in ASM, with no change in the other media; genes highlighted in blue were significantly reduced in M63 medium, with no change in the other media; and the gene highlighted in green showed no change in the three media. Shown are averages with standard deviations calculated from three independent experiments. *, P < 0.05.

Strikingly, the promoters of PA2818 (arr) and PA4396 exhibited the highest activity in Jensen’s medium (Fig. 1), which remained at the same level in either M63 or LB medium, and an even higher level in ASM. These results suggested the importance of these two genes. In addition, the promoter activity of PA3343 (hsbD) (marked in blue) was reduced only in M63 medium and kept the same level in the other three media. PA5487 (dgcH) (marked in green) was reported to be a conserved DGC with highly invariable expression, which showed similar levels of transcription in three media (22).

We also examined the biofilm biomass and c-di-GMP level of PAO1 grown in the three media. The biofilm biomass and c-di-GMP level went from high to low in Jensen’s medium, LB medium, ASM, and M63 medium (Fig. 4), and these results were consistent with the transcriptional profile of CMR genes in the three media.

Effects of subinhibitory concentrations of antibiotics on the expression pattern of CMR genes.

Previous reports showed that subinhibitory concentrations of aminoglycoside antibiotics induced biofilm formation by P. aeruginosa (47). We then examined the effect on our reporter library of three types of antibiotics at subinhibitory concentrations, including the aminoglycoside tobramycin, the fluoroquinolone ciprofloxacin, and the macrolides erythromycin and azithromycin. Our results showed that tobramycin increased the expression of PA2818 (arr) (Fig. 5A), which is consistent with the previous report about the contribution of PA2818 to tobramycin-induced biofilm formation (47). The expression of two genes, PA5295 (proE) and PA5442, was also induced under subinhibitory concentrations of tobramycin (Fig. 5A). ProE is a very active PDE with high enzymatic activity in the degradation of c-di-GMP and plays an important role in regulating EPS production of P. aeruginosa PAO1 (46), and the function of PA5442 is not determined. The transcription of 12 genes (PA0338, PA1107, PA2771, PA2870, PA4332, PA4396, PA2133, PA4781, PA1433, PA4601, PA4959, and PA5017) were not influenced by the addition of tobramycin. The expression of the other 26 genes was significantly reduced in the presence of tobramycin.

FIG 5.

Effects of subinhibitory concentrations of antibiotics on the promoter activity of CMR genes in P. aeruginosa PAO1. (A to D) Results of the CMR gene promoter-gfp fusion reporter library under tobramycin, ciprofloxacin, erythromycin, and azithromycin treatment, respectively. The fluorescence intensity per OD₆₀₀ of the corresponding strain grown in Jensen’s medium at 37°C for 24 h was normalized to the corresponding sample without antibiotic treatment. The genes with promoter activity significantly elevated in more than 2 antibiotics are indicated with arrows. Small asterisks mark the genes with significantly reduced levels, and large asterisks indicate those with significantly increased levels. Errors were calculated from three independent experiments. *, P < 0.05.

Among the four antibiotics, ciprofloxacin induced the expression of a large number of genes. As shown in Fig. 5B, the expression of 15 genes was significantly elevated, while the expression of 8 genes was decreased and that of 18 genes showed no change. It is worth noting that expression of PA2818 (arr), PA5295 (proE), and PA5442 was enhanced by both tobramycin and ciprofloxacin.

Erythromycin and azithromycin both repressed transcription of most c-di-GMP metabolism genes; the expression of only a few genes was enhanced (Fig. 5C). PA0285 (pipA) was the only gene whose transcription was enhanced by both erythromycin and azithromycin, while the transcription of PA1727 (mucR) was enhanced by azithromycin but reduced by erythromycin.

DISCUSSION

c-di-GMP is an important second messenger involved in bacterial switching from the motile to sessile lifestyle. It is critical for bacteria to control the intracellular c-di-GMP level in response to changing environments. The opportunistic pathogen P. aeruginosa can live in diverse ecological niches. Consistently, it has 41 genes that are predicted to encode proteins for the synthesis or degradation of c-di-GMP. Extensive studies have been done to investigate their functions with respect to biofilm formation and motility. However, it remains a mystery when and where those genes will be transcribed, and there is still a paucity of information concerning the systematic expression pattern of those genes. In this study, we have constructed a promoter-gfp transcriptional fusion reporter library to systematic investigate the transcription or regulation of the 41 c-di-GMP metabolism-related (CMR) genes.

We have examined this CMR gene promoter-gfp fusion library under different growth conditions, including different media, temperatures, and antibiotics. Each condition did affect the transcription of different genes, and the transcription of some genes was induced under a typical condition. For example, the expression of PA2567 was enhanced by lower temperatures (25°C > 30°C > 37°C). The subinhibitory concentrations of tobramycin induced the transcription of three genes, PA2818 (arr), PA5295 (proE) and PA5442. A previous study has shown that PA2818 (arr) is important for biofilm formation induced by aminoglycoside antibiotics (47), whereas the functions of PA5295 (proE) and PA5442 have not been linked with antibiotics. Our results showed that this library not only is helpful for studying the systematic expression pattern of CMR genes but also can reveal the new roles of those genes.

It is worth noting that PA2818 (arr) had the highest transcription among the 41 CMR genes. Moreover, its transcript was consistently stable under several tested conditions and elevated when P. aeruginosa PAO1 was grown in artificial sputum media or in the presence of subinhibitory concentrations of aminoglycoside antibiotics. These results suggested that PA2818 (arr) might play a key role in biofilm-related persistent infections caused by P. aeruginosa. Given that P. aeruginosa can cause life-threatening lung infections in cystic fibrosis patients, our results have also suggested a potential therapeutic target.

The promoter-probe vector pPROBE-AT′ used in this study is a broad-host-range vector; thus, we examined the promoter’s activity in both P. aeruginosa and E. coli. The promoters of 26 genes can initiate the transcription of gfp in E. coli, suggesting that these promoters might function in Gram-negative bacteria. Therefore, this promoter-gfp library also provides a library of biobricks for synthetic biology.

In summary, the CMR gene promoter-gfp transcriptional fusion reporter library we have constructed in this study is a versatile and useful tool. The library can be used in the following aspects: (i) investigating the systematic expression pattern and regulatory network of CMR genes in P. aeruginosa, (ii) determining the regulatory mechanism of any factor that affects intracellular c-di-GMP in P. aeruginosa, (iii) discovering compounds or drugs that target the intracellular c-di-GMP of P. aeruginosa, (iv) elucidating the role of CMR genes and their regulated pathways, and (v) as a promoter library for future applications in synthetic biology.

MATERIALS AND METHODS

Bacterial strains and growth conditions.

P. aeruginosa strain PAO1 was cultured in Luria-Bertani (LB) medium (74), Jensen’s medium (75), M63 medium (24), or artificial sputum medium (ASM) (76). E. coli DH5α (Tsingke Biotechnology Co., Beijing, China) was inoculated into LB medium at 37°C and 200 rpm. The chemical compositions of all media used in this study can be found at the National Microbiology Data Center (NMDC; https://nmdc.cn/cyclicdigmp/library/media?type=Jensen%27s). When necessary, antibiotics were added as follows: 300 μg/mL of carbenicillin for P. aeruginosa and 100 μg/mL of ampicillin for E. coli. The subinhibitory concentration of antibiotics was applied as 30 μg/mL of erythromycin (77), 2 μg/mL of azithromycin (78), 0.3 μg/mL of tobramycin (47), or 0.0625 μg/mL of ciprofloxacin (79).

Construction of promoter-gfp transcriptional fusion reporter plasmid.

The information for plasmids constructed in this study and a schematic diagram of the cloned promoter regions are shown in Fig. 1. The pPROBE-AT′ (Apr) was used as a vector for promoter-gfp transcriptional fusions (72). The information about cloned promoter regions, their corresponding genes, and PA number is also listed in Fig. 1. The cloned promoter sequences consist of the entire intergenic region, or encompassing up to 500 bp upstream of the translational starting site and 138 bp of the coding region of corresponding open reading frames (ORFs) (Fig. 1). The intergenic region was based on the Pseudomonas Genome Database (https://www.pseudomonas.com). The promoter regions and ribosome binding site (RBS) were predicted by the software at the following websites: http://www.phisite.org/main/index.php?nav=tools&nav_sel=hunter, https://services.healthtech.dtu.dk/service.php?Promoter-2.0, and https://www.fruitfly.org/seq_tools/promoter.html (80–82). For those genes within an operon or without software-predicted promoter characteristics in their intergenic region, their promoter regions were further confirmed by RT-PCR and/or 5′ rapid amplification of cDNA ends (5′-RACE) (https://nmdc.cn/cyclicdigmp/library/analysis). All promoter regions were amplified from the genomic DNA of P. aeruginosa PAO1 and then cloned into the EcoRI-BamHI sites of pPROBE-AT′. The cloned regions in each corresponding plasmid were all confirmed by sequencing. The sequence of cloned promoter regions and primers used in this study can be found at National Microbiology Data Center (NMDC; https://nmdc.cn/cyclicdigmp/library/promoter?domian=GGDEF); primers are also listed in Table 1. The promoter-Gfp fusion plasmid was introduced into the P. aeruginosa strain by the chemical transformation in order to determine the promoter’s activity in P. aeruginosa.

TABLE 1.

Primers used in this study

Primer	Sequence (5′→3′)
Primers for qRT-PCR
RT-rpsL-F	GTAAGGTATGCCGTGTACG
RT-rpsL-R	CACTACGCTGTGCTCTTG
RT-PA0169-F	GATGGACATCCTCGACCTGC
RT-PA0169-R	CGGCTCATCGTCGGCTACT
RT-PA0285-F	GGGACTTGGGAGTGATGAG
RT-PA0285-R	CGGGCAGTATAAATGATGG
RT-PA2818-F	GCTCGTCCTGGTCTCCTTTACT
RT-PA2818-R	CCCGGAACGAATCTTACCC
RT-PA4332-F	GCGTGTTGTCCTTGGTGTTCT
RT-PA4332-R	GGATCGTCACCGTGTTCGTC
RT-PA4601-F	GCATACCCTGGAGCAGATGTT
RT-PA4601-R	CGGCTGTCGAGGCACTTT
RT-PA5487-F	ATTTCCAGCTACATCCAGGGTC
RT-PA5487-R	GGTCACGGGTTCCAGCATT
Primers for cotranscriptional analysis
F1-PA0171-PA0169-F	GCCAATCTCAAGGGCTACGG
R1-PA0171-PA0169-R	CTCGGACAGCGACAGGTTCT
F2-PA0170-PA0172-F	CTGGCGCCGGGCTGGACCTTCTACC
R2-PA0170-PA0172-R	GTGGACTGGGTGCCGGGTATGTGC
F3-PA0336-PA0337-F	GAAGAAGTCGGGCTGGAGG
R3-PA0336-PA0337-R	GATGGAACGCTTGTTGAGGC
F4-PA0337-PA0338-F	ACTTCCTTTCGGTCGGTTCG
R4-PA0337-PA0338-R	CGCCCGTTCGTCCATTTT
F5-PA1120-PA1121-F	GGCCAATCTCGACTCCATCGC
R5-PA1120-PA1121-R	CGTCGTGCAGGATCGGTTGTT
F6-PA1433-PA1434-F	TGGACAACCTGATCGGCGAGAT
R6-PA1433-PA1434-R	GCTGAAGGCGACGACGAGGAA
F7-PA2128-PA2129-F	CGCCTTCACCCTGCAACTCA
R7-PA2128-PA2129-R	CGCAGCATCAGCAGCATCTGG
F8-PA2199-PA2200-F	CGCCTGTCTATCGCAACTACGG
R8-PA2199-PA2200-R	GAAATCTCCAGTGCCCGCTCC
F9-PA2869-PA2870-F	GATGGCCCGCTATCCTGAG
R9-PA2869-PA2870-R	AGATGCCTGAACAACGACTGG
F10-PA3257-PA3258-F	CCATCCAGCCGCTCAACCAC
R10-PA3257-PA3258-R	TTCGCTCAGCCGTCCGCATT
F11-PA3702-PA3703-F	CGCCGAACCGCGTTCTTT
R11-PA3702-PA3703-R	TAGGCAGCGAGCAGCGTGAG
F12-PA4958-PA4959-F	CGGCGGCTATTGCCTGAGCT
R12-PA4958-PA4959-R	CGAGCAGCAACTGGCAGCGTTT
F13-PA5487-PA5489-F	ATGGTGGTCAATGGCAAATATCGCTTC
R13-PA5487-PA5489-R	GCTCCTTCATGCACTGGTCGACC

Isolation of total RNA and qRT-PCR.

PAO1 was cultivated in Jensen’s medium at 37°C and 200 rpm until the optical density at 600 nm (OD₆₀₀) reached 3.0, at which point 1 mL of cells was harvested and RNA samples were extracted using the RNAprep pure bacterial kit (Tiangen Co., Beijing, China). Total cDNA was generated from 500 ng of total template RNA using the HiScript III All-in-one RT SuperMix Perfect for qPCR (Vazyme Co., Nanjing, China). The ChamQ universal SYBR qPCR master mix (Vazyme Co., Nanjing, China) and gene-specific primers were used to perform qRT-PCR on 5 ng/μL of cDNA in a real-time PCR system (Applied Biosystems ViiA 7; USA). Real-time PCR data were analyzed, validated, and calculated according to the instructions of the manufacturer. All samples were normalized to constitutively produced rpsL transcripts.

Western blotting.

Cells were lysed for 30 min on ice with lysis buffer (YEASEN Biotech Co., Ltd.) containing a phosphatase inhibitor cocktail (YEASEN Biotech Co., Ltd.) and a protease inhibitor cocktail (YEASEN Biotech Co., Ltd.). The mixture was treated with an ultrasonic homogenizer (Ningbo Xinzhi Biotechnology Co., Ltd., China) at 300 W and 4°C for 5 min. After centrifugation at 12,000 × g and 4°C for 15 min, the supernatant was boiled with 5× protein SDS-PAGE loading buffer (ZOMANBIO) for 10 min, and the proteins were separated by SDS-PAGE and transferred onto a polyvinylidene difluoride (PVDF) membrane (Merck Millipore). The membrane was preincubated with 5% skim milk in TBST buffer (20 mM Tris-HCl [pH 8.0], 150 mM NaCl, and 0.05% Tween 20) for 1 h, following 1 h of incubation with a green fluorescent protein (GFP) antibody (ABclonal) in TBST buffer with 5% skim milk at 25°C, and then washed with the TBST buffer and incubated with an alkaline phosphatase-conjugated secondary antibody (Abcam) for 1 h. Lastly, the 5-bromo-4-chloro-3-indolylphosphate (BCIP)/nitroblue tetrazolium (NBT) alkaline phosphatase color development kit (ZOMANBIO) was used to detect the targeted proteins.

Measurement of fluorescence intensity.

To determine the fluorescence intensity of the promoter, overnight cultures of the plasmid-carrying strains grown in Jensen’s medium with carbenicillin or ampicillin were inoculated at 1% into 200 μL of Jensen’s medium (unless otherwise specified) containing carbenicillin or ampicillin in 96-well plates (NEST Co., Wuxi, China) and grown with shaking at 700 rpm in a constant-temperature microplate shaker (MIULAB Co., Hangzhou, China) at 37°C (unless another specific temperature was required) until the time for measurement (8 h for exponential-phase culture or 24 h for stationary-phase culture). The fluorescence intensity of Gfp was measured on a Synergy H4 hybrid reader (BioTek, USA) at an excitation wavelength of 488 nm and emission wavelength of 520 nm with a gain of 50. Biomass was determined at 600 nm with a gain of 20 as scattered light (22). Fluorescence signal value was normalized to OD₆₀₀, and the value from the vector pPROBE-AT′ was subtracted as background fluorescence. All experiments were performed with a minimum of three biological triplicates. Statistically significant differences were determined using a two-tailed Student t test and one-way analysis of variance (ANOVA) (P < 0.05).

Biofilm assay.

The biofilm assay was performed as previously described (83). Briefly, an overnight culture was diluted 1:100 in Jensen’s medium (unless other specific medium was required) and grown in triplicate in a polyvinylchloride plate (Costar) at 37°C (unless another specific temperature was required) for 24 h, then stained with 0.1% crystal violet at room temperature for 15 min, and washed twice by being dipped into a standing water bath. The adherent stain was solubilized in 40% acetic acid and quantified by measuring its optical density at 560 nm.

Motility assays.

A swimming assay was performed as previously described (84). The strains were stab-inoculated with a sterile toothpick onto the surface of Jensen’s medium plates (0.3% BD Bacto agar) and cultivated at 25°C, 30°C, or 37°C overnight, and the swimming motility zones were measured.

Quantification of c-di-GMP level by using pCdrA::gfp.

The c-di-GMP levels of PAO1 were determined according to a previously described method by using pCdrA::gfp as a reporter (85). The fluorescence intensity was measured on a Synergy H4 hybrid reader (BioTek, USA) at an excitation wavelength of 488 nm and an emission wavelength of 520 nm with a gain of 50.