Artificial Control of Nitrate Respiration through the lac Promoter Permits the Assessment of Oxygen-Mediated Posttranslational Regulation of the nar Operon in Pseudomonas aeruginosa

Abstract

In this study, oxygen and nitrate regulation of transcription and subsequent protein expression of the unique narK1K2GHJI respiratory operon of Pseudomonas aeruginosa were investigated. Under the control of P_LAC, P. aeruginosa was able to transcribe nar and subsequently express methyl viologen-linked nitrate reductase activity under aerobic conditions without nitrate. Modulation of P_LAC through the LacI repressor enabled us to assess both transcriptional and posttranslational regulation by oxygen during physiological whole-cell nitrate reduction.

Pseudomonas aeruginosa is a ubiquitous gram-negative bacterium capable of growth and/or survival anaerobically through arginine catabolism (36), pyruvate fermentation (6), or denitrification in the presence of nitrogen oxides (37). The latter process allows this organism to persist in soil as part of the global nitrogen cycle. Additionally, denitrification has been implicated in infections by this opportunistic pathogen in the airways of cystic fibrosis patients (9, 18, 35).

During anaerobic growth of Escherichia coli, the Fnr protein is responsible for activation of the synthesis of anaerobic respiratory enzymes such as nitrate reductase (28, 31). In addition, the presence of external nitrate induces the transcription of the nitrate reductase operon through the dual two-component regulatory systems of narX-narL (31, 32) and narQ-narP (3, 21, 22). Parallel studies of P. aeruginosa have resulted in the characterization of a unique nar operon (24, 27) regulated by the proteins Anr and Dnr (40) as well as narX and narL (24). However, a narQ homologue has not been identified (30, 34).

Posttranslationally, oxygen also has the capacity to inhibit denitrification immediately at the level of nitrate uptake and nitrite excretion (10, 11, 33) as well as through the diversion of electron flow to oxygen in E. coli and in Paracoccus denitrificans (4, 33). Despite these studies, an experimental method for the measurement of posttranslational regulation by oxygen has been lacking.

In the present study, the effects of oxygen and nitrate on the expression of the narK1K2GHJI operon (27) were examined during aerobic or anaerobic growth with and without nitrate. In addition, a P_LAC element was inserted upstream of the respiratory nitrate reductase genes (narK2GHJI) of P. aeruginosa to overcome transcriptional regulation of the nar operon by oxygen and nitrate. The levels of transcription, respiratory nitrate reductase activity, and whole-cell physiological reduction of nitrate to nitrite were measured under both aerobic and anaerobic conditions, thus allowing quantitative assessment of posttranslational regulation by oxygen.

The bacterial strains and plasmids used in this study are listed in Table 1. All bacteria were grown at 37°C from single-colony isolates or overnight cultures in Luria-Bertani (LB) broth (Fisher Scientific, Pittsburgh, PA). The medium was supplemented with 1% (wt/vol) KNO₃ (LB-NO₃) when indicated. Aerobic cultures were set up as 50-ml volumes of LB or LB-NO₃ in a 500-ml Erlenmeyer flask by inoculating 500 μl of cells and shaking at 250 rpm to an optical density at 600 nm of 0.3 to 0.4. Cultures used to measure anaerobic transcription of narG and to conduct in vitro anaerobic respiratory nitrate reductase assays were first grown aerobically to an optical density at 600 nm of 0.3 to 0.4, after which time they were shifted to complete anaerobic conditions (BBL Anaerobic GasPak System) for 3 h. For anaerobic growth cultures and nitrite excretion studies, 1 ml of mid-log-phase cells was inoculated in 100-ml volumes of LB-NO₃ and incubated with magnetic stirring in a 125-ml Erlenmeyer flask with rubber stoppers equipped with ports for sample withdrawal and one-way gas release valves. To ensure complete anaerobiosis of the system, the medium was supplemented with 2% (wt/vol) Oxyrase (Oxyrase, Inc., Mansfield, OH) and flushed with argon. The antibiotics (BioWorld, Dublin, OH) used for E. coli were as follows: ampicillin (100 μg/ml), gentamicin (15 μg/ml), and tetracycline (25 μg/ml). Gentamicin (300 μg/ml) and carbenicillin (500 μg/ml) were used for P. aeruginosa.

TABLE 1.

Strains, plasmids, and oligonucleotide primers

Strain, plasmid, or primer	Relevant genotype, description or sequence (5′ to 3′)a	Source or reference
Strains
E. coli
DH5α	recA1 endA1 gyrA96 thi-1 hsdR17 supE44 relA1 ΔlacU169(φ80lacZΔM15)	Gibcoe
TOP10F′	F′ [lacIq Tn10(Tetr)] mcrA Δ(mrr-hsdRMS-mcrBC) φ80lacZΔM15 ΔlacX74	Invitrogene
	recA1 araD139 Δ(ara-leu)7697 galU galK rpsL (Strr) endA1 nupG
SM10	Kmr; mobilizer strain	29
P. aeruginosa
PAO1	Wild type	C. Manoil
JVC	PAO1 with narK1-Plac	This study
Plasmids
pUCP18	Apr; broad-host-range cloning vector	26
pUCGMlox	Gmr; Gmr cassette flanked by lox sequences	20
pUCGMxlac	Gmr; PLAC promoter ligated within the ClaI and SacII sites of pUCGMlox	This study
pGem-lacIq	Apr; 1-kb PCR fragment containing lacIq from TOP10F′ in pGEM-T Easy vector (Promega)	This study
pREP	Apr; ligation of a blunt-ended 1-kb EcoRI fragment containing lacIq from pGem-lacIq into the blunt-ended EcoRI and SapI sites of pUCP18	This study
pFLP2	Apr; Flp vector containing sacB and flp (Flp recombinase)	12
pCM157	Tcr; source of cre recombinase	15
pCR2.1-cre	Apr; Kmr; 1.3-kb PCR fragment containing cre recombinase in pCR2.1 (Invitrogen)	This study
pCRE	Apr; 1.3-kb EcoRI fragment of pCR2.1-cre ligated into an EcoRI digest of pFLP2	This study
pEX18Ap::narK1K2	Apr; 2.7-kb EcoRI fragment containing narK1 and narK2 ligated into the EcoRI site of pEX18Ap	27
pEXJVC	Apr; Gmr; 1.5-kb SmaI fragment from pUCGMxlac blunt ended into the NotI site of pEX18Ap::narK1K2	This study
Primer/templateb
Plac/Pucp18 (+)	TGTATCGATTCGCCACCTCTGACTT	This study
Plac/Pucp18 (−)	CTCCGCGGCGTAATCATGGTCATAG	This study
lacIq/Top 10F′ DNA (+)	GTGCAAAGCTCTTCGCGGTAT	This study
lacIq/Top 10F′ DNA (−)	CGCGAATTCACATTAATTGCGTTG	This study
cre/pCM157 (+)	GAATTCGCAAACCGCCTCTC	This study
cre/pCM157 (−)	CCAGTGAATTCTTACTAATCGCCATC	This study
narK1K2 (+)c	CCTGTCACTACCTCCAAAG	27
narK1K2 (−)c	AGAAGCTGATATTGGACATG	27
narG (+)d	ACGACCTCAACACCTCCGAC	This study
narG (−)d	GATCTCCCAGTCGCTCTTGG	This study
rpoD (+)d	GGGCGAAGAAGGAAATGGTC	23
rpoD (−)d	CAGGTGGCGTAGGTGGAGAA	23

^aKm^r, Ap^r, Gm^r, and Tc^r indicate resistance to kanamycin, ampicillin, gentamicin, and tetracycline, respectively. Underlined nucleotides were modified to accommodate a restriction site.

^bTemplate DNA used with the corresponding primer is indicated; plus and minus symbols indicate forward and reverse primers, respectively.

^cUsed for mutant confirmation.

^dUsed for real-time PCR.

^eLocated in Carlsbad, CA.

For the construction of strain JVC, a 1.5-kb SmaI digest containing Gmxlac was blunt ended into the NotI site of pEX18Ap::narK1K2. Single-copy, chromosomal gene disruptions were created using a gene replacement technique previously described (25, 29). loxP excision of the gentamicin resistance marker was conducted by transformation with pCRE into a P. aeruginosa strain with a narK1-Gmxlac disruption. Several passages of growth in LB containing 500 μg/ml carbenicillin were followed by selection for the loss of growth in LB containing 300 μg/ml gentamicin. The pCRE plasmid was cured from the strain through one passage in LB, followed by growth in LB containing 5% sucrose and selection for carbenicillin sensitivity. Constructs were confirmed by PCR using primers listed in Table 1.

JVC was constructed by inserting the Gmxlac cassette within narK1 for two reasons. First, it has been shown for P. aeruginosa that while narK1K2GHJI is transcribed as an operon, only NarK2 is required for respiratory nitrate reduction (27). Second, we did not want to interfere with physiologically regulated anaerobic transcription of the nar promoter since expression levels manipulated through a lac promoter could be either limiting or overexpressed and, thus, will not reflect physiological responses. Using the same reasoning, transcriptional termination sequences were not placed upstream of P_LAC in the pUCGmxlac vector. Thus, the transcriptional activity of narK2GHJI would be under the control of both the nar promoter and the P_LAC element during anaerobic conditions and be under the control of solely P_LAC during aerobic conditions.

Transcription of nar under aerobic and anaerobic conditions in JVC (narK1-P_LAC).

To test the ability of P_LAC to aerobically transcribe narK2GHJI, the mRNA levels of narG in PAO1 and JVC were measured through quantitative reverse transcription-PCR in LB with or without nitrate in aerobic and anaerobic conditions. The results further corroborated the previous experimental proof that in P. aeruginosa, narK1 and narK2 were contained in the nar operon (27) and demonstrated that the presence of nitrate alone was not sufficient for the transcription of narG in wild-type PAO1 and that anaerobic conditions were required (Table 2). In contrast, JVC transcribed narG aerobically in both the presence and the absence of nitrate (15.5 ± 3.8 and 23.4 ± 6.0 relative severalfold increases, respectively). The level of narG transcription anaerobically was significantly higher for both PAO1 and JVC since the operon was under the physiological control of the nar promoter. Finally, we manipulated the transcription of narG by transforming a plasmid containing lacI^q (pREP) into JVC. As expected, the addition of pREP significantly reduced aerobic narG transcriptional levels in JVC (Table 2).

TABLE 2.

Transcriptional levels of narG in P. aeruginosa PAO1 and JVC

Straina	Relative fold increase under indicated conditionsb
	Aerobic		Anaerobic
	LB	LB + 1% nitrate	LB + 1% nitrate
PAO1	1	1.1 ± 0.1	31.3 ± 2.2
PAO1 + pREP	NDc	1.5 ± 0.9	ND
JVC	23.4 ± 6.0	15.5 ± 3.8	55.5 ± 4.4
JVC + pREP	ND	6.2 ± 0.9	ND

^aThe PAO1 and JVC strains were grown in LB with or without 1% (wt/vol) KNO₃ under aerobic and anaerobic conditions as described in the text.

^bThe difference (n-fold) was calculated using the Pfaffl model (19), setting PAO1 in LB aerobically as 1. Values shown are means ± standard errors as calculated from three independent cell suspensions. PCR was accomplished by a 3-min denaturation step at 95°C, followed by 40 cycles of 30 s at 95°C, 45 s at 60°C, and 45 s at 72°C.

^cND, not determined.

The addition of P_LAC upstream of narK2GHJI enabled us to constitutively transcribe the operon in both the presence and the absence of oxygen and nitrate and to modulate aerobic transcriptional activity by utilizing pREP.

Respiratory nitrate reductase activity.

To determine whether the same trends are observed at the enzyme level, methyl viologen-linked respiratory nitrate reductase activity was measured (14) in cell extracts of strains grown in LB with or without nitrate aerobically and anaerobically (Table 3). The results indicated that enzyme activity was detected only in wild-type PAO1 under anaerobic conditions in the presence of nitrate (Table 3), confirming the narG transcriptional activities measured in Table 2 and the requirement of both nitrate and anaerobiosis for expression of the respiratory nitrate reductase. P. aeruginosa also contains a periplasmic nitrate reductase (38), but we did not detect any enzyme activity for P. aeruginosa under aerobic conditions (17, 27). Nitrate reductase activity in JVC was consistently detected even in the absence of nitrate and under aerobic conditions due to the expression of narGHJI through P_LAC (Table 3). Eighty percent of the specific activity detected aerobically without nitrate was retained within the insoluble fraction of the crude extract, indicating that the aerobic nitrate reductase measured was predominantly membrane bound (data not shown). Anaerobically, the nitrate reductase activity of JVC was approximately twice the level observed in wild-type PAO1 (1,120 ± 150 and 560 ± 77 nmol nitrite min⁻¹ mg⁻¹ protein, respectively [Table 3]). We also determined the ability of pREP to inhibit nitrate reductase activity and found both aerobic and anaerobic nitrate reductase activities to be significantly reduced in JVC with pREP (Table 3). Although pREP did not completely diminish anaerobic nitrate reductase activities, the level of inhibition was sufficient to significantly hinder anaerobic growth (data not shown).

TABLE 3.

Nitrate reductase activities of P. aeruginosa PAO1 and JVC

Straina	Nitrate reductase activity (nmol nitrite min−1 mg−1 protein) under indicated conditionsb
	Aerobic		Anaerobic
	LB	LB + 1% nitrate	LB	LB + 1% nitrate
PAO1	<1	<1	<1	560 ± 77
PAO1 + pREP	<1	<1	<1	548 ± 64
JVC	222 ± 27	210 ± 9	367 ± 66	1,120 ± 150
JVC + pREP	23 ± 7	28 ± 12	36 ± 13	218 ± 7

^aPAO1 and JVC were grown in LB with or without 1% (wt/vol) KNO₃ under aerobic and anaerobic conditions as described in the text.

^bNitrate reductase activities were determined in cell extracts by using reduced methyl viologen as the electron donor. Values shown are means ± standard errors as calculated from three independent cell suspensions.

Posttranslational oxygen regulation of nitrate reduction.

Through quantitative reverse transcription-PCR and methyl viologen-linked respiratory nitrate reductase activity, we have shown that oxygen inhibition of transcription and translation of the nar operon was overcome by modulating transcription of narK2GHJI through P_LAC. The use of pREP also allowed us to express nitrate reductase activity anaerobically in JVC over a wide range from 218 ± 7 to 1,120 ± 150 nmol nitrite min⁻¹ mg⁻¹ protein. More importantly, the levels of enzyme activity in JVC with pREP under anaerobic conditions are similar to those observed aerobically in JVC alone (218 ± 7 and 210 ± 9 nmol nitrite min⁻¹ mg⁻¹ protein, respectively). These expression levels allowed us to measure whole-cell nitrate reduction to nitrite aerobically and anaerobically in strains with comparable nitrate reductase activities and thus explore posttranslational regulation by oxygen (Table 3). We examined JVC for physiological whole-cell nitrate reduction (non-methyl viologen driven) to nitrite indirectly by measuring the levels of nitrite in the growth medium. Specifically, the supernatant was collected by centrifugation (10,000 × g, 4°C, 10 min) and measured for nitrite (16). The disappearance of nitrite was not considered since previous studies have shown that the presence of nitrate inhibits the reduction of nitrite and that its subsequent secretion by whole cells is stoichiometrically correlated to the disappearance of nitrate (5, 13). The JVC strain excreted nitrite when grown aerobically (1.5 ± 1 nmol nitrite μg⁻¹ protein [Table 4]) while the levels of aerobic nitrite excretion in wild-type PAO1 remained undetectable. During anaerobic growth, the JVC strain containing pREP excreted nitrite at a rate approximately sixfold higher than the levels observed under aerobic conditions in the same strain without pREP (9.4 ± 0.5 and 1.5 ± 0.1 nmol nitrite μg⁻¹ protein, respectively [Table 4]). Although similar nitrate reductase activities were observed in aerobic JVC and anaerobic JVC with pREP (Table 3), the levels of nitrite excretion were significantly lower under aerobic conditions. The P_LAC insertion enabled JVC to aerobically transcribe nar and produce functional wild-type levels of respiratory nitrate reductase activity under aerobic conditions, thus clearly demonstrating the physiological significance of posttranslational regulation by oxygen at the level of nitrate transport and/or diversion of electron flow (4, 10, 11).

TABLE 4.

Nitrite excretion levels of P. aeruginosa PAO1 and JVC

Straina	Nitrite excretion (nmol nitrite μg−1 protein)b
	Aerobic	Anaerobic
PAO1	<0.1	131.1 ± 5.6
PAO1 + pREP	<0.1	128.0 ± 9.9
JVC	1.5 ± 0.1	123.6 ± 4.7
JVC + pREP	0.5 ± 0.1	9.4 ± 0.5

^aPAO1 and JVC were grown aerobically and anaerobically in LB with 1% (wt/vol) KNO₃ as described in the text.

^bNitrite levels were determined as described in the text. Values shown are means ± standard errors as calculated from three independent measurements.

The results in this investigation are consistent with those of past studies of transcriptional regulation of nar (1, 2, 8, 27, 39-41) that were recently challenged by a DNA microarray-based study (7) which reported no significant changes in narG transcription aerobically versus anaerobically in P. aeruginosa grown on complex medium supplemented with nitrate. The authors of that recent study concluded that nitrate alone was sufficient for nar expression in both aerobic and anaerobic environments.

Finally, the modification of pUCGMlox allowed us to modulate the expression of the narK2GHJI operon for physiological studies which would otherwise be difficult to perform by utilizing plasmid expression vectors currently available for P. aeruginosa. With the combination of pREP and P_LAC, we were able to conditionally express nitrate reductase activities chromosomally. Intermediate levels of nitrate reductase activities can even be obtained using IPTG (isopropyl-β-d-thiogalactopyranoside) (data not shown). Therefore, the use of this technique enables the investigator to integrate a regulatory element(s) and modulate gene expression of operons and multiple open reading frames to further take advantage of the available P. aeruginosa genome database (34).