Doc-mediated cell killing in Shigella flexneri using a C1/LacI controlled expression system

Abstract

In this report we describe the development of a highly stringent and dually regulated promoter system for Shigella flexneri. Dual regulation was provided by utilizing a promoter susceptible to control by the bacteriophage P1 temperature-sensitive C1 repressor that in turn was under the transcriptional control of LacI. The level of induction/repression ratios observed was up to 3700-fold in S. flexneri. The general utility of this promoter system was evaluated by demonstrating that the bacteriophage P1 post-segregational killer protein Doc mediates a bactericidal effect in S. flexneri. This represents the first report of Doc (death on curing)-mediated killing in this Gram-negative species.

Regulated gene expression, Vectors, Bacteriophage P1, Post-segregational killer protein, Shigella flexneri

1 Introduction

Shigella species are capable of causing acute, debilitating diarrheal disease in humans. While Shigella dysenteriae causes the most severe diarrheal illness reflected in high mortality rates, Shigella flexneri remains the leading cause of shigellosis in most of the developing world 1,2]. To aid bacterial virulence gene studies, it is desirable to use a regulated promoter system to study the expression or depletion of a particular gene product. However, while a wide variety of regulated promoter systems exist for use in Escherichia coli3,4], few systems have been described in bacteria other than E. coli. Moreover, the range of regulation and level of stringency using analogous systems adapted to other species are inferior as compared to E. coli[5].

Bacteriophage P1 lysogenizes E. coli in a stable fashion, in part, due to the plasmid addiction system that kills plasmid-free segregants via a toxin known as Doc (death on curing;[6]). In E. coli, Doc-mediated post-segregational killing requires the antitoxin/toxin system MazEF[7]. As mazEF is chromosomally encoded and activated by starvation conditions, it has been suggested that this system may play a role in programmed cell death 8,9]. In silico analysis has identified orthologous systems in both Gram-negative and -positive species which suggests mazEF may be conserved among prokaryotes[10].

In this report, we describe the development of a regulated promoter system that exhibits a similar range of regulation, and a high level of stringency irrespective of its use in either E. coli or S. flexneri. To control gene expression, we placed the lacZ reporter system under the control of a promoter regulated by the temperature-sensitive C1 repressor from the broad-host-range bacteriophage P1. c 1 was placed under the transcriptional control of LacI, thereby providing a dual means of regulation by varying both the temperature and concentration of IPTG. Using the C1/LacI-regulated promoter system to control expression of the bacteriophage P1 post-segregational killer protein Doc, we demonstrate that Doc mediates a bactericidal effect in S. flexneri.

2 Materials and methods

2.1 Strains and transformation

E. coli XL1-Blue MRF′ and DH5α were obtained from Stratagene and Gibco-BRL, respectively. S. flexneri 12022 was obtained from the American Type Culture Collection. Both species were grown in Luria–Bertani (LB) medium supplemented as needed with 25 μg ml⁻¹ chloramphenicol and 50 μg ml⁻¹ kanamycin. E. coli was transformed according to Sambrook et al.[11] and S. flexneri was transformed by the method of Lederberg and Cohen[12].

2.2 Construction of reporter plasmids

The reporter plasmids were constructed in the vector pAM401, which contains both a chloramphenicol and tetracycline resistance marker[13]. The presence of both a p15A and pGB354 replicon allows the vector to replicate in both enteric Gram-negative bacteria and many different Gram-positive species[13]. Enzymes were used according to the manufacturers’ instructions and cloning was performed as described by Sambrook et al.[11]. The lacZ gene was PCR amplified using the plasmid pMC1871 (Pharmacia) as template with the upstream primer AGGACGGTCGACGGAGGTGTAGTATGGTCGTTTTACAACGTCG and downstream primer TCCTCCGCATGCTCCCCCCTGCCCGGTTAT which contained a Sal I and Sph I site (underlined sequence) and cloned into the respective sites of pAM401. The upstream primer also contained a ribosome binding site (5′-GGAGG-3′) positioned upstream of a start codon (bold) to initiate translation. Where indicated in the text, the doc gene was PCR amplified from the thermoinducible bacteriophage P1Cm carrying the c 1.100 mutation (kindly provided by Dr. Michael Yarmolinsky) using the upstream primer CAGAAGGTCGACTAGTTGTTTATGAGGCATATATC and downstream primer TCGATAGCATGCCTACTCCGCAGAACCATACAATC and cloned into the Sal I/Sph I sites. To prevent read-through from cryptic promoters and runaway transcription, the transcriptional terminators rrnB T1T2[14] and TL₁₇[15] were cloned 5′ (Sac II site) and 3′ (Eco RV site) of the expression cassette, respectively. The Pro1 promoter was derived from bacteriophage P1 and consists of two partially overlapping C1 operators[16]. The underlined sequences illustrate the C1 repressor binding sites and the proposed −35/−10 promoter elements are shown in bold (5′-TATATTGCTCTAATAAATTTATTAGTGTAATATCGCCTCAATG-3′). The Pro1 promoter was obtained by annealing complementary oligonucleotides and cloned into the Sal I site in the same orientation relative to lacZ. To control gene expression, the temperature-sensitive C1 repressor[17] under the transcriptional control of a LacI-regulated promoter (Pro2), was PCR amplified using the plasmid pBHRC1[16] as template with the upstream primer TCGTGCGGATCCATTGTAATACGACTCACTATAGG and downstream primer GTAGTAGCATGCGGTGAGCAAACAGCCAT and cloned into the Bam HI and Sph I sites. The LacI-regulated promoter contained a lac operator site (underlined sequence) flanked by consensus E. coli−35/−10 hexamers (bold sequence) 5′-AATTGACATTGTGAGCGGATAACAATATAATGTGTGGAAGCT-3′. The final lacZ expression vector was designated pAM402 (Fig. 1A).

1

Construction and map of the lacZ reporter plasmid and lacI expression plasmid. A: pAM402 reporter plasmid. To stop read-through from cryptic promoters and to prevent runaway transcription, the transcriptional terminators rrnB T1T2[14] and TL₁₇[15] were cloned 5′ and 3′ of the expression cassette, respectively. The lacZ gene was placed under the control of the Pro1 promoter (arrows denote direction) that consists of two partially overlapping C1 operator sites[16]. To control gene expression, the temperature-sensitive C1 repressor[17] was placed under the transcriptional control of a LacI-regulated promoter (Pro2). Where indicated in the text, lacZ was excised and doc was cloned into the respective sites. The plasmid pAM402 contains the p15A origin of replication, the origin of replication derived from pGB354 and the chloramphenicol (Cm) resistance markers from pACYC184 and pGB354[13]. B: pBHRlacI expression plasmid. The lacI gene was cloned into the chloramphenicol resistance gene of the broad-host-range plasmid pBBR122 (Mobitech). Transcriptional expression of lacI therefore relied on either cryptic promoters in the plasmid and/or the promoter driving the chloramphenicol resistance gene. The broad-host-range plasmid pBBR122 contains genes encoding the mobilization protein (mob), replication protein (rep) and kanamycin resistance marker (kan). The sequences of both plasmids are available on request.

To construct a LacI expression plasmid, the lacI gene was PCR amplified using E. coli DH5α as template with the upstream primer CGAATTGGATCCGGAGGTGGAATGTGAAACCAGTAACG and downstream primer CGGCGGAATTCCTAATGAGTGAGCTAACT. The upstream primer contained a ribosome binding site upstream of the ATG start codon (bold) to initiate translation. The PCR-generated fragment was then cloned into the Dra I site within the chloramphenicol resistance gene of the broad-host-range plasmid pBBR122 (Mobitech, Fig. 1B). Transcriptional expression of lacI therefore relied on either cryptic promoters in the plasmid and/or the promoter driving the chloramphenicol resistance gene.

2.3 Northern hybridization

RNA from both E. coli and S. flexneri was prepared using Qiagen's RNeasy kit according to the manufacturers’ instructions. RNA (5 μg) was electrophoresed on a denaturing formaldehyde gel[11] and vacuum blotted onto Duralon UV membrane (Stratagene). Hybridization was performed in 5×Denhardts’ solution, 0.5% SDS, 5×SSPE, 100 μg ml⁻¹ salmon sperm and 45% formamide at 45°C using a Sal I/Sph I-generated lacZ fragment randomly prime labeled (Roche) with [α³²P]dCTP. Washing was performed at 50°C at a final stringency of 0.25×SSPE and 0.1% SDS. To ensure equal loading of the RNA, the membrane was then stripped by incubation in a boiling solution of 0.1×SSC and 0.1% SDS, and reprobed with a ³⁵S-tailed (Roche) oligonucleotide (5′-ACTTTATGAGGTCCGCTTGCTCTCGC-3′) complementary to both E. coli and S. flexneri 16s rRNA. Hybridization was performed in 1×Denhardts’ solution, 4×SSC, 50 μg ml⁻¹ poly(A), 500 μg ml⁻¹ salmon sperm, 10% dextran sulfate, 10 mM DTT, and 45% formamide at 37°C. Washing was performed at 37°C at a final stringency of 0.5×SSC and 0.1% SDS.

3 Results and discussion

3.1 Characteristics of the reporter plasmid

The reporter system was placed under the transcriptional control of a promoter (Pro1) that contained two partially overlapping C1 operator sites[16]. To control gene expression, the temperature-sensitive C1 protein from bacteriophage P1 was used[17]. This promoter system has been shown to function well in E. coli but to a much lesser extent in S. flexneri, primarily due to the inability to achieve derepression at elevated temperatures[16]. To circumvent this problem, the c 1 gene was placed under the transcriptional control of a LacI-regulated promoter, thereby providing a dual means of regulation in species that express LacI. As S. flexneri lacks a functional lacI homolog, a lacI expression plasmid was constructed and, where indicated, was co-transformed with the lacZ reporter plasmid into S. flexneri. At low temperatures and in the presence of IPTG, C1 is expressed and is thermally stable which in turn switches off the expression of the reporter, lacZ. At elevated temperatures and in the absence of IPTG, C1 is switched off and is thermally unstable which results in LacZ expression.

3.2 Analysis of β-Gal activity in E. coli and S. flexneri

To demonstrate the functionality of the dual promoter system, the activity of the protein produced by the lacZ gene (β-Gal activity) was measured in E. coli DH5α (lacI) and XL1-Blue MRF′ (lacI^q) that express and over-express LacI, respectively. Since the promoter driving c 1 contained consensus −35/−10 hexamers, it was expected that the construct would produce an excess of C1 resulting in the efficient repression of the C1-regulated promoter but might only result in the partial derepression at elevated temperatures. In support of this hypothesis basal expression in DH5α was below the limits of detection of the assay and upon induction at elevated temperature, only a modest level of induction was observed (Table 1). In contrast, basal expression in XL1-Blue MRF′ cells was extremely high suggesting that the expression of the chromosomally encoded and over-expressed LacI was effectively switching off c 1 expression. Upon addition of IPTG a dramatic decrease in β-Gal expression was observed to levels nearly undetectable by the assay. Furthermore, following exposure to IPTG at low temperature, high levels of induced expression were achieved after only 100 min induction. Therefore, the results indicate it was possible to achieve low levels of basal expression, and high-induced activity using a combination of C1 to control lacZ expression, and LacI to control levels of c 1 produced.

1

Basal and induced activities of lacZ fusions to the C1-regulated promoter in E. coli strains DH5α and XL1-Blue MRF′

Construct	Strain (lacI status)	IPTG (mM)	Activity (Miller units)
			Basal (31°C)	Induced (42°C)
Control	DH5α (lacI)	0	<0.5	<0.5
pAM402	DH5α (lacI)	0	<0.5	11(0.3)
Control	XL1-Blue (lacIq)	0	<0.5	<0.5
pAM402	XL1-Blue (lacIq)	0	471(24)	1578(26)
pAM402	XL1-Blue (lacIq)	2	<0.5	0.5(0.1)
pAM402	XL1-Blue (lacIq)	0.2	<0.5	84(17)
pAM402	XL1-Blue (lacIq)	0.06	<0.5	617(47)

Overnight cultures grown at 31°C at the stated concentration of IPTG, were diluted 1:100 and grown to an OD₆₀₀ of approximately 0.15 in LB under the same conditions. Cells were collected at 2500×g for 10 min at room temperature and resuspended in fresh LB. Cultures were divided equally and incubated at 31°C with IPTG at the same concentration or at 42°C without IPTG for 100 min (OD₆₀₀ approximately 0.6). The control strain carried a plasmid containing a promoterless lacZ gene. Miller units[18] are averages of results for multiple cultures (n=3) followed by the standard deviation in parentheses. <0.5 indicates below the limits of detection for the assay.

The functionality of the dual expression system was tested in S. flexneri. As S. flexneri does not contain a functional homolog of LacI, it was supplied in trans from a lacI expression plasmid (Fig. 1B, pBHRlacI). Since an insufficient intracellular concentration of LacI would have little effect on c 1 expression, and an intracellular excess of LacI might generate leakiness from the C1-regulated promoter, a number of different lacI expression plasmids were constructed and evaluated in order to find the optimal concentration of LacI required for control of the desired transcriptional elements (data not shown). In the absence of LacI at low temperatures, β-Gal activity in S. flexneri was below the limits of detection with only modest induction observed at the elevated inducing temperature (Table 2). Co-transformation of both the lacZ and lacI expression plasmids however, resulted in a dramatic increase in basal expression that could be regulated to concentrations below detectable limits by the addition of IPTG. Furthermore, high levels of induced expression were achieved by the elevation of temperature and the titration of IPTG (Table 2). This level of induced expression was significantly higher using the dual C1/LacI-regulated promoter system as compared to the system regulated by C1 alone. Because the basal expression levels was below the limit of detection of the standard colorimetric assay for β-Gal, the activity of the enzyme was also measured using a chemiluminescent substrate in order to determine the level of expression from the regulated genetic elements (Table 2). The activity observed ranged from 2.1×10⁴ U during basal conditions to 8.1×10⁷ U under induced conditions. This represented an approximate 3700-fold range of regulation. This observation is similar if not better than regulated promoter systems described for E. coli3,4] and far superior to the previously described promoter system in S. flexneri[16].

2

Basal and induced activities of lacZ fusions to the C1-regulated promoter in S. flexneri

Construct	LacI repressor	IPTG (mM)	Activity
			Basal (31°C)		Induced (42°C)
			Miller units	R.L.U.	Miller units	R.L.U.
Control	–	0	<0.5	nd	<0.5	nd
pAM402	–	0	<0.5	nd	18(0.3)	nd
pAM402a	+	0	324(30)	7.7×107	392(44)	8.7×107
pAM402a	+	1	<0.5	9.8×103b	283(4)	7.9×107
pAM402a	+	0.2	<0.5	2.1×104	317(24)	8.1×107
pAM402a	+	0.06	0.8(0.3)	3.1×105	303(6)	7.4×107

Overnight cultures grown at 31°C at the stated concentration of IPTG, were diluted 1:100 and grown to an OD₆₀₀ of approximately 0.1 in LB under the same conditions. Cells were collected at 2500×g for 10 min at room temperature and resuspended in fresh LB. Cultures were then divided equally and incubated at 31°C with IPTG at the same concentration or at 42°C without IPTG for 80 min (OD₆₀₀ approximately 0.6). The control strain carried a plasmid containing a promoterless lacZ gene. Miller units[18] are averages of results for multiple cultures (n=3) followed by the standard deviation in parentheses. Where indicated, lysates were also measured using the galacto-star chemiluminescent reporter gene assay (Applied Biosystems) and are presented as relative light units (R.L.U.)/OD₆₀₀ of culture. <0.5 indicates below the limits of detection for the assay. nd, not determined.

^aDenotes S. flexneri co-transformed with the lacI expression plasmid.

^bDenotes below the linear range of the luminometer.

3.3 Northern analysis of lacZ expression from E. coli and S. flexneri

To analyze the regulation of lacZ expression at the transcriptional level, Northern blot analysis was performed. RNA was prepared from cultures carrying promoterless lacZ constructs and from cultures carrying the reporter plasmids under repressed and derepressed conditions. Transcripts were not detected from control cultures or from cultures prepared under repressed conditions using lacZ as a probe for either S. flexneri or E. coli (Fig. 2, lanes 1–7). In contrast, under induced conditions, transcripts were detected from both S. flexneri and E. coli harboring the reporter constructs (Fig. 2, lanes 4 and 8). Thus, Northern analysis confirmed that the regulation of lacZ expression occurs primarily at the transcriptional level and suggests that the promoter system is tightly repressed.

2

Northern analysis of lacZ expression in E. coli and S. flexneri. Overnight cultures were diluted 1:100 and grown to an OD₆₀₀ of approximately 0.15 in LB containing 1 mM IPTG (S. flexneri, lanes 1–4) or 60 μM IPTG (E. coli, lanes 5–8) at 31°C. Cells were collected at 2500×g for 10 min at room temperature and resuspended in fresh LB. Cultures were then divided equally and incubated at 31°C with additional IPTG (repressed, lanes 1, 3, 5 and 7) or at 42°C without IPTG (induced, lanes 2, 4, 6 and 8) for 90 min. Control cultures (lanes 1, 2, 5 and 6) carried a promoterless lacZ construct while the test cultures (lanes 3, 4, 7 and 8) carried the lacZ/lacI expression plasmids.

3.4 C1 controlled doc expression in S. flexneri

Bacteriophage P1 lysogenizes E. coli in a stable fashion, in part, due to the plasmid addiction system that kills plasmid-free segregants. As bacteriophage P1 lysogenizes a wide variety of Gram-negative bacteria including Shigella species[19], it was postulated that Doc would also be functional in S. flexneri. To test this hypothesis and to demonstrate the utility of the regulated promoter system in S. flexneri, Doc was placed under the control of the C1-regulated promoter. No difference in the growth of the cultures harboring the doc expression plasmid was observed upon induction using temperature shift alone (data not shown). However, when the same cultures carrying the doc expression plasmid were co-transformed with the lacI expression plasmid, induction using a temperature shift in the absence of IPTG resulted in growth arrest (Fig. 3A). This indicated that LacI was required to switch off c 1 expression in order to achieve sufficient levels of Doc. Intriguingly, expression of the E. coli toxic protein Gef[20] did not mediate growth inhibition under the same conditions (data not shown).

3

A: Effect of Doc expression on the growth of S. flexneri. Overnight cultures, grown under repressed conditions (31°C, 1 mM IPTG), were diluted 1:100 and grown for 130 min under identical conditions. Cells were collected at 2500×g for 10 min at room temperature and resuspended in fresh LB. Cultures harboring the doc/lacI expression plasmids, were then divided equally and incubated at 31°C with additional IPTG (•) or at 42°C without IPTG (◯). Control cultures harboring the lac Z/lacI plasmids were also grown under both repressed (█) and induced conditions (□). Figure depicts representative numbers from three separate experiments. Arrows denote time points at which samples were taken to determine viable counts. B: The ability of S. flexneri to recover from Doc expression. Samples from cultures harboring the doc/lacI expression plasmids (□) were taken at 0 and 80 min induction (A, arrows) and plated in triplicate (values reported are averages±standard deviation) onto selective medium and grown under repressed conditions (31°C, 1 mM IPTG). As a control, the number of colony forming units were also measured for cultures harboring the lacZ/lacI plasmids (█) incubated under the same conditions.

In E. coli, Doc is postulated to trigger cell death via the activation of the programmed cell death antitoxin/toxin module MazEF by the inhibition of protein synthesis[7]. To investigate whether Doc exerts a bacteristatic or bactericidal effect in S. flexneri, cultures where plated out immediately prior to induction and after 80 min induction, and were allowed to recover overnight under repressed conditions (31°C, 1 mM IPTG). This resulted in a 10⁴ reduction in the number of colony forming units (Fig. 3B). A reduction in colony forming units was not observed for the control cultures. These data suggest that Doc exerts a bactericidal effect in S. flexneri. Although the target of Doc is unknown, as P1 can lysogenize a wide variety of Gram-negative species, it is not unreasonable to speculate that the target of Doc may be conserved. In silico analysis has identified mazEF orthologs in both Gram-negative and -positive bacteria[10] leading to the possibility that Doc-mediated cell death by mazEF may also occur in species other than E. coli.

Acknowledgements

This work was supported by Hexal Gentech Forschungs GmbH. DNA sequencing data was obtained by the Biotechnology Resource Laboratory of the Medical University of South Carolina.