Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2013 Apr 1.
Published in final edited form as: FEMS Microbiol Lett. 2012 Feb 8;329(1):28–35. doi: 10.1111/j.1574-6968.2012.02499.x

CsrA modulates luxR transcript levels in Vibrio fischeri

Joshua W Williams 1, A L Ritter 2, Ann M Stevens 1,*
PMCID: PMC3298366  NIHMSID: NIHMS350840  PMID: 22250984

Abstract

The quorum-sensing and CsrA regulons of Vibrios control overlapping cellular functions during growth. Hence, the potential exists for regulatory network interactions between the pathways that enable them to be coordinately controlled. In Vibrio cholerae, CsrA indirectly modulates the activity of LuxO in the quorum-sensing signaling pathway. In this study, it was demonstrated that in Vibrio fischeri, CsrA causes an increase in the transcript levels of a downstream quorum-sensing regulatory gene, luxR, which does not exist in the V. cholerae system. In V. fischeri, the increase in luxR transcripts caused by CsrA does not depend on the LitR transcriptional activator nor does the CsrA effect seem to occur through the global regulator cAMP-CRP. Thus there appears to be more than one mechanism whereby the CsrA and quorum-sensing pathways integrate regulatory outputs in Vibrios.

Keywords: quorum sensing, factorial design, CsrB, LitR, CRP

Introduction

The quorum-sensing response of Vibrio fischeri involves a complex signal transduction pathway that regulates many cellular processes including bioluminescence, host-association, certain metabolic functions, and motility (Fidopiastis, 2002; Lupp et al., 2003; Visick, 2005; Waters & Bassler, 2005; Studer et al., 2008). Many of the major regulatory genes in the quorum-sensing regulon have been identified and characterized through mutagenesis in V. fischeri or analysis of function studies in recombinant Escherichia coli (Engebrecht, 1984; Dunlap, 1985; Lupp et al., 2003) (Fig. 1). Much of the work on this system has focused on understanding interactions that lead to drastic changes in gene expression, such as a hyper-luminescent response, or a completely dark response. However there are potentially important interactions that may remain to be discovered. In a complicated regulatory network, where there are many downstream components and multiple pathways functioning coordinately, even a small change in the expression of one component can potentially lead to much larger differences in others. In this paper both standard laboratory experiments as well as the statistical technique of factorial design, based on analysis of variance (ANOVA), were applied to facilitate study of potentially subtle interactions between the quorum-sensing and CsrA networks of V. fischeri.

Figure 1. The quorum-sensing pathway of V. fischeri.

Figure 1

Open in a new tab

In V. fischeri, AinS produces N-octanoyl-L-homoserine lactone, which is thought to be sensed by AinR. Based on the V. harveyi model, LuxS produces (2S,4S)-2-methyl-2,3,3,4-tetrahydroxytetrahydrofuran-borate which is detected by LuxP. LuxI is known to produce N-3-oxo-hexanoyl-L-homoserine lactone (AHL), the ligand for the lux operon activator LuxR in V. fischeri. According to the V. harveyi model, the LuxU/O phosphorelay stimulates production of a sRNA that represses LitR at low cell density; at high cell density the sRNA is not expressed so that LitR is produced. See (Visick, 2005) for more details on this pathway.

Since the quorum-sensing response of V. fischeri leads to activation of luminescence, which consumes a large amount of cellular energy and oxygen (Bose et al., 2008), there are mechanisms in place that regulate the response based on the metabolic state of the cell. For example, the secondary metabolism regulatory complex cAMP-CRP activates transcription of luxR (Dunlap, 1985; Dunlap, 1988) whereas the redox sensitive regulator ArcA represses both luxR and the lux operon (Bose et al., 2007). While this links metabolism with quorum sensing, there may be additional points of convergent regulation. It was hypothesized that the global regulatory RNA-binding protein CsrA may have some role in controlling the quorum-sensing response in relation to the metabolic state of the cell. CsrA is an important component in regulating carbon storage and utilization in the cell during exponential-growth phase (Liu et al., 1995; Romeo, 1998; Baker et al., 2002), which is the point where the quorum-sensing response is induced. CsrA has also been shown to play a regulatory role in the quorum-sensing response of other Vibrio species (Lenz et al., 2005; Jones et al., 2008). For example, in V. cholerae, CsrA is regulated by three sRNAs (CsrB, CsrC and CsrD) and it in turn indirectly affects the activity of LuxO (Lenz et al., 2005). In V. fischeri, CsrA is regulated by two sRNAs (CsrB1 and CsrB2) (Kulkarni et al., 2006), but its interaction with the quorum-sensing system is unknown. In this study possible connections between CsrA and quorum sensing were probed by examining the influence of CsrA levels on the luminescence output of wild-type and mutant strains of V. fischeri.

Materials and methods

Strains and growth conditions

Strains and plasmids are described in Table 1. E. coli strains were grown with aeration at 37°C in Luria-Bertani broth. V. fischeri strains were grown with aeration at 30°C in RMS minimal medium (2% casamino acids, 1X M9 salts [12.8 g Na2HPO4 7H2O, 3 g KH2PO4, 0.5 g NaCl, and 1 g NH4Cl per liter], 0.4% glucose, 0.1% MgCl2, 15 g NaCl per liter); no serious growth defects were observed using these conditions. Ampicillin (Ap) (50 or 100 µg/ml), kanamycin (Km) (50 µg/ml), cAMP (5 mM), or N-(β-ketocaproyl)-L-homoserine lactone (AHL) (20 nM) were added to media as specified.

Table 1.

Bacterial strains, plasmids, and primers used in this study

Relevant Description Reference
V. fischeri
ES114 Wild-type strain of V. fischeri (Boettcher & Ruby, 1990)
PMF8 ES114 ∆litR, Kmr (Fidopiastis et al., 2002)
E. coli
DH5αλpir F- endA1 glnV44 thi-1 recA1 relA1
gyrA96 deoR nupG Φ80dlacZΔM15
Δ(lacZYA-argF)U169, hsdR17(rK− mK+),
λpir 		(Hanahan, 1983)
Top10 F- mcrA Δ(mrr-hsdRMS-mcrBC)
φ80lacZΔM15 ΔlacX74 nupG recA1 araD139
Δ(ara-leu)7697 galE15 galK16 rpsL(StrR)
endA1 λ− 		(Invitrogen)
Plasmids:
pKK223-3-CsrA Apr, encodes Ptac-csrA expression cassette (Kulkarni et al., 2006)
pKK223-3-CsrB1 Apr, encodes Ptac-csrB1 expression cassette (Kulkarni et al., 2006)
pBBRMCS2 Kmr, broad-host-range cloning vector (Kovach et al., 1995)
pGEM-T Apr, lacZ-MCS, used in TA-cloning (Promega)
pVSV104 Kmr, R6K ori, pES213 ori, RP4 oriT,
lacZα-(SphI-AvrII-HpaI-SalI-NcoI-Kpn-SacI)		 (Dunn et al., 2006)
pEVS104 RP4-based conjugal helper plasmid (Stabb & Ruby, 2002)
pJW3 Ptac-csrA expression cassette from
pKK223-3-CsrA cloned into pVSV104		 This study
pJW4 Ptac-csrB1 expression cassette from
pKK223-3-CsrB1 cloned into pVSV104		 This study
Primers:
PtacUP1 5’GGTACCGGAGCTTATCGACTGCACG3’ This study
PstcsrB1right 5’GTTCTGCAGAAAAACCCCACCAAGCTCTC3’ (Kulkarni et al., 2006)
RT-luxRF 5’TGGCAGCGGTTAGTTGTATTG3’ (Williams et al., 2008)
RT-luxRR 5’TAGCGTGGGCGAGTGAAG3’ (Williams et al., 2008)
RT-crpF 5’TTTCTTATTGATGGGTTTTGTCATTC3’ This study
RT-crpR 5’AACCCAATCTCCTTTCCAATAAAAT3’ This study
gRT-Vf16SF 5’GGGTTAAGTCCCGCAACGA3’ This study
RT-Vf16SR 5’CCATTACGTGCTGGCAAACA3’ This study
RT-CsrAF 5’ATGCTAATTTTGACTCGCCGTGTAG3’ This study
RT-CsrAR 5’GGTGTACCTTTTTCCGCTTGAATGC3’ This study
Open in a new tab

DNA manipulation

Standard molecular biology techniques for DNA cloning and manipulation were used for all cloning steps. PCR purification, gel extraction, and plasmid purification kits were obtained from Qiagen.

Creation of CsrA and CsrB1 overexpression plasmids for use in V. fischeri

The Ptac-csrA expression cassette from pKK223-3-CsrA (Kulkarni et al., 2006) was removed by digestion at the HindIII-BamHI sites and ligated into vector pBBRMCS2 (Kovach et al., 1995) digested with the same enzymes. A KpnI-SacI fragment from this intermediate construct was then ligated into pVSV104 (Dunn et al., 2006), which had also been digested with KpnI-SacI, to create pJW3. The Ptac-csrB1 expression cassette from pKK223-3-csrB1 (Kulkarni et al., 2006) was PCR amplified with Deep Vent DNA polymerase using primers PtacUP1 and PstcsrB1right (Table 1). Adenine was added to the end of the PCR product using Taq polymerase, and it was then ligated into pGEM-T (Promega). The Ptac-csrB1 expression cassette was removed from pGEM via SalI-SphI digestion, and ligated into pVSV104, which had been digested in the same way, to create pJW4. The integrity of pJW3 and pJW4 were confirmed by sequencing. pJW3 or pJW4 were used to transform E. coli DH5αλpir and were subsequently introduced into V. fischeri ES114 or PMF8 via tri-parental conjugation using the helper strain E. coli (pEVS104) (Stabb & Ruby, 2002).

To introduce pJW3 or pJW4 (KmR) into PMF8 (KmR), Ap (50 µg/ml) was added to the selection plates to select against the E. coli donor with no impact on V. fischeri. Presence of the vector in PMF8 was verified by plasmid purification followed by PCR to amplify the expression cassette. pVSV104-based vectors are known to be stably maintained in V. fischeri without antibiotic selection (Dunn et al., 2006). To confirm this, plasmid stability was examined by growing KmR strains without selection for 3 days followed by plasmid isolation and PCR screening.

Experimental design

Two methods of experimental design were employed in this study to enable a side-by-side comparison of the approaches. All experiments were performed using standard laboratory set-ups (at least 2 independent experiments assayed in triplicate). In addition, factorial design was simultaneously used to test the efficacy of this approach for laboratory-based studies (where it has not been commonly adopted). The design and analysis of the factorial experiments were done using the statistical application program Design Expert from Stat-Ease (Minneapolis, MN). For all experiments, data collection was done in random order to minimize systematic error from uncontrolled factors such as drift in measurement instruments. The ANOVA analysis allows identification of statistically significant model terms, based on p-values, that will be included in the multivariate regression analysis of the response variables (luminescence and transcript levels).

Assays for V. fischeri luminescence

V. fischeri strains were grown to mid-exponential phase (OD600 nm 0.6). Once this OD was reached, 200 µl samples were taken and added in triplicate to a white 96-well microtiter plate for luminescence readings. Data was collected on a Beckman-Coulter LD400 luminometer, with an integration time of 1 sec per well, and with the photometer wavelength set to 492 nm.

Quantitative RT-PCR analysis

V. fischeri was grown as described above, and 500 µl cell samples were collected. Qiagen RNAprotect Bacteria Reagent was used to stabilize RNA in cell pellets prior to storage at −70°C. Total RNA was extracted using a Qiagen RNeasy mini kit and stored at −70°C. RNA was analyzed for integrity and concentration using a Bio-Rad Bioanalyzer, and converted to cDNA using an Applied Biosystems High-Capacity cDNA Reverse Transcription Kit. cDNA samples were stored at −20°C until use as templates in an Applied Biosystems 7300 Real-Time PCR system. Primers used in the analysis of csrA, luxR, and crp, transcripts are in Table 1. 50 ng of cDNA was the template for the RT-PCR reaction, with primer concentrations of 250 µM. 2x SYBR Green master mix (Applied Biosystems) and H2O were added to a final reaction volume of 50 µl/well in a MicroAmp Optical 96-well reaction plate (Applied Biosystems). Thermal cycler settings were programmed for 52°C for 2 min, 95°C for 10 min, then 45 cycles of the following: 95°C for 15s, 51°C for 15s, and 60°C for 1 min, which was the data collection point.

Results and discussion

Modulation of CsrA levels

Ideally, a csrA partial deletion strain would have been used for experiments as has been possible in other systems (Liu et al, 1995; Lenz et al, 2005). However repeated attempts failed to generate the desired construct. Therefore an alternative strategy was employed to modulate CsrA levels whereby either csrA (pJW3) or csrB1 (pJW4) was overexpressed from a stable plasmid construct in two V. fischeri strains, ES114 (wild type) and PMF8 (ΔlitR). This approach was followed, because in factorial design just two levels of each experimental factor are permitted and they should be as far apart from one another as possible. A 20 nM level of AHL was chosen for experiments because it permitted for detection of luminescence from ES114 strains without fully saturating the system. The amount of csrA transcript was measured to ensure that there were significantly different levels expressed from pJW3 and pJW4. As anticipated, there were higher levels of csrA transcripts in cells overexpressing csrA (pJW3) in the presence of 20 nM AHL in comparison to the cells overexpressing csrB1 (pJW4) (Fig. 2). Further, because CsrB1 post-translationally sequesters CsrA (Timmermans & Van Melderen, 2010; Romeo, 1998), the actual decrease in the cellular activity of CsrA in strains overexpressing csrB1 is likely greater than what is observed by simply measuring differences in csrA transcript levels.

Figure 2. Quantitative RT-PCR analysis of csrA transcripts via factorial design.

Figure 2

Open in a new tab

The levels of csrA transcripts in wild-type ES114 background (panel A) or PMF8 ΔlitR background (panel B) with overexpression of csrB1 (pJW4) or overexpression of csrA (pJW3) in the presence of 20 nM AHL. Data was obtained from two independent triplicate sets. The mean values, represented by the two columns in each panel, were obtained by fitting the entire data set using multivariate regression. The error bars represent the 95% confidence intervals for the means.

Probing the relationship between CsrA and the quorum-sensing pathway upstream of luxR

The V. fischeri ES114 (wild type) and PMF8 (ΔlitR) strains carrying pJW3, pJW4, or the control pVSV104 were next examined for luminescence expression. LitR was chosen as the quorum-sensing factor to be examined due to the fact that it is a critical link between the upstream quorum-sensing regulatory network, and the downstream luminescence response regulated by LuxR (Fig. 1). The level of luminescence in the wild-type strain V. fischeri ES114 was independent of the expression level of CsrA (over the range studied) (Fig. 3A). In contrast, the ΔlitR strain of V. fischeri (PMF8) produced the lowest level of luminescence when CsrA activity was depressed (strain PMF8 (pJW4)), an intermediate level for the control (strain PMF8 (pVSV104)), and the highest level when csrA was overexpressed (strain PMF8 (pJW3) (Fig. 3B). The results showed that there was a significant interaction between litR and the CsrA level (p < 0.0001). Thus CsrA did not affect the luminescence level in V. fischeri ES114 (Fig. 3A), but in the absence of LitR luminescence was dependent on CsrA (Fig. 3B). The upstream quorum-sensing regulatory network functions to control the expression of litR, which in turn activates luxR, leading to luminescence expression. The fact that a significant physiological effect was seen when CsrA was overexpressed in a ΔlitR strain suggests that the regulatory components upstream of litR are not involved in mediating the observed increase in luminescence. For example, if the V. fischeri system was regulated in a manner similar to V. cholerae through LuxO, then CsrA levels would have had no impact on luminescence output in the ΔlitR strain. Instead, CsrA appears to be regulating luminescence levels at some point in the quorum-sensing pathway downstream of LitR.

Figure 3. Analysis of strain luminescence outputs via factorial design.

Figure 3

Open in a new tab

Luminescence (RLU/OD) in wild-type ES114 background (panel A) or PMF8 ΔlitR background (panel B) with overexpression of csrB1 (pJW4), the vector control (pVSV104), or overexpression of csrA (pJW3) in the presence of 20 nM AHL. Data was obtained from two independent triplicate sets. The mean values, represented by the columns in each panel, for the factorial data points (pJW4 and pJW3) were obtained by fitting the entire data set using multivariate regression. The mean value for the center point, pVSV104, was obtained by averaging the two replications of that level. The error bars represent the 95% confidence intervals for the means.

At high cell density, the upstream quorum-sensing signaling cascade in V. fischeri results in derepression of litR (Fig. 1). LitR in turn not only activates luxR transcription, but also other processes in the cell that are important for host-colonization, motility, and metabolism (Fidopiastis, 2002; Studer et al., 2008). In V. cholerae, CsrA is known to indirectly control the activity of LuxO, which in turn modulates the activity of four Qrr sRNAs and the LitR homologue HapR (Lenz et al., 2005). Interestingly, although the quorum-sensing pathways of V. cholerae and V. fischeri contain some homologous components, the regulation and role of these components has evolved in a different manner. The V. cholerae system has no equivalent of LuxR in its regulatory cascade, therefore it could be speculated that it needs to have more sensitive control of expression of its LitR homologue, HapR, through CsrA, LuxO and multiple Qrr sRNAs (Lenz et al., 2005). However, in the V. fischeri system differential regulation of LitR and LuxR may work together to give the cells the flexibility they need to adapt to changing environmental or metabolic conditions.

Altering CsrA levels changes the abundance of luxR transcripts

It was hypothesized that CsrA must in some way cause activation of luxR in a LitR-independent manner. Because LitR is a transcriptional activator of luxR, its disruption leads to lower levels of luxR transcription, and therefore lower levels of luminescence expression, since the LuxR-AHL complex controls luminescence. The effect of CsrA on the system may be masked in the wild-type strain due to luxR transcription already being highly activated.

To determine if the increase in luminescence observed in PMF8 (pJW3) was due to an increase in luxR transcript levels, quantitative RT-PCR was performed on cDNA samples obtained from ES114 (wild-type) and PMF8 (ΔlitR) strains carrying pJW3 or pJW4 to modulate CsrA levels. In ES114, the luxR transcript level insignificantly decreased with increasing CsrA expression, but in PMF8, the amount of luxR transcript significantly increased as the amount of csrA transcript was increased (Fig. 4A & 4B). Thus the impact of CsrA on luminescence described above was manifested by the different dependence of the luxR transcript level on CsrA expression in ES114 vs. PMF8 (ΔlitR) (p < 0.0001). Because CsrA regulation of direct targets occurs post-transcriptionally, it is unlikely that CsrA controls the rate of luxR transcription directly. However it is possible that CsrA might impact the stability of the luxR mRNA.

Figure 4. Quantitative RT-PCR analysis of luxR transcripts via factorial design.

Figure 4

Open in a new tab

The levels of luxR transcripts in wild-type ES114 background (panel A) or PMF8 ΔlitR background (panel B) with overexpression of csrB1 (pJW4) or overexpression of csrA (pJW3) in the presence of 20 nM AHL. Data was obtained from two independent triplicate sets. The mean values, represented by the two columns in each panel, were obtained by fitting the entire data set using multivariate regression. The error bars represent the 95% confidence intervals for the means.

CsrA does not appear to impact the LuxR regulator cAMP-CRP

Several factors are known to directly regulate luxR transcription, including LuxR itself (Dunlap & Ray, 1989; Shadel, 1991; Chatterjee et al., 1996; Williams et al., 2008). Because LuxR levels are very low in a ΔlitR strain, it is considered unlikely that the effect seen in a csrA overexpression strain was due to LuxR autoregulation. Therefore experiments were performed to probe for interactions between CsrA and the known LuxR regulator cAMP-CRP. Activation of the cAMP-CRP activator by CsrA would result in an increased luxR transcription rate.

Quantitative RT-PCR was performed on cDNA samples obtained from ES114 (wild-type) and PMF8 (ΔlitR) strains with pJW3 or pJW4 in 20 nM AHL to examine crp transcript levels. In contrast to the dependence of luxR level on CsrA expression, the quantity of crp transcript did not depend on the expression level of csrA or on strain (p > 0.14) (data not shown).

Finally, in an effort to rule out any influence of cAMP levels on the increase in luminescence seen between PMF8 (pJW4) and PMF8 (pJW3), the luminescence experiment (Fig. 3AB) was repeated with 5 mM exogenous cAMP (Fig. 5AB). If cya activity were in some way being positively affected by CsrA, then addition of high levels of cAMP would be predicted to make luminescence output in PMF8 CsrA-independent. A relatively high concentration of cAMP was chosen because V. fischeri is capable of metabolizing cAMP, and it therefore needed to be provided in excess in order to ensure that there was enough to generate a response. When 5 mM cAMP was added to the growth medium, the luminescence levels did increase for both the wild type and PMF8 strains (compare Fig. 3AB to Fig. 5AB). However, the degree of change in luminescence between PMF8 (pJW3) and PMF8 (pJW4) was the same for each strain whether the concentration of cAMP was 0 (Fig. 3B) or 5 mM (Fig. 5B). Hence it can be concluded that regulation of cAMP levels did not produce the CsrA-dependent observed effects on luxR transcription.

Figure 5. Comparison of standard statistical and factorial design analysis of strain luminescence outputs in the presence of 5 mM cAMP.

Figure 5

Open in a new tab

Luminescence (RLU/OD) in wild-type ES114 background (panel A) or PMF8 ΔlitR background (panel B) with overexpression of csrB1 (pJW4) or overexpression of csrA (pJW3) in the presence of 20 nM AHL and 5 mM cAMP. The light gray columns represent data analyzed by the standard approach and the dark gray columns represent data analyzed by the factorial design approach. The two independent triplicate sets were averaged to obtain the means in the standard approach. The mean values in the factorial design approach were obtained by fitting the entire data set using multivariate regression. The error bars are 95% confidence intervals in both the standard and factorial design approach.

Comparison of standard laboratory versus factorial experimental design

All of the above experiments were performed simultaneously using both factorial design and standard laboratory design of at least two independent experiments with samples analyzed in triplicate. This enabled for a direct comparison of the analysis of the data via these two methods. Factorial design is a standard method of experimental design and data analysis (for example, see (Box, 1978; Montgomery, 1997)) widely used in agricultural and industrial research and development. It provides significant enhancement of statistical power vs. standard experimental designs, to identify subtle interactions between various regulatory elements. This typically can be achieved with fewer replications offering significant cost savings in time and materials that could be very beneficial when applied to expensive molecular-based studies. Due to the magnitude of the observed differences, the findings of this paper were the same using the factorial design method and the equivalent standard laboratory experiment, as demonstrated by the comparison of the two approaches in Fig. 5 for the luminescence measurements. The difficulty in using factorial design comes from the necessity of having personnel with the required statistical expertise. The advantage comes from the greater statistical power it affords that may enable smaller differences between measurements to be recognized, which might otherwise be missed, though this enhanced power was not critical in the current study.

Concluding thoughts

CsrA is capable of increasing the amount of luxR transcript in V. fischeri cells, which in turn elevates luminescence output. The mechanism by which this occurs is not yet precisely known, but the results indicate that CsrA does not act on the quorum-sensing network components upstream of luxR, nor does it influence the level of crp transcripts or adenylate cyclase activity. The manner in which CsrA-mediated regulation of the quorum-sensing system occurs in V. fischeri is distinct from that used in V. cholerae. The fact that this control is LitR-independent indicates that the regulation may occur prior to activation of the quorum-sensing network, and be important in generating an increase in luxR levels separately from the quorum-sensing pathway. This could occur in response to certain environmental cues or metabolic changes, and be an important factor in the timing of the quorum-sensing response in relation to metabolic state. CsrA is most active during exponential growth phase, and its levels become lower as the cell enters into late log and early stationary phase. The opposite is true of the quorum-sensing system, which becomes increasingly active as the cells transition from exponential growth to a high cell density stationary phase. Thus interactions between these two regulatory networks may be important in timing induction of quorum sensing and warrant further investigation.

Acknowledgements

Thanks to Edward Ruby and Cheryl Whistler for providing strains, Eric Stabb for both strains and experimental advice, Alison Kernell for technical assistance, Andre Levchenko and Rahul Kulkarni for their support of this work, and Mark Anderson (StatEase, Inc.) for help with factorial design. This work was funded by a subcontract from NIH R01 GM066786, NSF IGERT DGE-0504196 and ICTAS at Virginia Tech.

References

  1. Baker CS, Morozov I, Suzuki K, Romeo T, Babitzke P. CsrA regulates glycogen biosynthesis by preventing translation of glgC in Escherichia coli. Mol Microbiol. 2002;44:1599–1610. doi: 10.1046/j.1365-2958.2002.02982.x. [DOI] [PubMed] [Google Scholar]
  2. Boettcher KJ, Ruby EG. Depressed light emission by symbiotic Vibrio fischeri of the sepiloid squid Euprymna scolopes. J Bacteriol. 1990;172:3701–3706. doi: 10.1128/jb.172.7.3701-3706.1990. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Bose JL, Kim U, Bartkowski W, Gunsalus RP, Overley AM, Lyell NL, Visick KL, Stabb EV. Bioluminescence in Vibrio fischeri is controlled by the redox-responsive regulator ArcA. Mol Microbiol. 2007;65:538–553. doi: 10.1111/j.1365-2958.2007.05809.x. [DOI] [PubMed] [Google Scholar]
  4. Bose JL, Rosenberg CS, Stabb EV. Effects of luxCDABEG induction in Vibrio fischeri: enhancement of symbiotic colonization and conditional attenuation of growth in culture. Arch Microbiol. 2008;190:169–183. doi: 10.1007/s00203-008-0387-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Box GEP, Hunter WG, Hunter JS. Statistics for Experimenters: An Introduction to Design, Data Analysis, and Model Building. New York: John Wiley and Sons; 1978. [Google Scholar]
  6. Chatterjee J, Miyamoto CM, Meighen EA. Autoregulation of luxR: the Vibrio harveyi lux-operon activator functions as a repressor. Mol Microbiol. 1996;20:415–425. doi: 10.1111/j.1365-2958.1996.tb02628.x. [DOI] [PubMed] [Google Scholar]
  7. Dunlap PV, Ray JM. Requirement for autoinducer in transcriptional negative autoregulation of the Vibrio fischeri luxR gene in Escherichia coli. J Bacteriol. 1989;171:3549–3552. doi: 10.1128/jb.171.6.3549-3552.1989. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Dunlap PV, Greenberg EP. Control of Vibrio fischeri luminescence gene expression in Escherichia coli by cyclic AMP and cyclic AMP receptor protein. J Bacteriol. 1985;164:45–50. doi: 10.1128/jb.164.1.45-50.1985. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Dunlap PV, Greenberg EP. Control of Vibrio fischeri lux gene transcription by a cyclic AMP receptor protein-LuxR protein regulatory circuit. J Bacteriol. 1988;170:4040–4046. doi: 10.1128/jb.170.9.4040-4046.1988. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Dunn AK, Millikan DS, Adin DM, Bose JL, Stabb EV. New rfp- and pES213-derived tools for analyzing symbiotic Vibrio fischeri reveal patterns of infection and lux expression in situ. Appl Environ Microbiol. 2006;72:802–810. doi: 10.1128/AEM.72.1.802-810.2006. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Engebrecht J, Silverman M. Identification of genes and gene products necessary for bacterial bioluminescence. P Natl Acad Sci USA. 1984;81:4154–4158. doi: 10.1073/pnas.81.13.4154. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Fidopiastis P, Miyamato C, Jobling M, Meighan E, Ruby G. LitR, a new transcriptional activator in Vibrio fischeri regulates luminescence and symbiotic light organ colonization. Molec Microbiol. 2002;45:131–143. doi: 10.1046/j.1365-2958.2002.02996.x. [DOI] [PubMed] [Google Scholar]
  13. Hanahan D. Studies on the transformation of Escherichia coli with plasmids. J Mol Biol. 1983;166:557–580. doi: 10.1016/s0022-2836(83)80284-8. [DOI] [PubMed] [Google Scholar]
  14. Jones MK, Warner EB, Oliver JD. CsrA inhibits the formation of biofilms by Vibrio vulnificus. Appl Environ Microbiol. 2008;74:7064–7066. doi: 10.1128/AEM.01810-08. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Kovach ME, Elzer PH, Hill DS, Robertson GT, Farris MA, Roop RM, II, Peterson KM. Four new derivatives of the broad-host-range cloning vector pBBR1MCS, carrying different antibiotic-resistance cassettes. Gene. 1995;166:157–176. doi: 10.1016/0378-1119(95)00584-1. [DOI] [PubMed] [Google Scholar]
  16. Kulkarni PR, Cui X, Williams JW, Stevens AM, Kulkarni RV. Prediction of CsrA-regulating small RNAs in bacteria and their experimental verification in Vibrio fischeri. Nucleic Acids Res. 2006;34:3361–3369. doi: 10.1093/nar/gkl439. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Lenz DH, Miller MB, Zhu J, Kulkarni RV, Bassler BL. CsrA and three redundant small RNAs regulate quorum sensing in Vibrio cholerae. Mol Microbiol. 2005;58:1186–1202. doi: 10.1111/j.1365-2958.2005.04902.x. [DOI] [PubMed] [Google Scholar]
  18. Liu MY, Yang H, Romeo T. The product of the pleiotropic Escherichia coli gene csrA modulates glycogen biosynthesis via effects on mRNA stability. J Bacteriol. 1995;177:2663–2672. doi: 10.1128/jb.177.10.2663-2672.1995. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Lupp C, Urbanowski M, Greenberg EP, Ruby EG. The Vibrio fischeri quorum-sensing systems ain and lux sequentially induce luminescence gene expression and are important for persistence in the squid host. Mol Microbiol. 2003;50:319–331. doi: 10.1046/j.1365-2958.2003.t01-1-03585.x. [DOI] [PubMed] [Google Scholar]
  20. Montgomery DC. Design and Analysis of Experiments. 4th edn. New York: John Wiley and Sons; 1997. [Google Scholar]
  21. Romeo T. Global regulation by the small RNA-binding protein CsrA and the non-coding RNA molecule CsrB. Mol Microbiol. 1998;29:1321–1330. doi: 10.1046/j.1365-2958.1998.01021.x. [DOI] [PubMed] [Google Scholar]
  22. Shadel GS, Baldwin TO. The Vibrio fischeri LuxR protein is capable of bidirectional stimulation of transcription and both positive and negative regulation of the luxR gene. J Bacteriol. 1991;173:568–574. doi: 10.1128/jb.173.2.568-574.1991. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Stabb EV, Ruby EG. RP4-based plasmids for conjugation between Escherichia coli and members of the Vibrionaceae. Methods Enzymol. 2002;358:413–426. doi: 10.1016/s0076-6879(02)58106-4. [DOI] [PubMed] [Google Scholar]
  24. Studer SV, Mandel MJ, Ruby EG. AinS quorum sensing regulates the Vibrio fischeri acetate switch. J Bacteriol. 2008;190:5915–5923. doi: 10.1128/JB.00148-08. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Timmermans J, Van Melderen L. Post-transcriptional global regulation by CsrA in bacteria. Cell Mol Life Sci. 2010;67:2897–2908. doi: 10.1007/s00018-010-0381-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Visick KL. Layers of signaling in a bacterium-host association. J Bacteriol. 2005;187:3603–3606. doi: 10.1128/JB.187.11.3603-3606.2005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Waters CM, Bassler BL. Quorum sensing: cell-to-cell communication in bacteria. Annu Rev Cell Dev Biol. 2005;21:319–346. doi: 10.1146/annurev.cellbio.21.012704.131001. [DOI] [PubMed] [Google Scholar]
  28. Williams JW, Cui X, Levchenko A, Stevens AM. Robust and sensitive control of a quorum-sensing circuit by two interlocked feedback loops. Mol Syst Biol. 2008;4:234. doi: 10.1038/msb.2008.70. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES