Effects of luxCDABEG induction in Vibrio fischeri: enhancement of symbiotic colonization and conditional attenuation of growth in culture

Abstract

Production of bioluminescence theoretically represents a cost, energetic or otherwise, that could slow Vibrio fischeri growth; however, bioluminescence is also thought to enable full symbiotic colonization of the Euprymna scolopes light organ by V. fischeri. Previous tests of these models have proven inconclusive, partly because they compared nonisogenic strains, or undefined and/or pleiotropic mutants. To test the influence of the bioluminescence-producing lux operon on growth and symbiotic competence, we generated dark luxCDABEG mutants in strains MJ1 and ES114 without disrupting the luxR-luxI regulatory circuit. The MJ1 luxCDABEG mutant out-competed its visibly luminescent parent approximately 26% per generation in a carbon-limited chemostat. Similarly, induction of luminescence in the otherwise dim ES114 strain slowed growth relative to DeltaluxCDABEG mutants. Some culture conditions yielded no detectable effect of luminescence on growth, indicating that luminescence is not always growth limiting; however, luminescence was never found to confer an advantage in culture. In contrast to this conditional disadvantage of lux expression, ES114 achieved approximately fourfold higher populations than its luxCDABEG mutant in the light organ of E. scolopes. These results demonstrate that induction of luxCDABEG can slow V. fischeri growth under certain culture conditions and is a positive symbiotic colonization factor.

Fig. 1

Schematic representation of chromosomal gene organization in lux mutants. Black arrows indicate the direction of transcription of lux or other genes. An asterisk denotes location of a frame-shifting 4-bp insertion in strain EVS101 resulting in a non-functional LuxI. “T” indicates where three bidirectional transcriptional terminators were added between luxI and lacI^q in strains JB22 and JB23.

Fig. 2

Effect of 3-oxo-C6-HSL on the growth and luminescence of V. fischeri ES114 in culture. ES114 was grown with (filled symbols) or without (empty symbols) addition of 1 μM 3-oxo-C6-HSL. For cells grown in SWTO at 24°C (panels a, b, and c), the data shown include the specific luminescence plotted as a function of culture density (panel a), the culture density over time (panel b), and the doubling time as a function of OD₅₉₅ (panel c). The doubling time as a function of OD₅₉₅ is also reported for cells grown in LBS at 28°C (panel d).

Fig. 3

Relative growth of strains EVS100 and ES114. The ΔluxA::ermR mutant EVS100 (triangles) and wild type ES114 (diamonds) were grown in SWTO medium at 24°C with (filled symbols) or without (empty symbols) addition of 1 μM 3-oxo-C6-HSL.

Fig. 4

Specific induction of bioluminescence affects growth of strains EVS101 and JB22. Doubling time is reported as a function of culture OD₅₉₅ for cells grown in SWTO at 24°C with (filled symbols) or without (empty symbols) addition of 0.5 mg ml⁻¹ IPTG. a Growth of wild type ES114 (diamonds) and luxR::ermR lacI^q P_tac-luxI-luxCDABEG mutant EVS101 (triangles). b Growth of ES114 (diamonds), the lacI^q P_A1/34-luxCDABEG mutant JB22 (squares), or the dark lacI^q P_A1/34-ΔluxCDABEG control strain JB23 (circles). c Specific luminescence of ES114 (diamonds), EVS101 (triangles), and JB22 (squares).

Fig. 5

3-oxo-C6-HSL slows growth of ES114 but not ΔluxCDABEG mutant. Doubling time is reported as a function of culture OD₅₉₅ for wild type ES114 (diamonds) or ΔluxCDABEG mutant EVS102 (triangles) grown in SWTO at 24°C with (filled symbols) or without (empty symbols) addition of 1 μM 3-oxo-C6-HSL.

Fig. 6

ΔluxCDABEG mutant EVS103 out-competes wild-type parent MJ1. In each panel data is from one representative experiment of three. a Relative competitiveness (RCI) of EVS103 (ΔluxCDABEG) co-cultured with MJ1 in SWTO at 24°C. The dotted line follows the best fit of the data (RCI=1.05 per generation). b RCI of EVS103 and MJ1 co-cultured continuously in a carbon-limited chemostat in BGMYE with 4 mM glycerol at 24°C (see Materials and Methods), conditions under which they are highly luminescent (Table 2). The dotted line follows the best fit of the data (RCI=1.26 per generation). RCI is defined as the ratio of EVS103 to MJ1 at each generation divided by the ratio of these strains at the start of the experiment.

Fig. 7

Symbiotic colonization of E. scolopes by lux mutants and wild type. a Average colonization levels 48h after inoculation with the indicated strain. Error bars represent standard error (n=17). Asterisks denote a significant (p<0.005) difference from wild type. b Competitiveness of EVS102 (ΔluxCDABEG) when presented in a mixed (~1:1) inoculum with wild type and recovered from squid after 48 h. Either ES114 (solid symbols) or EVS102 (empty symbols) contained pVSV3 (lacZ), which enabled strain identification by blue/white screening after patching on LBS plates supplemented with 50 μg/ml X-Gal. Each symbol represents the RCI determined from one squid, defined as the ratio of EVS102:ES114 in the squid divided by the ratio in the inoculum. An arrow marks the average RCI of 0.32, which was significantly <1 (p<0.01).