Attenuation of host NO production by MAMPs potentiates development of the host in the squid-vibrio symbiosis

Abstract

Bacterial pathogens typically upregulate the host's production of nitric oxide synthase (NOS) and nitric oxide (NO) as antimicrobial agents, a response that is often mediated by microbe-associated molecular patterns (MAMPs) of the pathogen. In contrast, previous studies of the beneficial Euprymna scolopes/Vibrio fischeri symbiosis demonstrated that symbiont colonization results in attenuation of host NOS/NO, which occurs in high levels in hatchling light organs. Here, we sought to determine whether V. fischeri MAMPs, specifically lipopolysaccharide (LPS) and the peptidoglycan derivative tracheal cytotoxin (TCT), attenuate NOS/NO, and whether this activity mediates the MAMPs-induced light organ morphogenesis. Using confocal microscopy, we characterized levels of NOS with immunocytochemistry and NO with a NO-specific fluorochrome. When added exogenously to seawater containing hatchling animals, V. fischeri LPS and TCT together, but not individually, induced normal NOS/NO attenuation. Further, V. fischeri mutants defective in TCT release did not. Experiments with NOS inhibitors and NO donors provided evidence that NO mediates apoptosis and morphogenesis associated with symbiont colonization. Attenuation of NOS/NO by LPS and TCT in the squid-vibrio symbiosis provides another example of how the host's response to MAMPs depends on the context. These data also provide a mechanism by which symbiont MAMPs regulate host development.

Fig 1. Juvenile E. scolopes and its light organ

A. Left: A dorsal view of a juvenile E. scolopes showing the location of the light organ beneath the mantle (box). B. An LSM confocal micrograph of the surface of a hatchling light organ stained with acridine orange, which reveals the general morphology of the organ. Each lateral surface (box) is covered by a complex, juvenile specific superficial ciliated epithelium C. An illustration of the internal morphology of the light organ. aa = anterior appendage, ac = antechamber, c = crypt, d = duct, p = pores, pa = posterior appendage

Fig 2. The visualization and quantification of NOS attenuation in the presence and absence of V. fischeri MAMPs by LSM confocal microscopy

A. Representative micrographs of duct cells labeled with a uNOS antibody and a FITC-conjugated secondary antibody (green). The left column depicts duct tissues from representative animals. The orange box indicates the field shown in the right hand column, which is a magnified view of individual duct cells. [Counterstain: Actin, rhodamine phalloidin (red) and nuclei, TOTO-3 (blue).] (Bars, 10 μm.) B. The quantification (see Materials and Methods) of a single representative microscopy experiment (APO n=10, SYM n=12, LPS n=8, TCT n=9, LPS+TCT n=14). Bars, standard error. (*), data points that were significantly different from SYM.

Fig 3. The visualization by LSM confocal microscopy of NO attenuation in the presence and absence of V. fischeri MAMPs

Lower images, changes in NO levels due to bacterial products. Upper images, control aposymbiotic and symbiotic tissues. Representative micrographs of duct cells stained for NO production with DAF-FM (green). Bar, 20 μm.

Fig 4. The visualization by LSM confocal microscopy and quantification of NOS attenuation in the presence of V. fischeri MAMP mutants and in the presence of LPS isolated from non-symbiotic bacteria

A. Quantification of antechamber cells stained with a commercial uNOS antibody. Animals were exposed to mutants defective in TCT release (ltg⁻) or complemented pharmacologically with 1 μM TCT (ltg⁻ + TCT) or genetically with one of the lytic transglycosylase genes in a multicopy plasmid (ltg⁻ + ltgA or ltg⁻ + ltgD). (APO n=9, SYM n=13, ltg⁻n=8, ltg⁻+TCT n=12, ltg⁻+ltgA n=14, ltg⁻+ltgD n=11, ltg⁻+plasmid n=14) B. Quantification of antechamber cells stained with a commercial antibody from animals exposed to LPS purified from non-symbiotic bacteria (N. gonorrheae, N. meningitides, H. influenzae) and 1 μM TCT (APO n=6, SYM n=7, V. fischeri LPS n=9, N. gonorrheae n=9, N. meningitides n=6, H. influenzae n=8). Graphs are of single representative experiments. [Bars, standard error. (*), data points which were significantly different from SYM.]

Fig 5. The role of NO in the induction of apoptosis

A. Representative micrographs of the anterior appendages of light organs stained with either acridine orange (Top row) or TUNEL (green) and rhodamine phalloidin (red) (bottom row). Arrows indicate apoptotic nuclei. B. Quantification of single representative experiments of animals exposed to NO donor (SNAP) or inhibitor (SMTC) as indicated and quantified for either early- stage (acridine orange) or late-stage (TUNEL) apoptosis. The number of apoptotic nuclei were counted per appendage for each treatment. C. Graph of single representative experiment of animals exposed to NO inhibitor (SMTC) and scored for stage of regression. D. Graphs of average of two experiments of animals exposed to NO inhibitor (SMTC) and/or bacterial MAMPs as indicated and quantified for late-stage (TUNEL) apoptosis. The number of apoptotic nuclei were counted per appendage for each treatment. Statistical analyses were done comparing the samples indicated by brackets using a Student’s T-test with a Bonferroni adjustment. (*), significance of p ≤ 0.05. n.s., “not siginificantly different”.

Fig 6. A model for the role of NOS/NO in developmental apoptosis of the light organ

In stage 1, LPS and TCT are recognized by the animal and have the synergistic effect of attenuating NOS/NO, which removes a block on early-stage apoptosis. Removal of this block allows late-stage apoptosis (stage 2) to proceed by LPS or TCT stimulating an as yet undetermined pathway, possibly p53.