Vibrio fischeri imports and assimilates sulfate during symbiosis with Euprymna scolopes

Abstract

Sulfur is in cellular components of bacteria and is, therefore, an element necessary for growth. However, mechanisms by which bacteria satisfy their sulfur needs within a host are poorly understood. Vibrio fischeri is a bacterial symbiont that colonizes, grows, and produces bioluminescence within the light organ of the Hawaiian bobtail squid, which provides an experimental platform for investigating sulfur acquisition in vivo. Like other γ-proteobacteria, V. fischeri fuels sulfur-dependent anabolic processes with intracellular cysteine. Within the light organ, the abundance of a ΔcysK mutant, which cannot synthesize cysteine through sulfate assimilation, is attenuated, suggesting sulfate import is necessary for V. fischeri to establish symbiosis. Genes encoding sulfate-import systems of other bacteria that assimilate sulfate were not identified in the V. fischeri genome. A transposon mutagenesis screen implicated YfbS as a sulfate importer. YfbS is necessary for growth on sulfate and in the marine environment. During symbiosis, a ΔyfbS mutant is attenuated and strongly expresses sulfate-assimilation genes, which is a phenotype associated with sulfur-starved cells. Together, these results suggest V. fischeri imports sulfate via YfbS within the squid light organ, which provides insight into the molecular mechanisms by which bacteria harvest sulfur in vivo.

Keywords: Vibrio; host-microbe interactions; sulfate assimilation; symbiosis; transport.

Fig. 1.

CysK promotes sulfate assimilation in V. fischeri. Growth curves of ES114 (WT) and NPW90 (ΔcysK) in sulfate-replete (50 mM sulfate) defined minimal medium. Each point represents a turbidity measurement (OD₆₀₀), and each curve represents an independent culture (N = 3). Experiment was performed two times with similar results obtained from both trials.

Fig. 2.

CysK is necessary for normal symbiosis establishment. A. Luminescence emission by juvenile squid (N = 30) at 48 h post-inoculation (p.i.) that were exposed to inoculums of ES114 (WT) or NPW90 (ΔcysK). Each point represents an individual animal, and each bar represents the median of the corresponding group. Dotted line represents the 95% percentile of luminescence measurements of animals (N = 10) exposed to a mock inoculum (Apo). Animals with luminescence emission levels above this line are deemed bioluminescent. RLU = relative light units. A two-tailed Mann-Whitney test was performed to test for statistical significance between group medians (**** = p < 0.0001). B. Abundance of V. fischeri in squid described in A. Each point represents an individual animal, and each bar represents the median of the corresponding group. Dotted line indicates limit of detection (14 CFU). A two-tailed Mann-Whitney test was performed to test for statistical significance between group medians (**** = p < 0.0001). Data shown in graphs are from one experimental trial. Experiment was performed twice using comparable N, and similar results were obtained from both trials.

Fig. 3.

CysB-dependent transcription is elevated within host-associated cells that lack CysK. A. GFP-based transcriptional reporter for P_tcyJ in plasmid pVF_0008P. Expression of mCherry depends on P_tetA. B. Images of light organs colonized by ES114 (WT; top) or NPW90 (ΔcysK; bottom) harboring the P_tcyJ transcriptional reporter at 72 h p.i.. Scale bar = 100 μm. C. Quantification of P_tcyJ expression at 72 h p.i. in light organs of squid colonized by ES114 (WT) or NPW90 (ΔcysK) harboring pVF_0008P (P_tcyJ::gfp). Each point represents P_tcyJ expression of host-associated V. fischeri populations within an individual squid (WT: N = 23; ΔcysK: N = 28), and each bar represents the mean of the corresponding group. An unpaired, two-tailed t test with Welch’s correction was performed to test for statistical significance between means (**** = p-value < 0.0001). Experiment was performed twice using comparable N, and similar results were obtained from both trials. Data shown in graph are from one experimental trial.

Fig. 4.

Cystine depletion promotes CysB-dependent transcription in cells that lack CysK. A. Growth curves of ES114 (WT; closed symbols) and NPW90 (ΔcysK; red lines and open symbols) in sulfate-replete, defined minimal medium supplemented with the indicated amount of cystine (Cys₂). For WT curves, black and blue line indicate 0 and 500 μM Cys₂, respectively. Each curve represents a single culture that was sampled at the indicated times. Experiment was performed twice with similar results obtained from both trials. Data shown in graphs are from one experimental trial. B. Left, Growth curves of ES114 (WT; closed symbols) and NPW90 (ΔcysK; open symbols) harboring the P_tcyJ transcriptional reporter (pVF_0008P) in sulfate-replete, defined minimal medium supplemented with 18.5 μM cystine. Each curve represents a single culture that was sampled at the indicated times. Arrowheads indicate samples examined for expression of P_tcyJ. Right, Levels of green fluorescence normalized by OD₆₀₀ for samples described in the left panel. Dotted line indicates autofluorescence associated with a non-fluorescent control strain. Experiment was performed twice with similar results obtained from both trials. Data shown in graph are from one experimental trial.

Fig. 5.

Growth of and CysB-dependent transcription in cysteine auxotrophs in rich medium. A. Growth curves of ES114 (WT; closed symbols) and NPW90 (ΔcysK; open symbols) in LBS medium. Each curve represents an independent culture (N = 3). Differences between replicates is smaller than resolution allows. Experiment was performed twice with similar results obtained from both trials. B. Images of spots of ES114 (WT), AGC10 (ΔcysK), or AGC09 (cysK⁺) harboring pVF_0008P (P_tcyJ::gfp) or pVSV105 (vector) grown on LBS for 24 h. Scale bar = 1 mm. C. Quantification of P_tcyJ expression in strains harboring pVF_0008P grown as described in panel B. Each point represents the green fluorescence of an individual spot, and bars represent the group means. Dotted line indicates mean auto-fluorescence associated with spots of the non-fluorescent vector control pVSV105/ES114 (N = 3). A one-way ANOVA test demonstrated statistical significance between group means (F_2,6 = 9,407, p-value < 0.0001). A Tukey’s post-hoc test was performed to statistically compare the means of each group, with different letters indicating p-value < 0.0001 and same letters indicating p-value > 0.05, with p-values adjusted for multiple comparisons. Experiment was performed twice with similar results obtained from both trials.

Fig. 6.

yfbS encodes a conserved transporter that promotes sulfate assimilation in V. fischeri. A. Phylogenetic tree for taxa that represent the major clades of Vibrionaceae. Tree is based on the multi-gene supertree reported previously (Sawabe et al., 2013). All taxa listed encode a CysC homolog. Taxa highlighted in black encode a YfbS homolog. Locus tags of CysC and YfbS homologs are listed in Table S1. B. Turbidity of cultures of ES114 (WT) and NPW88 (ΔyfbS) harboring either pTM214 (vector) or pNW032 (P_trc-yfbS) after 21 h of incubation in sulfate-replete DMM supplemented with 1 mM IPTG. Each point represents the optical density (OD₆₀₀) of an individual culture (N = 3). A one-way ANOVA revealed statistically significant differences between means (F_2,6 = 2011, p-value < 0.0001). A Tukey’s post-hoc test was performed to statistically compare the means of each group, with p-values adjusted for multiple comparisons (* = p < 0.05, **** = p < 0.0001). Experiment was performed twice with similar results obtained from both trials. C. Turbidity of cultures of E. coli strains BW25113 (WT) and JW2415 (ΔcysA) harboring either pTrc99a (vector) or pNW039 (P_trc-yfbS) after 21 h of incubation in minimal A glucose medium ± 100 μM IPTG. Each point represents the optical density (OD₆₀₀) of an individual culture (N = 3). A two-way ANOVA test revealed statistically significant differences for genotype (F_3,16 = 773, p-value < 0.0001) and interactions between genotype and IPTG treatment (F_3,16 = 58.37, p-value < 0.0001), but not IPTG treatment alone (F_1,16 = 0.6470, p-value = 0.433). A Tukey’s post-hoc test was performed to statistically compare the means of each group, with p-values adjusted for multiple comparisons. Comparisons between groups labeled with different letters indicate statistical significance between their means (p-value < 0.0001) and those with the same letters not significant (p-value > 0.05). D. Sulfate uptake by cultures of ES114 (WT) and NPW88 (ΔyfbS) that were grown in defined minimal medium supplemented with 0.1 mM sulfate and 18.5 μM cystine. At an OD₆₀₀ of 0.8, cells were harvested and exposed to an [³⁵S] sulfate cocktail. Radioactivity of approximately 9.0x10⁸ CFU was measured at the indicated time points. cpm = counts per minute. Experiment was repeated three times with similar results obtained from each trial.

Fig. 7.

YfbS promotes sulfate assimilation during symbiosis establishment. A. Abundance of ES114 (WT) or NPW88 (ΔyfbS) harboring pTM214 (vector) or pNW032 (P_trc-yfbS) in FSSW supplemented with 1 mM IPTG. Each point represents the concentration of recovered CFU (CFU/ml), and each line connecting points indicates an individual culture (N = 3). B. Left, Luminescence emission by juvenile squid (N = 13–15) at 48 h p.i. that were exposed to inoculums of ES114 (WT) or NPW88 (ΔyfbS). Each point represents an individual animal, and each bar represents the median of the corresponding group. Dotted line represents the 95% percentile of luminescence measurements of animals (N = 5) exposed to a mock inoculum (Apo). Animals with luminescence emission levels above this line are deemed bioluminescent. RLU = relative light units. A two-tailed Mann-Whitney test was performed to test for statistical significance between group medians (*** = p < 0.001). Right, Abundance of V. fischeri in squid shown in left. Each point represents an individual animal, and each bar represents the median of the corresponding group. Dotted line indicates limit of detection (14 CFU). A two-tailed Mann-Whitney test was performed to test for statistical significance between group medians (** = p < 0.01). Data shown in graphs are from one experimental trial. Experiment was performed twice using comparable N, and similar results were obtained from both trials. C. Images of light organs colonized by ES114 (WT; top) or NPW88 (ΔyfbS; bottom) harboring the P_tcyJ transcriptional reporter at 72 h p.i. Scale bar = 100 μm. D. Quantification of P_tcyJ activity at 72 h p.i. in light organs of squid colonized by ES114 (WT) or NPW88 (ΔyfbS) harboring pVF_0008P (P_tcyJ::gfp). Each point represents an individual squid (WT: N = 20; ΔyfbS: N = 28), and each bar represents the median of the corresponding group. Image analysis described in methods. An unpaired, two-tailed t test with Welch’s correction was performed to test for statistical significance between means (**** = p-value < 0.0001). Data shown in graph are from one experimental trial. Experiment was performed twice using comparable N, and similar results were obtained from both trials.

Fig. 8.

Model of sulfate assimilation in V. fischeri. This model represents an updated version of one presented previously (Wasilko et al., 2019). Nodes indicate intermediate sulfur compounds, cofactors, and other substrates involved in sulfate assimilation. Arrows are labeled with the genetic factors encoded by V. fischeri that facilitate each step. APS = adenosine-5’ phosphosulfate; PAPS = 3'-phosphoadenosine-5'-phosphosulfate; OAS = O-acetyl-serine

Fig. 9.

Model of CysB-dependent sulfur import in V. fischeri. Left, Within the marine environment, sulfate is the primary sulfur source and imported into V. fischeri via YfbS. The sulfate is assimilated by synthesizing cysteine (1), which promotes viability and growth in seawater. High O-acetyl-serine (OAS) levels promote CysB-dependent expression of factors involved in sulfate assimilation, sulfate import (YfbS), and cystine importers (e.g., TcyP). Right, In the light organ, V. fischeri growth is accelerated due by accessing host-derived peptides, which are presumed to serve as a source of cysteine that auto-oxidizes to cystine. To meet the demand for sulfur, cells supplement sulfate assimilation by importing cystine, which boost cysteine pools. This increased flux of cysteine leads to lower OAS levels through feedback inhibition over CysE, which in turn decreases CysB-dependent transcription.