An acidic microenvironment produced by the V-type ATPase of Euprymna scolopes promotes specificity during Vibrio fischeri recruitment

Abstract

Animals often acquire their microbial symbionts from the environment, but the mechanisms underlying how specificity of the association is achieved are poorly understood. We demonstrate that the conserved proton pump, V-type ATPase (VHA), plays a key role in the establishment of the model light-organ symbiosis between the squid Euprymna scolopes and its bacterial partner, Vibrio fischeri. Recruitment of V. fischeri from the surrounding seawater is mediated by juvenile-specific ciliated fields on the organ's surface. These epithelia produce acidic mucus containing antimicrobials with low-pH optima, creating a chemical environment fostering specific recruitment of V. fischeri. We provide evidence that this critical acidic landscape is created by activity of VHA. VHA inhibition abolished epithelial-cell acidity, resulting in increased mucus pH and inefficient symbiont colonization. Thus, VHA provides a mechanistic link between host modulation of microenvironmental acidity, immune function, and selection of microbial symbionts, a strategy for specificity that may govern other symbioses.

Fig. 1. Features of the juvenile E. scolopes light organ that foster recruitment of environmental V. fischeri cells.

Left, diagram of a ventral view of a hatchling squid showing the internal organ’s position in the center of the mantle cavity (white arrow). Middle, diagrams of the internal and external features of the juvenile organ, before and after mucus shedding in response to environmental peptidoglycan; aa/pa, anterior/posterior appendages on the organ surface; P, pores near which environmental bacteria aggregate before migrating into and colonizing the deep crypts, C. Bacteria in the ambient seawater include V. fischeri (VF; blue), and other gram negative (‘-’; red) or gram positive (‘+’; yellow) cells. Right, zoomed in diagram of one side of the organ surface, showing the gradient of acidification that develops in the mucus during the first hours following hatching. This acidification, together with an array of antimicrobial proteins and peptides in the mucus, select against cells that are not V. fischeri. The resulting biochemical and biomechanical environment created near the pores fosters the dominance of V. fischeri cells in the aggregates within ~3 h of hatching. Modified from images created by BioRender.

Fig. 2. Transcriptional expression profiles of juvenile E. scolopes genes encoding subunits of the V-type proton ATPase (VHA).

A Cartoon of VHA predicted from the genome of E. scolopes. B Heatmap showing the genes encoding VHA subunits and assembly proteins, as well as their expression profiles across three different organs [light organ (LO), eye and gill] 24 h after hatching (data from ref. ). In these animals the light organ either remained aposymbiotic or was colonized with either the native bioluminescent strain ES114 (WT), or a dark-mutant derivative of ES114 (ΔLUX). C Heatmap depicting the expression of E. scolopes carbonic anhydrase (CA) genes and a sodium-proton transporter (NHE3) predicted to be involved in acid-base regulation. Gene counts (rows) were Z-transformed, i.e., scaled to have mean of zero, with a standard deviation of 2.1 (B) or 1.9 (C); dark magenta indicates highest expression and dark green indicates lowest. Only genes with a >1 TMM expression value in at least one sample are listed. See Supplementary Data Set 1 for the raw dataset including those genes with <1 TMM expression, and for the entire transcriptomic and genomic blast alignments.

Fig. 3. Phylogenetic and structural analyses showing a strong conservation of subunit B of VHA (VHA_B).

A Phylogenetic relationships among VHA_B proteins. Numbers at nodes indicate bootstrap values; an asterisk (*) indicates a value above 75%. Scale bar, number of amino acid substitutions per site. B AlphaFold-derived models predicting that VHA_B proteins show a high degree of conformational conservation between humans (Homo sapiens) and bobtail squid (Euprymna scolopes). The position of the epitope used to develop the VHA_B antibody used in this study is indicated by the brackets (see Methods for sequence of this region). Protein coloring denotes secondary structure: red, beta-pleated sheet; blue, alpha helix; black, random coil.

Fig. 4. Immunocytochemical (ICC) localization of the VHA_B protein within the ciliated fields of the juvenile light organ’s superficial epithelium.

A Illustration indicating the position of the light organ in the mantle cavity, and the appendage tissues (dashed yellow oval indicates one anterior appendage) showing the region of cells displayed in the rest of this figure. B, B’ Confocal microscopy image of areas of VHA_B-antibody (αVHA_B) labeling. B Labeling occurred in the apical surfaces of the epithelium (white box), enlarged below and in (B’). Inset: ICC control staining showed undetectable labeling. B’ Imaris 3D rendering highlighting the close association of the antibody with the epithelial nuclei. C, C’ At higher sensitivity, labeling of host cilia was detected. C Confocal image of antibody labeling (white arrows). C’ Imaris rendering. D, D’ Increased magnification and Imaris rendering. E Upper, low and high magnification images of LysoSensor-stained acidic regions associated with epithelial nuclei; lower, Imaris rendering at even higher magnification, providing evidence that the nuclear regions wrap around (white arrows) the acidic regions. F Immuno-electron micrograph of the distal edges of ciliated epithelial cells. Left, an image defining the ciliated region; right, enlargement of the area in the black box, including points of VHA_B labeling (red arrows). LIS, lateral intracellular space; MV, microvilli; TW, terminal web.

Fig. 5. Evidence for VHA activity in the living squid light organ at different colonization times, and under different conditions, during the early stages of symbiosis development.

The relative acidity of regions of the light organ’s superficial ciliated fields in 0-h hatchling, 24-h aposymbiotic (APO) and 24-h symbiotic (SYM) animals was revealed by incubation with LysoSensor (green) in the presence or absence of the VHA inhibitor concanamycin A (ConcA); samples were visualized by confocal microscopy (see Methods for details). A, A’ representative images at low and high magnifications, respectively, of the anterior appendages. The appendages of both hatchling and 24-h APO animals had similar patterns of LysoSensor labeling, with bright regions at the apical and basal surfaces of the cells. After ConcA exposure, both hatchling and APO squid had attenuated signal in these regions. In contrast, the apical and basal surfaces of the SYM animals had similar labeling, with additional labeling both in hemocytes that had entered the blood sinus of the appendage and in epithelial cells that were undergoing symbiosis-induced cell death. The presence of ConcA also diminished the pattern of acidity in SYM animals. B Higher magnification images revealed areas of acidity in the region of the cilia. ConcA diminished the degree of staining in both hatchling and APO animals; SYM animals had no labeling in the cilia with or without ConcA. C LysoSensor staining around the pores was similar in hatchling animals under both untreated and ConcA-treated conditions but was diminished in ConcA treated 24-h APO animals. Both untreated and ConcA-treated SYM animals showed similar, comparatively low-level staining with LysoSensor. All scale bars, 25 µm; n = 15 for each condition (3 clutches, 5 animals/clutch).

Fig. 6. The impact on colonization of the mucosal pH gradient present along the hatchling light-organ surface.

A, A’ Tissue pH is indicated by the ratio of fluorescence emissions of the fluorochrome SNARF at 580 and 650 nm. A Fluorescence of SNARF associated with tissue surfaces indicated a sharp pH gradient from the more acidic pore region to the less acidic appendages (upper row), which was reduced when the VHA-inhibitor ConcA was present (lower row); aa, anterior appendage; p, pore; pa, posterior appendage. A’ Effect of ConcA treatment on the relative fluorescence intensities in two tissue regions, the appendages (Append.) and the pores, of the organ surface. Each point represents a single squid; n = between 21–24 squid; Bonferroni-corrected, independent Mann–Whitney U tests; *p < 0.05, **p < 0.01. Impact of VHA inhibition on light-organ colonization. B Colonization efficiency, as measured by the proportion of squid becoming luminous after 18 h, was reduced when the squid had been inoculated for 3 h in the presence of 1 nM VHA-inhibitor ConcA. Letters (a, b) represent statistically different mean values (p < 0.05); each point represents a replicate; n = between 7–15 squid/treatment per replicate. Error bars represent 95% confidence intervals. Juveniles treated for 3 h with ConcA, washed, and then inoculated (Relief) were as efficiently colonized as untreated animals (B’). Average number of colony-forming units (CFUs) per light organ after 18 h under different treatments. V. fischeri cells were not present in the light organs of the uncolonized (non-luminous) ConcA-treated juveniles from (B); n = 5 squid/treatment.

Fig. 7. The impact of VHA activity on V. fischeri aggregate formation at or near the pores on the light-organ surface.

Measurements were performed based on confocal images. A Two strains of V. fischeri showed significantly reduced aggregate size in the presence of the VHA inhibitor ConcA: wild-type (WT) strain ES114 and its ΔmotB derivative (a nonmotile mutant in the flagellar motor); n = 18; **p < 0.01. In contrast, while ConcA decreased the average size of aggregates made by either another ES114 derivative, ΔflrA (a mutant lacking any flagella; n = 19, p = 0.33), or wild-type MB13B2 (a strain that typically makes very large aggregates; n = 12, p = 0.93), the differences were not statistically significant (ns). Each point is representative of the average aggregate size for each light organ (including both left and right sides) at 3 h post-hatch. B V. campbellii strain KNH1, n = 23, typically formed a smaller aggregate than V. fischeri ES114, n = 19, p = 0.03; however, while VHA inhibition by ConcA resulted in a significantly reduced aggregate formation by ES114, n = 20, ***p < 0.001, KNH1 showed no such effect, n = 23, p = 0.54. (C) V. fischeri ES114 typically dominates over V. campbellii KNH1 in co-aggregation assays, n = 20; however, in the presence of ConcA, V. campbelli became dominant, n = 20, **p < 0.01, suggesting that the advantage shown by V. fischeri cells was dependent on normal VHA activity of the host. Each point represents the proportion of KNH1 cells relative to the total number of bacteria in the aggregate at 3 h post-hatch. (D) In a typical animal used in (C), at one pore (p), V. fischeri was the only aggregating bacterium (D’), while at another, V. campbellii was also present (D”). E, E’ In contrast, with the addition of ConcA, V. campbellii became numerically dominant over V. fischeri in the majority of aggregates. aa, anterior appendage; pa, posterior appendage.