Exploring different methods of Exaiptasia diaphana infection to follow Vibrio parahaemolyticus dissemination in the whole animal

Abstract

An increase in wastewater rejection and rising seawater temperature are the two main causes of the spreading of pathogenic bacteria in the ocean that present a risk to the health of marine organisms, i.e., corals. Deciphering the infectious mechanism is of interest to better disease management. The quantity of infecting bacteria as well as method of pathogen administration is an important parameter in studying host–pathogen interactions. In this study, we have tested two models of infection (bathing or injection) of Exaiptasia diaphana (E. diaphana) with a clinically isolated strain of Vibrio parahaemolyticus expressing constitutively a Green Fluorescent Protein (Vp-GFP). We followed Vp-GFP dissemination over time with confocal microscopy at 6, 24, and 30 h. During the early time of infection, bacteria were observed adhering to the ectoderm in both infection methods. In later stages of the infection, Vp-GFP were lost from the ectoderm and appeared in the gastroderm. Compared to bathing, the injection method was supposed to provide better control of the bacteria quantity introduced inside the animal. However, injection induced a stress response with contraction and rejection of bacteria thus making it impossible to control the number of infecting bacteria. In conclusion, we recommended using the bathing technique that is closer to the infection route found in the environment and, moreover, did not cause injury to the animal. We also demonstrated, by using Vp-GFP, that we could track pathogenic bacteria in different tissues of E. diaphana over the time of infection and quantify them in the whole animal, thus opening a technical approach for developing new strategies to fight infection disease.

Introduction

Rising global temperatures are causing ocean acidification, melting ice, fires, and changing rainfall patterns on land, which are becoming more frequent and intense, contributing to increased sewage and agricultural runoff into the oceans [1,2,3,4]. These runoffs introduce a variety of pathogens, including enteric bacteria (such as Salmonella spp., Brucella spp., pathogenic Escherichia coli, and Vibrio spp.), which pose a significant health risk [5, 6]. Other consequences of climate change include significant changes in ocean circulation patterns and currents, leading to the spread of bacteria such as Vibrio spp. from the southern to the northern oceans [7]. Amongst Vibrio spp., Vibrio parahaemolyticus (V. parahaemolyticus or Vp) is a marine bacterium pathogenic to humans that is frequently responsible for foodborne diarrheal outbreaks, usually caused by contaminated or undercooked seafood (usually shellfish) [8, 9]. Their mechanism of infection has been partially described in humans and includes several virulence factors: i) adhesins and pili involved in adhesion and biofilm formation, and ii) various effector proteins injected into the host through secretion systems that induce inflammation, apoptosis, and autophagy [10].

Deciphering the infection mechanisms of Vibrio is particularly important, as members of this genus are pathogenic not only for humans but also for various marine organisms [8, 11, 12]. Recent studies have shown that Vibrio spp. can infect mussels, shrimps, lobsters, crabs, and cnidarians (e.g., Pocillopora damicornis or Exaiptasia diaphana) [13,14,15,16,17,18]. Cnidarians provide crucial ecosystem services like coastal protection, fisheries support, and tourism. However, they are increasingly afflicted by diseases due to stressors such as rising temperatures, acidification, and sewage pollution. Cnidarians are known to live in symbiosis with algae which are crucial for their health and coloration. The phenomenon of color loss occurs when cnidarians experience stress due to environmental factors such as increased water temperature, pollution, light conditions and infection [19, 20]. Vibrio coralliilyticus is an example of a Vibrio species known to infect the coral P. damicornis, resulting in expulsion of algal symbionts and bleaching [21]. Other species of Vibrio, such as V. alginolyticus, are associated with both white syndrome and yellow band disease in Porites andrewsi [22, 23]. Corals become particularly vulnerable to yellow band disease when their immune systems are weakened by environmental stressors, such as increasing ocean temperatures [23].

In order to unravel the complexity of bacterial infections in cnidarians and develop effective strategies to mitigate their impact, it is essential to use a suitable model organism that provides valuable insights into host–pathogen interactions. In this context, the sea anemone E. diaphana has emerged as an advantageous model for studying bacterial infections and their impact on the health of cnidarians [24,25,26,27]. While natural corals are less and less available for laboratory research, E. diaphana can be cultivated in large quantities and live in facultative symbiosis with Symbiodiniaceae algae. Like corals, E. diaphana has an open circulatory system but lacks the calcium carbonate skeleton characteristic of corals. Data from our group has shown that V. parahaemolyticus can infect the anemone E. diaphana and induce overexpression of genes involved in the regulation of apoptosis, inflammation, and immune response [28]. We have also reported that an increase in seawater temperature (+ 4 °C) promotes the virulence of V. parahaemolyticus by causing an upregulation of factors involved in adhesion and biofilm formation [29]. This increase in virulence of the bacterium, combined with the reaching of immunity limits at higher temperatures observed in various cnidarians, could contribute to the decline of cnidarians such as corals or anemones in their natural environment [11, 13].

It is important to consider the structure of E. diaphana for a better understanding of their interactions with pathogens. E. diaphana comprises two main tissue layers: the ectoderm, covered by a protective mucus layer, and the gastroderm, which houses symbiotic algae (Symbiodiniaceae) [30] (Fig. 1). These layers are separated by the mesoglea, a structure rich in muscle cells and sparse in phagocytic cells [31]. Both the ectoderm and gastroderm contain various cell types, including epithelial cells, mucocytes, nematocysts, and granule-secreting cells secreting digestive enzymes [32]. These cell types may be involved in maintaining symbiosis, microbial tolerance, and initiating innate immune responses [33, 34].

Fig. 1

Schematic cross-section of the pedal disk of the anemone. Exaiptasia is composed of two cell layers, the ectoderm and the gastroderm, separated by the mesoglea. The ectoderm contains epithelial cells, different types of mucocytes, and nematocysts. Mesoglea and mesenteries contain phagocytic cells (amebocytes). Algal symbionts are localized in the gastroderm cells. Mesenteries float in the internal cavity named the coelenteron. Based on microscopy observations and [35]

Research into the localization and visualization of native bacteria within cnidarian tissues, especially in corals and the model organism Aiptasia, remains limited. Understanding these microbial communities is important, as they play significant roles in the health and ecology of coral reefs. Techniques like fluorescence in situ hybridization (FISH) have emerged as essential tools for identifying and visualizing various microbial taxa associated with coral hosts [36]. This method allows researchers to delve into the complex interactions between microorganisms and their cnidarian hosts, facilitating the identification of both symbiotic algae and bacteria within the coral holobiont.

The technique of Scanning Electron Microscopy (SEM) underscores the importance of the topography of the tissue surface, which serves as a habitat for bacteria, influencing their distribution and interactions within the host in symbiotic or aposymbiotic anemone [37].

Among the advancements in this field, recent studies have significantly enhanced our understanding of host–pathogen interactions using GFP fluorescence [38]. One notable study elucidated a mechanistic pathway for how coral pathogens are acquired through the ingestion of zooplankton by E. diaphana. In this experiment, anemones were exposed to fluorescent Vibrio-colonized Artemia in the surrounding water, demonstrating a direct link between zooplankton and pathogen transmission [38].

Another innovative approach involved injecting a stable GFP-labeled strain of V. coralliilyticus into the tissues of Aiptasia, a type of sea anemone, to visualize the interactions between the pathogen and the coral host. This was achieved using epi-fluorescence video microscopy, allowing researchers to observe these interactions in real-time and gain insights into the dynamics of pathogen colonization and host response [35].

Today, there is a research gap regarding the comparative analysis of infection methods, in particular the different effects of bathing and injection techniques on the infection processes in E. diaphana. This study addresses the question of how these methods affect the number of infecting bacteria, the spread of the pathogen over time and the associated stress responses in the host organism. For this purpose, we use a Vp that constitutively expresses the green fluorescent protein (Vp-GFP) [39].

In the first method, which mimics the "natural" pathway of infection, we added the bacteria directly to the medium. This method was previously used in the experimental infection of E. diaphana by V. coralliilyticus and resulted in darkening of tissues, retraction of the tentacles, and degradation of the polyps [12]. It was also used to infect Anemonia sulcata with E. coli and Vibrio alginolyticus, which caused an inflammatory zone in the animal's body and triggered a molecular response against the pathogenic bacteria [40]. In the second method, we injected a solution containing the bacteria directly into the anemone pedal. This technique of injection has already been used for stem cell transplantation in the anemone Nematostella vectensis [41]. We applied both infection methods to track the distribution of Vp-GFP in the whole E. diaphana.

Methods

Bacterial strain

The clinically isolated strain Vibrio parahaemolyticus RIMD2210633 serotype O3:K6 expressing GFP (Vp-GFP) was provided by Kim Orth from UT Southwestern, Dallas, Texas [39]. The strain was maintained at -80 °C in a 25% glycerol solution. To revive from storage, Vp-GFP was inoculated into 5 mL of Luria–Bertani medium (LB) containing 3% NaCl supplemented with 50 µg/mL of spectinomycin (standard condition) for 6 h. Fluorescence did not drift over the duration of the experiments.

Animal husbandry

Exaiptasia diaphana strains CC7 were acquired from the Pringle laboratory at Stanford University. Sea anemones were maintained in 2 L tanks incubated in 0.22 µm filtered seawater (FSW) at 27 °C and with 60 μmol m − 2 s − 1 of light (12 h/12 h). FSW was changed, and the tanks were cleaned twice per week with cell scrapers and pipettes. If any anemones were dying, they were removed from the tanks. The anemones were fed with artemia twice a week. They were starved one week before the experiment to remove the effect of Artemia feeding.

E. diaphana infection with V. parahaemolyticus-GFP

Prior to the exposure with Vp, we selected twelve anemones per condition. The anemones oral disk diameter was half a centimeter (relaxed bodies). Anemones were transferred from the stock tanks into 12-well plates with 2 mL of FSW supplemented with 50 µg/mL of spectinomycin. The use of spectinomycin was for selection pressure of the GFP expressed by Vp following the protocol of Chimalapati et al. (2020). Anemones were incubated at culturing conditions overnight until acclimated and settled to the bottom of the well plate.

Vp-GFP solution was subcultured overnight in LB medium supplemented with 3% NaCl and 50 μg/mL of spectinomycin (150 rpm, 37 °C). Cultures were harvested by centrifugation (2500 g, 15 min) and were resuspended in 1 mL of FSW supplemented with 50 μg/mL spectinomycin to an OD620 nm ~ 0.8, corresponding to 10⁹ CFU/mL. 100 µL of the bacterial solution containing 10⁸ bacteria were seeded in the seawater for infection by bathing or, in case of internal infection, injected into the E. diaphana pedal with an insulin syringe. All animals, (“Control” and “Infected”), were maintained in FSW supplemented with 50 μg/mL spectinomycin during the experiment to exclude any toxicity effect of the antibiotic on the animal. 100 µL of FSW containing the antibiotic were injected or added to an external medium in “Control” animals (Fig. 2).

Fig. 2

Experimental design for assessing the V. parahaemolyticus distribution over infection of E. diaphana. Individual anemones (strain CC7) were kept in 12-well plates with 2 ml of FSW supplemented with 50 µg/mL of spectinomycin in each well. Controls are represented by Ø (anemone kept in FSW without added bacteria) and FSW-injected anemones (injection of 100 µL FSW). For infected anemones, 100 µL of the bacterial solution containing 10⁸ CFU of Vp-GFP was seeded in the FSW for bathing or directly injected into the E. diaphana pedal disk. Plates were incubated at 27 °C. For each time point (6, 24, and 30 h), 12 anemones were sampled. Bacterial visualization and quantification were analyzed at 6, 24, and 30 h post-inoculation

Fluorescent staining and immunohistochemistry

After 6, 24, and 30 h, the anemones were relaxed with 7.14% of MgCl₂ in FSW for 10 min and fixed overnight in 4% PFA in FSW. Anemones were placed vertically into a mounting cup (Simport Scientific) and were frozen in Cryomatrix Frozen Embedding Medium (Thermo Scientific™ Shandon). 10 µm thick sections were cut using a cryostat (Leica CM350). Slides were washed twice with 300 µL of PBS Tween (PBS + Tween 0.1%) and incubated with 300 µL of 0.1% Triton X-100 for 20 min to permeabilize the cells. After incubation, slides were washed twice with PBS Tween prior to a blocking step with 300 µL of 1% Bovine Serum Albumin (BSA) diluted in PBS Tween for 1 h. Ectodermal epithelial cells were labeled with phalloidin conjugated to Alexa Fluor 647 (Invitrogen™ Ref 10,053,252), and gland cells (highlighted by the lectins from the mucus) were labeled with wheat germ agglutinin (WGA) conjugated with Texas Red (Invitrogen™ Ref W21405), both via 30 min of incubation. The WGA label was chosen following the protocol described by [42]. For DNA labeling, 4′,6-diamidino-2-phenylindole (DAPI) was used (Invitrogen™ Ref D21490). After washing twice in PBS Tween, the slides were mounted in a small drop of Mowiol mounting medium and observed under a Nikon A1R Eclipse Ti2 microscope. ImageJ (version 2.14.0) was used to process and analyze our acquisitions.

Bacterial quantification by spectral cytometry

After 6 and 24 h of infection, whole anemones were washed two times in 10 mL of filtered seawater by vortexing for 30 s. Once washed, the anemones were crushed and filtered on a 40 µm filter. Cellular homogenates were analyzed on an Aurora spectral cytometer (Cytek) using SpectroFlow software (version 3.1.0). The cytometer was equipped with 4 lasers with different emission wavelengths (405 nm, 488 nm, 560 nm, and 630 nm). The spectrum of Vp-GFP was identified and used as a Fluorescence Tag.

Statistical analysis

For statistical analysis, we used the Prism software (version 10.3.1). The results were expressed as the mean ± standard error of the mean (SD). Statistical tests were performed by analysis of means (unpaired t-test assuming a gaussian distribution). The p-values < 0.05 were considered as statistically significant.

Results

Morphological observation over the period of infection

Localizing infecting bacteria in different tissues and quantifying the number of infecting bacteria is a critical issue in monitoring the host response to infection. To better control the number of bacteria infecting the sea anemone E. diaphana in experimental inoculations, we tested two infection methods: bathing and injection (Fig. 2). First, we monitored the morphological responses of animals infected by both methods over the duration of infection at 6, 24, and 30 h (Fig. 3).

Fig. 3

Morphological observation of E. diaphana control and infected animals through injection or bathing after 6, 24 and 30 h in 12-well plates. Both methods of infection impact the morphology of anemones, inducing darkening of tissues and tentacles retraction after 30 h. After Vp-GFP injection we observed white secreted substances immediately post infection, probably corresponding to mucus secretions by anemones expelling bacteria. Data are representative of 12 replicates per condition

We observed that regardless of the infection method, the animals showed a stress response characterized by tentacle retraction, shrinkage, and cell detachment. However, in E. diaphana that received Vp-GFP by injection, we observed that at the time of infection, the animal retracted, causing an expulsion of a "mucus-like" white substance containing bacteria (Fig. 3).

Distribution of V. parahaemolyticus-GFP by confocal microscopy

We used Vp-GFP to track the bacteria distribution in the whole E. diaphana animal. The distribution of Vp-GFP in different tissues was observed by confocal microscopy for both infection methods (Fig. 4). The autofluorescence emitted by anemones is less intense than the Vp-emitted fluorescence, allowing us to distinguish them finely. Furthermore, Vp-GFP expresses fluorescence even when the bacteria are fixed. In the early postinfection (PI) period, we visualized Vp-GFP attached to the ectoderm, while in the late period, Vp-GFP appeared in the gastroderm. In addition, after 6 h PI, the number of bacteria attached to the outer ectoderm membrane was higher in animals infected by injection than in animals infected by bathing (Fig. 4A, B). Vp-GFP was not found in “Controls”.

Fig. 4

Confocal microscopy acquisition of the E. diaphana infection kinetics. Acquisition of the ectoderm (A, B) and gastroderm (C, D) of the anemones after infection by injection (A, C) or bathing (B, D). Blue = nucleus, purple = actin, and green = Vp constitutively expressing GFP. The round purple fluorescence corresponds to algae, which are autofluorescent. Scale bar corresponds to 10 µm. Data are representative of 12 replicates. EcD = ectoderm; M = Mesoglea; Mes = Mesenteries; Algae = Algal symbiont. The orange arrow shows bacteria adhering to the ectoderm and the white arrows show the bacteria clusters

Furthermore, after 6 h of infection, we observed by confocal microscopy the co-localization between WGA (labeling lectins from the mucus) and GFP-expressing Vp bacteria (Fig. 5), indicating that the bacteria could potentially be trapped by mucocytes.

Fig. 5

Acquisition of the ectoderm of the anemones after 6 h of infection through bathing. Adhering Vp-GFP bacteria are colocalized with mucus secretions. The zoom focuses on the largest bacteria cluster trapped in the mucus. DAPI (blue) = Nucleus, WGA (red) = lectins from the mucus, and Vp-GFP (green) = Vp constitutively expressing GFP. EcD = Ectoderm; GaD = Gastroderm; M = Mesoglea; Algae = Algal symbiont. Data are representative of six independent experiments

After 24 and 30 h of infection, Vp-GFP has been lost from the surface of the ectoderm in both infection conditions, but its presence was detected in the gastroderm. Curiously, Vp-GFP was detected in the gastroderm of E. diaphana, which had undergone internal infection as early as 6 h after injection of the bacteria (Fig. 4C). In the anemone infected by bathing, we did not observe the bacteria in the gastroderm until 24 h after infection (Fig. 4D). In both infection methods, the Vp-GFP found in the gastrodermis were clustered. The cytometry technique allowed us to quantify the Vp-GFP in the whole animal (attached and inside the anemones). In animals infected for 6 and 24 h, we didn’t observe a difference in the number of bacteria between the two infection methods. However, within injected animals, we found a significant decrease in bacteria number after 24 h of infection when compared to 6 h (Fig. 6). The number of bacteria was not shown after 30 h of infection as the anemones were degraded.

Fig. 6

Quantification of Vp-GFP (attached and internalized) after 6 and 24 h of infection by bathing (white) or injection (grey). Percentages are shown as means ± standard deviations of experiments performed with three replicate samples. * p-value < 0.05

Discussion

Coastal ecosystems are under increasing seawater temperature and pressure from anthropogenic activities, including agricultural runoff, urban wastewater discharge, and industrial pollution. These activities introduce a variety of pollutants and pathogenic microorganisms into marine environments, leading to ecological disturbances [2,3,4]. Among the most concerning are bacteria from the Vibrio genus, which are known to proliferate in nutrient-rich waters, often resulting in harmful effects on marine life, including coral bleaching and disease outbreaks in marine invertebrates [21,22,23]. Understanding how cnidarians, such as Exaiptasia diaphana, respond to these pathogens is crucial for developing strategies to protect these vulnerable ecosystems.

In this study, we explored the infection trajectory of V. parahaemolyticus in E. diaphana, employing two different methods of bacterial exposure: bathing, which simulates natural environmental exposure, and injection, which directly introduces the bacteria into the host tissues. Our findings demonstrated that E. diaphana exhibited similar stress responses to both infection methods, characterized by tentacle retraction and tissue dissociation, particularly after 30 h of exposure (Fig. 3). These responses are consistent with previous studies showing that exposure to Vibrio species triggers similar stress reactions in cnidarians, leading to tissue degradation and, in severe cases, death [12].

By using GFP-expressing Vp, we were able to visualize and localize the bacteria within the host by confocal microscopy. Initially, bacteria adhered to the ectoderm in both bathing and injection methods, probably trapped by mucocytes present on the ectoderm, as shown in Additional file 1 and already reported in the literature [43]. Moreover, a higher bacterial load was observed on the ectoderm in the injected animals, probably due to the stress induced by injection, causing retraction of anemones, thus leading to the expulsion of a "mucus-like" substance (Figs. 3 and 4).

Mucus, an essential component of the innate immune defense in both humans and invertebrates, acts as a barrier that prevents pathogenic bacteria from accessing epithelial cells [44, 45]. In cnidarians, this mucus is rich in antimicrobial peptides (AMPs) and other immune factors that actively combat microbial invaders [45]. Our findings show that Vp-GFP co-localized with mucus on the ectoderm surface, supporting the hypothesis that mucus plays a crucial role in trapping and possibly neutralizing pathogenic bacteria shortly after infection (Fig. 5).

Interestingly, by the later stages of infection, the bacteria adhering to the ectoderm were no longer detectable. This could be due to several factors, including competition from the host’s commensal microbiota, which has been shown to outcompete potential pathogens and improve host fitness [46, 47]. Additionally, the host or associated microbes might have produced AMPs that recognized and eliminated the trapped pathogens. Sea anemones are known to produce a wide range of biologically active compounds with antimicrobial properties, which could play a role in clearing the infection [48, 49].

At a later time of infection, Vp-GFP was localized in the gastroderm. However, our results revealed differences in the kinetics of bacteria detection between the two infection methods. Vp-GFP was detected in the gastroderm as early as 6 h post-injection, while in the bathing method, bacteria did not appear in the gastroderm until 24 h post-infection. This suggests that the direct introduction of bacteria into the coelenteron via injection may allow them to bypass the initial defenses provided by the ectoderm and mucus, leading to earlier recognition by the immune cells in the gastroderm [45].

In cnidarians, the gastroderm is an important site for immune recognition and response, containing phagocytic cells known as amebocytes. These cells play a central role in engulfing and destroying pathogens [31, 50, 51]. Our observations, in both infection methods of Vp-GFP forming clusters in the gastroderm, suggest that these bacteria might be engulfed by amebocytes or similar immune cells. The size and morphology of these clusters (reaching approximately 10 µm) are similar to phagocytic cells observed in other marine invertebrates, such as sea urchins and the sea anemone Nematostella vectensis [52, 53]. Recently, by using spectral cytometry, we identified those Vp-GFP clusters as “amebocyte-like” cells mainly located in the mesenteries of E. diaphana [11]. These cells internalized the bacteria and interestingly presented a spectral profile of human monocytes.

Phagocytosis of bacteria by endodermal epithelial cells from the gastric cavity has already been reported in Hydra, demonstrating the implication of this tissue in the immune response to pathogens and non-pathogenic organisms [54]. This is supported by the localization of commensal algal symbionts found mainly along the gastroderm (Fig. 1). Phagocytosis was identified as the primary mechanism required for the establishment of symbiosis with the dinoflagellates [55]. Moreover, the Symbiodiniaceae presence can alter the immune responses of E. diaphana to the infection, opening a new area of research [56, 57].

In cnidarians, an amebocyte-like population comprises various cell types that migrate to injury sites in response to wound healing, bleaching, and disease [58]. The authors reported that corals respond to wound healing by immune cell infiltration around the wound site 6 h post-injury. Amongst the infiltrating cells they identified amebocytes similar to those documented within the mollusk Mytilus edulis where they are described to have phagocytic activity and strong superoxide radical production. As shown in Fig. 6, the number of Vp-GFP decreased significantly after 24 h in anemones infected by injection. We can’t exclude that this phagocytic amebocyte that infiltrates the wound healing by the syringe injury can internalize the Vp-GFP and, by the way, impact the host’s immune response.

In this study, we highlighted the importance of investigating the infection processes in E. diaphana as a model for understanding bacterial infections in cnidarians. By using GFP-expressing Vp, we were able to quantify bacteria in the whole organisms and localize them within the host tissues, validating both bathing and injection as effective infection methods. The stress induced by the injection underlines the need to be careful because, contrary to what was expected, it does not make it possible to control the quantity of bacteria. Furthermore, in this study, we also presented a bacterial visualization protocol which is faster and less tedious than other visualization techniques as FISH. Our findings underscore the importance of the ectodermal mucus barrier and the gastrodermal immune cells in defending against bacterial pathogens. Additionally, the injection method, despite inducing stress, offers a unique model for studying wound healing (caused in the environment (e.g., anchor) or in vitro) and its impact on the immune response.