Microbiome of wild oysters: assessing diversity and detecting potential pathogens using a metabarcoding approach

Abstract

Background

The Pacific oyster (Crassostrea gigas) and the hooded oyster (Saccostrea cucullate) are ecologically and economically important species in the northwestern Pacific Ocean. However, sustainable oyster farming faces challenges from pathogenic diseases and environmental changes. Understanding microbial diversity in oysters is essential for managing pathogens, maintaining healthy microbial communities, and addressing microbial imbalances. This study aimed to investigate the diversity and composition of bacterial and protist communities in wild oysters from South Korea (C. gigas), Taiwan, and the Philippines (S. cucullata) using a metabarcoding approach.

Results

Gill tissue and environmental DNA (eDNA) samples were analyzed to assess microbial community structure across species and regions. Bacterial richness significantly exceeded protist richness in all samples. Bacterial diversity was negatively correlated with sea surface temperature and positively correlated with latitude, indicating that temperature is a key driver of regional variation in bacterial community composition. Several potentially pathogenic protist and bacterial taxa were detected, including Perkinsus marinus, Bonamia ostreae, Haplosporidium costale, and Vibrio bathopelagicus. Notably, P. marinus and B. ostreae were identified in S. cucullata from Taiwan, while H. costale was detected in C. gigas from South Korea. Most pathogens occurred at low infection intensities and without clinical signs of disease. Challenges in detecting protist diversity due to sequencing depth and host-derived nontarget amplification were also noted, highlighting the importance of using protist-specific primers.

Conclusions

This study demonstrates the utility of metabarcoding for characterizing microbial communities and detecting pathogens in two oyster species, C. gigas and S. cucullate, across geographically distinct regions. Bacterial richness was influenced by environmental gradients such as sea surface temperature and latitude, while several key protist pathogens (P. marinus, B. ostreae, H. costale) were detected at low infection intensities in asymptomatic oysters. Limitations in protist detection due to sequencing depth and host amplification highlight the need for optimized primers and complementary approaches such as multiomics and microscopy. These findings provide a foundation for understanding host–microbe–environment interactions. Specifically, our results can inform targeted monitoring programs for the early detection of pathogens and guide selective breeding initiatives for disease-resistant oyster strains, thereby enhancing the long-term resilience of oyster aquaculture under changing environmental conditions.

Background

The Pacific oyster (Crassostrea gigas) is a key species in the aquaculture industry of East Asia, particularly in the Republic of Korea, where it is the predominant molluskan shellfish. In 2023, the production of Pacific oysters in Korea reached 310,753 metric tons [1]. Globally, C. gigas remains the most cultured oyster species, with China contributing approximately 83.5% of the total global oyster production in 2019 [2]. In addition to its economic value, oyster aquaculture offers ecological benefits, such as water quality improvement [3], and provides a sustainable protein resource with relatively low greenhouse gas emissions [4, 5].

To broaden the ecological and geographical scope of microbial surveillance, the hooded oyster Saccostrea cucullata was also included in this study. This species is widely distributed throughout tropical and subtropical regions of the Indo-Pacific, including Taiwan and the Philippines, where it contributes to local aquaculture production and plays an important ecological role in coastal ecosystems—although its commercial importance is generally lower than that of C. gigas [6]. These regions, like South Korea, are influenced by the Kuroshio Current—a warm western boundary current that flows northward from the Philippines past Taiwan and along the eastern coast of Japan. This current connects marine ecosystems across the western Pacific and is known to transport marine larvae, plankton, microorganisms, and even parasitic life stages over long distances [7]. For instance, Kato and Yamakawa [8] reported the northward transport of the freshwater parasite Lernaea cyprinacea on marine fish along the Kuroshio Current, demonstrating its role in cross-regional parasite dispersal. Similarly, recent studies have emphasized the importance of ocean currents, including the Kuroshio, in shaping microbial and parasite biogeography in marine ecosystems, with implications for disease emergence in aquaculture [9]. This oceanographic linkage raises the possibility of disease dispersal across national or regional boundaries, potentially introducing novel or emerging pathogens into oyster populations in South Korea, Taiwan, and the Philippines.

However, the oyster farming industry faces substantial challenges from pathogenic diseases, including perkinsiosis and bonamiosis, both which pose significant threats to oyster production. Perkinsiosis, caused by the protozoan Perkinsus marinus [10, 11], infects the hemolymph and tissues of oysters, often resulting in high mortality rates under warm-water conditions and leading to considerable economic losses. Bonamiosis, caused by intracellular protists of the genus Bonamia [12], targets oyster blood cells and digestive tissues, frequently leading to chronic infections that compromise oyster health and survivability. Additionally, human-induced environmental changes have destabilized marine ecosystems, negatively affecting the sustainable use of fishery resources, including oysters [13]. Environmental stressors can weaken host resilience and increase susceptibility to novel or emerging diseases [14,15,16]. The Kuroshio Current, a major oceanic current in the North Pacific, further influences regional marine ecosystems by affecting nutrient transport, sediment dynamics, storm tracks, and climate. ​This current also facilitates the translocation of parasites across continents, potentially introducing new parasitic threats to molluskan species in South Korean regions [17].​

Microbiomes contribute significantly to host health, influencing digestion, metabolism, immune regulation, and defence against opportunistic pathogens [18]. The complexity of host–parasite interactions, which often involve multiple pathogens, challenges the traditional “one pathogen–one disease” model. A holistic understanding of infectious diseases requires the examination of community-level diversity within the microbiome, encompassing both protists and bacteria. Investigating the microbiomes of wild oyster populations is particularly important, as wild populations can act as reservoirs for pathogens that may spread to aquaculture systems [19]. Factors influencing microbiome diversity include environmental conditions, biological interactions, and anthropogenic impacts [20].

The microbiome is crucial for maintaining homeostasis [21, 22] and supporting host performance, including growth rates [23]. Advances in next-generation sequencing (NGS) technologies provide cost-effective screening of community-level diversity within the host and facilitate the sensitive detection of novel pathogens [24, 25]. These capabilities support early detection and rapid response (EDRR) strategies—proactive frameworks designed to identify biological threats at an early stage and implement timely interventions to prevent their establishment and spread [26]. In aquaculture, EDRR approaches are essential for mitigating the impacts of infectious diseases before they result in large-scale mortality or economic loss. For example, molecular tools such as eDNA and metabarcoding have been increasingly applied to improve the early detection of protozoan parasites in oysters and other aquatic species, enhancing surveillance and informing management responses [27]. NGS-based monitoring, therefore, offers a valuable tool for applying EDRR principles to safeguard oyster health and aquaculture sustainability [26].

Elucidating microbial diversity in wild oysters offers insights into managing pathogenic microorganisms, understanding healthy microbial communities, and identifying microbial signatures associated with poor health status [28]. Gill tissues were collected from two oyster species—C. gigas in South Korea and S. cucullata in Taiwan and the Philippines—and water samples were collected from South Korea for environmental DNA (eDNA) extraction. These regions are influenced by the Kuroshio Current, which may facilitate the dispersal of microbiomes and associated parasites. To account for species-specific and regional differences, we characterized microbial communities within each oyster species and geographic site. Comparisons were made between gill-associated and environmental microbiota within Korea to ensure ecological relevance and minimize confounding effects between host species and geographic region.

The primary aim of this study was to investigate the diversity and composition of bacterial and protist communities in wild oysters using a metabarcoding approach, and to evaluate how microbiome patterns and pathogen occurrence are shaped by host species, environmental gradients, and potential for regional pathogen dispersal. By integrating host-associated and environmental microbiome data, this study contributes to understanding host–microbe–environment interactions and supports the development of early detection tools and sustainable disease management strategies in oyster aquaculture.

Materials and methods

Sampling

Pacific oysters (C. gigas, N = 20) were collected from eight sites across the eastern (Kangreung [KR]), southern (Busan [BS], Tongyoung [TY], Geomun Island, Yeosu [YS]), and western (Jangja Island, Gunsan [JJ], Gaeya Island, Gunsan [GY], Hongseong [HS], Incheon [IC]) coasts of South Korea. Hooded oysters (S. cucullata, N = 8) were collected from three sites: Kaohsiung (TG) in southwestern Taiwan, Yilan (TL) in northeastern Taiwan, and Manila (PH) in the northern Philippines. Sampling was conducted between April and June 2023, and exact collection dates for each site are summarized in Supplementary Table S1 (Fig. 1 A). Host species identification was performed using mitochondrial COI barcoding with the primers LCO1490 (5′ GGT CAA CAA ATC ATA AAG ATA TTG G 3′) and HCO2198 (5′ TAA ACT TCA GGG TGA CCA AAA AAT CA 3′) under conditions described in a previous report [29]. Sampling locations were selected based on ecological and epidemiological relevance. Sites in South Korea included major oyster production areas and locations near high-traffic international ports (Incheon and Busan), where ballast water discharge may facilitate the introduction of non-native pathogens [30]. In addition, sampling sites in Taiwan and the Philippines were chosen along the Kuroshio Current, a major oceanographic feature that transports marine larvae, microorganisms, and potential pathogens across the northwestern Pacific Ocean. This site selection aimed to capture microbial diversity influenced by both aquaculture intensity and oceanic connectivity (Fig. 1A).

[IMAGE OMITTED: SEE PDF]

Environmental and geographical variables, such as sea surface temperature (SST), chlorophyll-a concentration, and latitude, were obtained from public databases (https://www.seatemperature.org/asia/; https://meis.go.kr/) and previous studies [31, 32] (Fig. 1B–D). Environmental measurements represent monthly averages for the sampling period, collected from monitoring stations within 5 km of each sampling site. SST and chlorophyll-a concentration were selected due to their known influence on microbial diversity and primary productivity in marine environments. Latitude was included to capture broader biogeographic variation and its potential correlation with temperature.

Oysters were collected from locations where no mass mortality events had been reported, to increase the likelihood of capturing baseline microbiome diversity representative of healthy or stable populations. Each collected oyster was individually transported in sterile plastic bags to the laboratory and maintained at 4 °C for no more than 24 h before processing. Samples were thoroughly cleaned with autoclaved seawater to prevent cross-contamination and gill tissues were isolated. DNA extraction was performed using a Qiagen DNA extraction kit (Hilden, Germany) following the manufacturer’s protocols without modification, and negative extraction controls included to assess potential contamination. No significant symptoms or lesions were observed in the oyster samples. The extracted DNA was stored at −20 °C until library preparation. In addition, 4 L of seawater samples were collected in sterile buckets near the oyster sampling locations. The seawater samples were transferred to the laboratory under dark and cold (5 °C) conditions. Then, 2 L of seawater was filtered through a 0.2 μm pore size membrane. DNA extraction from the eDNA filter was performed using the Macherey-Nagel Nucleospin Tissue Kit (Macherey-Nagel, Duren, Germany) as per the manufacturer’s instructions. The concentration of the DNA extracted from both the tissue and water samples was measured using the Infinite® 200 Nanoquant (Tecan Group Ltd, Männedorf, Switzerland).

Metabarcoding sequencing

Using 16 S rRNA gene sequencing to target bacterial communities and 18 S rRNA gene sequencing to target protists, we applied a metabarcoding approach to characterize the microbial diversity associated with oyster gill tissues and seawater samples. Next-generation metabarcoding sequencing libraries were prepared using the Herculase II Fusion DNA Polymerase Nextera XT Index V2 Kit (Illumina Inc. San Diego, California, USA), following the 16 S metabarcoding sequencing library preparation protocol (Part # 15044223 Rev. B). To amplify the protistome amplicon, 18 S universal primers targeting the V9 region were used (forward: 5′-CCCTGCCHTTTGTACACAC-3′; reverse: 5′-CCTTCYGCAGGTTCACCTAC-3′). For the bacteriome amplicon, 16 S universal primers targeting the V3–V4 regions were used (forward: 5′-CCTACGGGNGGCWGCAG-3′; reverse: 5′- GACTACHVGGGTATCTAATCC-3′). The PCR conditions for 18 S amplification were as follows: 94 °C for 3 min, followed by 30 cycles at 94 °C for 30 s, 57 °C for 60 s, and 72 °C for 90 s, with a final elongation at 72 °C for 10 min. The PCR conditions for 16 S amplification were as follows: 95 °C for 3 min, 25 cycles for 95 °C for 30 s, 55 °C for 30 s and 72 °C for 30 s; and a final elongation at 72 °C for 5 min. Amplicon sequencing was performed using the Illumina MiSeq platform (301 bp paired-end) by Macrogen (Seoul, Republic of Korea). Negative PCR controls and mock community controls were performed to assess amplification bias and contamination.

Bioinformatic and statistical analysis

To minimize false signals, sequencing reads were trimmed using BBDuk (https://sourceforge.net/projects/bbmap/) with the following parameters: minimum quality = 30, minimum overlap = 24, minimum length = 100 bp, k-mer length = 27, maximum substitutions = 1, and maximum substitutions + INDELs = 0. Trimmed paired-end reads were merged with BBMerge (BBMap version 37.47) [33], and only successfully merged reads were carried forward for downstream analyses. Chimeric sequences were identified and removed using the UCHIME algorithm [34]. Amplicon Sequence Variants (ASVs) were inferred using the DADA2 pipeline implemented in QIIME2, which enables single-nucleotide resolution and avoids clustering ambiguities associated with traditional operational taxonomic units. This approach retains high-quality sequence variants regardless of database match status, allowing for improved ecological resolution and reproducibility. Taxonomic classification was performed using the SILVA database (version 138) for 16 S rRNA sequences and the PR2 database (version 5.0.0) for 18 S rRNA sequences. Classified ASV abundance tables were formatted for downstream statistical analysis using the R packages ranacapa [35] and phyloseq [36]. Taxonomic information was organized using TaxonKit [37], and separate ASV tables were generated for bacterial and protist communities.

Bacterial and protist abundances and alpha diversity were calculated using the R package phyloseq [36]. In this study, alpha diversity refers to within-sample diversity and was quantified using three complementary metrics:

*

Observed ASVs – a measure of species richness (number of distinct ASVs),

*

Shannon diversity index (H) – a combined measure of richness and evenness,

*

Simpson index – a measure of evenness or dominance within the community.

Statistical comparisons of each alpha-diversity metric between groups (e.g., sample type, region) were conducted using the Wilcoxon rank-sum test for two-group comparisons and analysis of variance (ANOVA) for multi-group comparisons, with post-hoc Tukey’s HSD tests where appropriate. Effect sizes were calculated as Cohen’s d for Wilcoxon tests and η² for ANOVA, with 95% confidence intervals.

Beta diversity was evaluated using Bray–Curtis dissimilarities and visualized via nonmetric multidimensional scaling (NMDS) ordination using the ordinate function in phyloseq. Differences in microbial community composition across sites and sample types were assessed using permutational multivariate analysis of variance (PERMANOVA) with 999 permutations. To address the confounding between oyster species and geographic region in our sampling design (C. gigas from Korea and S. cucullata from Taiwan and the Philippines), our analyses focused on site-level comparisons and within-region patterns. This approach enabled us to explore microbial variation across individual sites and sample types, while recognizing that the differences observed may reflect environmental factors (e.g., sea surface temperature, chlorophyll-a concentration, or local anthropogenic influences) in addition to host species. R² values from PERMANOVA were used to assess the strength of beta-diversity patterns. Heatmaps were generated to compare the relative abundance of bacterial and protist taxa among samples, and microbiome co-occurrence networks were visualized using the plotnet function, based on the Jaccard distance method, in phyloseq [36].

Results

Protist and bacterial ASV diversity across regions

Substantial differences in microbial richness (observed ASVs) were observed across sampling regions and oyster species, with bacterial richness far exceeding that of protists in both eDNA and gill tissue samples (Fig. 2). In South Korea, where C. gigas was collected, 1,740 bacterial ASVs were extracted from eDNA samples and 1,429 from gill tissues, with 861 ASVs shared between both sample types (Supplementary Fig.S1). In contrast, only 69 protist ASVs were detected in eDNA samples from both the southern and western coasts, and just 12 were found in tissue samples, indicating that bacterial richness was approximately 25 times higher than protist richness in eDNA and over 100 times higher in tissue samples.

Bacterial richness in eDNA was higher on the west coast (1,364 ASVs) than the south coast (1,219 ASVs), while protist richness remained similar (69 ASVs in both regions) (Fig. 2 A and B). In tissue samples from C. gigas in South Korea, bacterial ASV richness (1,252) far exceeded protist richness (12 ASVs), emphasizing a marked taxonomic imbalance (Fig. 2D).

[IMAGE OMITTED: SEE PDF]

In contrast, richness in gill tissues from S. cucullata in Taiwan and the Philippines was lower than that observed in C. gigas from South Korea. However, since oyster species and geographic region are not independent in this study, these differences are interpreted as site-specific patterns rather than species-level effects. The Philippines had the lowest total ASV count (165 bacterial ASVs, 69 of which were private, and 2 private protist ASVs), while Taiwan exhibited moderate richness (301 bacterial ASVs, 97 private bacterial ASVs, and 6 protist ASVs) (Fig. 2 C, D). Overall, the ratio of bacterial to protist ASVs in tissue samples ranged from approximately 50:1 to 150:1 depending on region. Protist ASV diversity from eDNA showed broader environmental representation than tissue samples. Higher protist richness was observed at west coast sites near inland bays (IC, HS) and at the southern open-water site YS, while no protist ASVs were detected in eDNA from Busan (BS) (Fig. 3).

[IMAGE OMITTED: SEE PDF]

Alpha diversity and geographical distribution of bacterial ASVs

Alpha diversity was evaluated using observed ASVs, the Shannon diversity index, and the Simpson index across both eDNA and tissue samples. Bacterial richness (observed ASVs) was significantly higher in eDNA samples than in tissue samples (Wilcoxon rank-sum test: W = 156, p < 0.01, Cohen’s d = 1.8, 95% CI: 1.1–2.5; Fig. 4A). In contrast, no significant difference was detected for the Shannon index (W = 89, p = 0.12, Cohen’s d = 0.6, 95% CI: − 0.1 to 1.3), suggesting comparable evenness despite higher richness in eDNA. Results from the Simpson index confirmed this pattern: eDNA samples (0.802 ± 0.089, range: 0.635–0.891) showed significantly greater evenness than tissue samples (0.713 ± 0.174, range: 0.211–0.925) (Wilcoxon test: W = 134, p < 0.05, Cohen’s d = 0.78; Supplementary Table S4).

[IMAGE OMITTED: SEE PDF]

Significant variation in bacterial richness among countries was also detected (ANOVA: F(2,25) = 12.4, p < 0.01, η² = 0.50; Fig. 4B). As different oyster species were sampled in each country (C. gigas in Korea and S. cucullata in Taiwan and the Philippines), these results are reported for descriptive context only, as species and region are not independent. No significant differences were observed among countries for the Shannon or Simpson indices (Supplementary Table S5, S6).

Within South Korean sampling sites (C. gigas only), Shannon diversity was significantly higher in oysters from the southern coast compared to those from the western coast (Tukey’s HSD: p < 0.05, Cohen’s d = 0.9, 95% CI: 0.2–1.6; Fig. 4 C). The Simpson index showed a consistent but non-significant trend in the same direction (Supplementary Table S4). These patterns suggest that site-specific environmental conditions influence richness more strongly than evenness in oyster-associated bacterial communities.

Environmental correlates of bacterial ASV richness

The bacterial ASV richness varied significantly across the study sites and was correlated with both environmental and geographic factors (Fig. 5). The number of bacterial ASVs ranged from 77 to 488, with the highest richness observed in Tongyoung (TY) and Gaeya (GY) (488 and 439 ASVs, respectively) and the lowest richness in Jangja (JJ) (Fig. 5 A, Supplementary Table S1). A significant negative correlation was observed between sea surface temperature (SST) and bacterial ASV richness (Pearson’s correlation: r = − 0.42, p < 0.05, 95% CI: −0.71 to −0.08), indicating that bacterial diversity decreases with increasing SST (Fig. 5B). In contrast, no significant correlation was detected between the chlorophyll-a concentration and bacterial richness (Pearson’s r = 0.012, p = 0.94, 95% CI: −0.36 to 0.38, Fig. 5 C). A positive correlation between latitude and bacterial ASV richness (Pearson’s r = 0.41, p < 0.05, 95% CI: 0.06–0.69) was noted, suggesting collinearity between latitude and SST (Fig. 5D).

[IMAGE OMITTED: SEE PDF]

Environmental correlates of bacterial evenness

In addition to richness, we evaluated bacterial community evenness using the Simpson index and tested its association with environmental variables (Supplementary Table S7). Simpson evenness showed no significant correlations with SST (r = − 0.185, p = 0.361), chlorophyll-a (r = 0.098, p = 0.634), or latitude (r = 0.147, p = 0.470). For comparison, richness (observed ASVs) was significantly associated with SST (r = − 0.423, p = 0.035) and latitude (r = 0.412, p = 0.041), but not with chlorophyll-a (r = 0.012, p = 0.954). These findings indicate that the environmental gradients examined relate more strongly to richness than to evenness in oyster-associated bacterial communities.

Microbial community network analysis

Microbial community network analysis based on Jaccard distances revealed clear patterns of spatial clustering among sampling sites (Fig. 6). Samples from the west and south coast of Korea showed substantial overlap and high connectivity, indicating shared bacterial community structures within C. gigas populations across Korean regions. This clustering likely reflects comparable environmental conditions (sea surface temperatures ranging from 15.3 to 18.5 °C) and the common host species identity (C. gigas) across Korean sites.

[IMAGE OMITTED: SEE PDF]

In contrast, Taiwanese samples exhibited heterogeneous network patterns. Most samples from Yilan, Taiwan (S. cucullata, TL) integrated into the central network cluster with Korean samples, suggesting partial microbial similarity despite differences in host species and geography. However, samples from Kaohsiung, Taiwan (TG) formed a distinct, disconnected cluster, indicating substantial divergence in bacterial community composition. This pattern may reflects a combination of (1) Species-specific effects—TG represents S. cucullata while most connected samples are C. gigas from Korea; (2) Environmental differences—Kaohsiung exhibits higher sea surface temperatures (26.7 °C) and distinct oceanographic settings; (3) Anthropogenic influences—Kaohsiung’s major industrial port and heavy shipping traffic may alter microbial inputs (e.g., via ballast water); and (4) Temporal variations—sampling occurred across different months (April-June 2023).

The Philippine sample (S. cucullata, PH) occupied an intermediate position between Korean and Taiwanese nodes in the network, indicating partial microbial similarity to both regions. This intermediate connectivity may be influenced by regional circulation (e.g., Kuroshio Current) that facilitate microbial exchange, and/or shared environmental gradients affecting S. cucullata populations across the Indo-Pacific region.

Overall, these network patterns complement the alpha- and beta-diversity results by highlighting broader spatial relationships in oyster-associated microbiomes and the interplay of host identity, environmental context, and potential anthropogenic inputs in shaping connectivity across the northwestern Pacific region.

Pathogenic protists and co-occurrence patterns

In gill tissue samples from South Korea, 17 parasitic ASVs were identified, with Rhinosporidium seeberi showing the highest relative abundance (4.48% of parasite assigned reads), whereas other taxa each accounted for < 2% (Fig. 7; Supplementary Ta

[IMAGE OMITTED: SEE PDF]

ble S2). In comparison, gill tissue samples from the Philippines contained two parasitic ASVs, and those from Taiwan contained eight ASVs, including two major oyster pathogens: P. marinus and Bonamia ostreae.

From eDNA samples collected in South Korea, over 200 protist ASVs were identified. Seven species including Condylostentor auriculatus, Euplotes charon, Margolisiella islandica, P. marinus, Pirum gemmata, Pseudobodo tremulans, and R. seeberi were detected in both eDNA and gill tissue samples, but their co-occurrence did not occur uniformly across all sites. For example, P. marinus was detected in both gill tissue and eDNA at IC, R. seeberi co-occurred at YS and IC, and Pirum gemmata was identified in both sample types at multiple west coast sites including HS and GY (Fig. 8; Supplementary Table S3). In addition, no clear geographic clustering was observed in relative read abundance of protozoan parasites. Universal primers targeting the V9 region of 18 S rRNA enabled broad detection of protist taxa, whereas primers targeting 16 S rRNA were specific to bacterial communities. Together, these results demonstrate the utility of metabarcoding for capturing both host-associated and environmental pathogen diversity.

[IMAGE OMITTED: SEE PDF]

Protist composition and abundance

A heatmap visualization of protist ASV abundances (Fig. 9) revealed compositional differences among samples, though no clear geographic clustering was observed. eDNA samples from the YS presented the greatest abundance of protists, including Thaumatomastix salina and Paramoeba pemaquidensis. Samples from Tongyoung (TY) and Incheon (IC), both near major ports, exhibited distinct protist assemblages, suggesting localized microbial communities potentially shaped by port-associated environmental factors or human activity. Although these assemblages were different from each other, their uniqueness at major shipping hubs raises the possibility that ports may act as entry points or reservoirs for non-native microbial taxa, including potential pathogens.

[IMAGE OMITTED: SEE PDF]

Bacterial community structure and diversity

We assessed compositional similarity and ecological coherence within oyster-associated microbiomes using non-metric multidimensional scaling (NMDS) on Bray–Curtis dissimilarities (Fig. 10). For each of the ten most abundant bacterial families and genera, ASV abundance matrices were constructed and ordinated separately; each point in the ordination represents a single ASV, and distances reflect its similarity to other ASVs within the same taxonomic group. We interpreted tighter groupings as evidence of ecological coherence and more diffuse patterns as potential signals of functional diversity or niche partitioning.

[IMAGE OMITTED: SEE PDF]

At the family level, clustering varied across groups. Families such as Arcobacteraceae, Endozoicomonadaceae, and Fusobacteriaceae displayed relatively tight groupings, consistent with greater ecological coherence. In contrast, Flavobacteriaceae, Paracoccaceae, and Roseobacteraceae were more widely dispersed, indicating broader functional diversity or adaptation to distinct microenvironments within oyster gill tissues (Fig. 10 A).

At the genus level, similar differences were observed. Endozoicomonas exhibited moderate clustering, whereas Pseudoalteromonas and Vibrio showed wider ordination spread, suggesting greater intra-taxonomic heterogeneity in ASV composition (Fig. 10B). Other genera, such as Geobacillus, Poseidonibactor, Massilia, and Ruritalea were represented by fewer ASVs and thus showed more limited clustering patterns.

This ASV-based ordination provides qualitative insights into the degree of taxonomic coherence but does not directly address our primary questions on pathogen detection or environmental drivers of community-wide diversity. The observed dispersion may partly reflect limits of taxonomic resolution or uneven sampling rather than true ecological differentiation. Accordingly, these NMDS patterns should be considered a complementary context for the alpha- and beta-diversity analyses rather than stand-alone evidence of ecological processes.

Discussion

Oyster aquaculture plays a crucial role in ensuring global food security [4, 5]. In this study, a metabarcoding approach was used to detect potentially novel or regionally unreported pathogens—including B. ostreae, P. marinus, and H. costale—and to examine the environmental and geographical determinants of bacterial and protist ASV richness in the northwestern Pacific Ocean, covering Korea, Taiwan, and the Philippines. Significant variations in bacterial ASV richness were observed, with the SST negatively correlated with bacterial diversity, indicating that bacterial richness decreases as the SST increases. However, no significant relationship was found between chlorophyll-a concentration and bacterial ASV richness, suggesting that primary productivity does not play a major role in shaping bacterial diversity in these oyster populations. Interestingly, latitude was positively correlated with bacterial ASV richness, likely due to its collinearity with SST. These results highlight the complex interplay of geographical and environmental factors influencing bacterial diversity in marine ecosystems, with temperature emerging as a key driver.

The inclusion of both Shannon and Simpson indices allowed us to capture complementary aspects of community structure, balancing richness and evenness, when assessing microbiome diversity across regions and sample types. While Shannon diversity showed significant regional variation within Korea (higher in southern compared to western coasts) and no significant differences between sample types (eDNA vs. tissue), the Simpson index revealed significantly higher evenness in eDNA samples than in tissue samples. Notably, environmental gradients such as SST and latitude were significantly correlated with bacterial richness but showed no significant associations with evenness (Simpson), indicating that temperature-driven changes primarily affect the number of taxa rather than their relative abundance distributions.

Given the species-specific nature of microbiome composition, we analyzed C. gigas (South Korea) and S. cucullata (Taiwan and the Philippines) separately to minimize confounding. However, because oyster species and geographic region are intrinsically linked in our sampling design, it is not possible to disentangle whether observed microbiome differences are driven by host identity, geography, or environmental conditions. Therefore, our interpretations focus on site-level patterns rather than direct comparisons between species or regions. For example, bacterial richness was highest in C. gigas from Tongyoung and Gaeya, while protist pathogens such as B. ostreae were detected only in S. cucullata from Taiwan. These patterns reveal that potentially pathogenic organisms—including P. marinus, B. ostreae, and H. costale—are present across distant locations and host species, despite the absence of visible disease symptoms. These findings support the need for localized microbiome surveillance and suggest that oceanographic features like the Kuroshio Current may facilitate long-range pathogen dispersal.

The network analysis complements the diversity and environmental correlation results by revealing broader spatial relationships among oyster-associated microbial communities. Notably, C. gigas populations from both the west and south coasts of Korea formed a highly connected cluster, reflecting strong similarity in bacterial community structure within Korean regions. In contrast, S. cucullata samples from Taiwan displayed more heterogeneous patterns, with most TL samples integrated into the network and TG samples forming a distinct, disconnected group. The Philippine sample (S. cucullate) occupied an intermediate position between Korean and Taiwanese samples, indicating partial microbiome similarity. Although species identity and regional factors are confounded in the current design, the partial grouping of samples by region—especially within Korean sites and some Taiwanese sites—supports a pattern of limited biogeographic clustering. These patterns, along with variation in alpha and beta diversity, indicate that microbiome connectivity is likely shaped by both environmental gradients (e.g., SST) and host-specific factors such as species and local adaptation.

Protists exhibited site-specific patterns of occurrence, while bacterial community profiles varied across regions and host populations. These differences may be influenced by host-associated factors, including tissue microenvironments and species-specific physiological traits. For example, the observed divergence in microbiome composition between C. gigas and S. cucullata across Korea, Taiwan, and the Philippines may reflect inherent differences in immune function, habitat preference, or microbial filtering capacity. Previous studies have shown that oyster microbiomes are shaped by complex host–microbe interactions, which can reflect long-term coevolution and local adaptation [16, 22]. Such interactions may contribute to microbiome stability, influencing both symbiont maintenance and pathogen colonization.

The present study revealed significantly lower protist richness than did previous reports, which detected approximately 6,000 protist ASVs in coastal benthic and surface samples [38]. This discrepancy may be attributed to differences in the targeted ocean regions and 18 S rRNA gene regions, and sequencing depth. On average, approximately 50,000 reads per sample were obtained, which is consistent with standard recommendations for microbial diversity profiling in marine invertebrates. While this depth was sufficient to detect dominant taxa, rare protist diversity may have been underestimated, as reported in previous studies [39]. A previous study also identified a higher number of ASVs at coastal sediment sites in the Yellow Sea [40]. In contrast, the number of ASVs detected in our study suggests that sequencing depth remains a critical factor for detecting rare or low-abundance taxa, particularly under low-infection conditions. In addition to these methodological factors, species-specific differences between C. gigas and S. cucullata—such as variation in gill structure, filtration rates, or susceptibility to protists—may also influence observed richness. Despite this limitation, our metabarcoding approach effectively identified several key protist pathogens, reinforcing its utility for early detection and disease surveillance in oyster populations.

Our findings also suggest that the V9 region of 18 S rRNA may not be optimal for studying protist diversity in bivalves such as Pacific oysters, because of similarities in primer-binding sites between bivalves and target protists, which can lead to substantial nontarget amplification. Future studies should consider the use of protist-specific primers to minimize this issue. Previous studies have shown that the number of protist ASVs can be increased by the use of primers targeting multiple marker regions [41] and protist-specific primers [22, 42], which could help avoid nontarget amplification in host species. Despite the challenges posed by host amplicons, this study successfully identified protist diversity from multiple samples. It is also possible that differences between oyster species, such as variation in genomic architecture or amplicon interference, may influence primer efficiency and detection sensitivity. To overcome these limitations and improve the resolution of protist diversity, future studies should increase sequencing depth and design universal primers to reduce off-target amplification. The generation of more reads per sample would enable the detection of rare and low-abundance taxa, providing a more comprehensive understanding of protist communities. Complementary methods, such as microscopy or targeted PCR assays, could validate the presence of specific protist taxa and increase the robustness of diversity assessments.

The metabarcoding approach demonstrated high sensitivity and reliability in detecting pathogenic parasites, which is vital for managing aquaculture resources such as Pacific oysters (C. gigas) and other ecologically important oyster species like S. cucullata. However, this method remains underutilized in Korea, despite the country being the second-largest producer of Pacific oysters globally. This study identified novel pathogenic protists and bacterial ASVs in gill tissues and eDNA samples from oysters in the northeastern Pacific. These findings underscore the importance of monitoring pathogens to understand the environmental and geographical factors driving their emergence and spread. Climate change, particularly rising SSTs, may influence pathogen prevalence and virulence, ultimately affecting the health and resilience of different oyster host species.

In this study, P. marinus, a pathogen that causes perkinsosis and significant oyster mortality [10, 11], was detected in S. cucullate from Taiwan and eDNA samples from Incheon, South Korea. While P. marinus was previously isolated from C. gigas in South Korea [43], its identification in a different oyster species (S. cucullate) and through eDNA highlights its broader regional distribution and potential for environmental persistence. The low number of detected copies suggests a low infection intensity, and no clinical signs of perkinsosis were observed. B. ostreae, the causative agent for bonamiosis [12], was identified exclusively in S. cucullate from Taiwan, marking both the first report of B. ostreae in this oyster species and its first detection in the northwestern Pacific region. Although B. ostreae typically induces hemocyte lysis, hemocytic nodules, and tissue degeneration in heavily infected oysters, no such clinical manifestations were observed, suggesting a low-level or early-stage infection. As with P. marinus, B. ostreae infections at low intensities may remain subclinical, particularly in tolerant or carrier hosts. Its absence in South Korean oysters, primarily C. gigas, may reflect their relative resistance to infection [12], though C. gigas could still serve as a carrier, potentially posing risks to more susceptible oyster species. Haplosporidium costale, a pathogenic protist causing SSO disease [44,45,46], was identified in C. gigas from Incheon. Although no mortality was observed, the possibility of C. gigas acting as a carrier warrants further surveillance. In addition, Polyplicarium citrusae, a marine apicomplexan that parasitizes polychaetes [47], was detected in oysters from both South Korea and Taiwan. While its direct impact on oysters remains unknown, its consistent presence in asymptomatic individuals may indicate a commensal or environmental association, and warrants further study to determine its ecological role in oyster microbiomes.

These findings offer important insights into oyster disease ecology. The detection of low-abundance but high-risk pathogens such as P. marinus, B. ostreae, and H. costale, particularly in asymptomatic oysters, demonstrates the utility of metabarcoding for early detection and proactive surveillance. The presence of these pathogens in both gill tissue and eDNA samples across geographically distant sites suggests the possibility of long-range dispersal. Although this study did not directly test ocean current–mediated transport, the identification of overlapping parasite species—such as P. marinus and R. seeberi—in South Korea and Taiwan may reflect oceanographic connectivity among these regions. The Kuroshio Current, which flows northward through all three countries, has been shown to facilitate the movement of parasites and microorganisms across marine ecosystems, including the northward drift of L. cyprinacea and other pathogens [8, 9]. This suggests that natural oceanic circulation, along with anthropogenic activities, may contribute to the spread of emerging oyster diseases. These observations underscore the importance of ongoing monitoring of both susceptible and carrier species, including C. gigas and S. cucullata, which may act as reservoirs for pathogen transmission within and between aquaculture zones. Additionally, the strong association between environmental gradients (e.g., SST) and microbiome composition suggests that climate-driven changes may influence disease risk by altering microbial stability or host susceptibility. Together, these findings support the use of integrated metabarcoding approaches for assessing pathogen dynamics and informing disease management strategies in oyster aquaculture.

Among bacteria, Vibrio bathopelagicus showed the highest relative read abundance in gill tissues and eDNA from the southern Korean coast. This species, which is pathogenic to bivalves in Europe [48], was relatively more abundant at high-production oyster sites, indicating a possible link to localized environmental or anthropogenic factors. Although no visible disease symptoms were observed, its elevated relative abundance warrants further investigation into its potential impact on oyster health. Additionally, Plasmodium ovale wallikeri, a malaria parasite primarily affecting humans, and Vermamoeba vermiformis [49], a free-living amoeba known to harbor pathogenic bacteria, were detected through eDNA screening. Although their presence in oyster-associated environments is unlikely to reflect direct infection, it may indicate anthropogenic contamination or serve as indicators of environmental conditions relevant to both oyster and human health. Taken together, these findings contribute directly to our understanding of oyster disease ecology by documenting both widespread and emergent pathogens in C. gigas and S. cucullata populations. This underscores the value of using both gill tissue and eDNA metabarcoding as effective surveillance tools for aquaculture health monitoring and ecosystem risk assessment.

Conclusion

The diversity of marine microbiomes is influenced by host–microbe interactions, environmental conditions, and ecological processes. Understanding these drivers is critical for developing effective conservation strategies and ensuring the health and sustainability of marine ecosystems. By examining microbial communities in two ecologically important oyster species—C. gigas and S. cucullata—this study highlights the importance of species-specific monitoring in different environmental contexts. Future research should integrate multiomics approaches, such as transcriptomics and metabolomics, to elucidate the functional roles and interactions of microbes within the host. Long-term monitoring across diverse environments is essential for predicting and mitigating the impacts of global change on oyster populations and their microbiomes.

Data availability

The datasets generated and analyzed during this study are available in publicly accessible repositories. The names of the repositories and the corresponding accession numbers are provided in **Supplementary Tables S8** and **S9**.

Abbreviations

eDNA:

Environmental DNA

NGS:

Next-generation sequencing

EDRR:

Early detection and rapid response

SST:

Sea surface temperature

ASV:

Amplicon sequence variants

NMDS:

Nonmetric multidimensional scaling

References

FIPS. Available from: https://www.fips.go.kr/p/Main/

Yu S, Hou X, Huan C, Mu Y. Comments on the oyster aquaculture industry in China: 1985–2020. Thalassas. 2023;39:875–82. https://doi.org/10.1007/s41208-023-00558-1.

Parker M, Bricker S. Sustainable oyster aquaculture, water quality improvement, and ecosystem service value potential in Maryland Chesapeake Bay. J Shellfish Res. 2020;39:269. https://doi.org/10.2983/035.039.0208.

Nakayama K, Kawahara Y, Kurimoto Y, Tada K, Lin HC, Hung MC, et al. Effects of oyster aquaculture on carbon capture and removal in a tropical mangrove lagoon in southwestern Taiwan. Sci. Total Environ. 2022;838:156460. https://doi.org/10.1016/j.scitotenv.2022.156460.

Ray NE, Maguire TJ, Al-Haj AN, Henning MC, Fulweiler RW. Low greenhouse gas emissions from oyster aquaculture. Environ. Sci. Technol. 2019;53:9118–27. https://doi.org/10.1021/acs.est.9b02965.

Nowland SJ, O’Connor WA, Osborne MWJ, Southgate PC. Current status and potential of tropical rock oyster aquaculture. Rev Fish Sci Aquac. 2020;28:57–70. https://doi.org/10.1080/23308249.2019.1670134.

Hung JJ, Tsai SH, Lin YH, Hsiang ZY. Characterizing dissolved inorganic and organic nutrients in the oligotrophic Kuroshio Current off eastern Taiwan during warm seasons. Front Mar Sci. 2024;11:1383244. https://doi.org/10.3389/fmars.2024.1383244.

Kato S, Yamakawa U. First record of an anchor worm Lernaea cyprinacea on a tropical abortively migrating fish, Lutjanus argentimaculatus, in eastern Japan. Biogeography. 2020;22. https://doi.org/10.11358/biogeo.22.16.

Wu Y, Hirai J, Zhou F, Iwataki M, Jiang S, Ogawa H, Inoue J, Hyodo S, Saito, H. Diversity and biogeography of dinoflagellates in the Kuroshio region revealed by 18S rRNA metabarcoding. Front Mar Sci. 2024;11:1361452. https://doi.org/10.3389/fmars.2024.1361452.

Andrews JD. Epizootiology of the disease caused by the oyster pathogen Perkinsus Marinus and its effects on the oyster industry. American Fisheries Society. 1988;18:47–63.

Villalba A, Reece KS, Camino Ordás M, Casas SM, Figueras A. Perkinsosis in molluscs: A review. Aquat Living Resour. 2004;17(4):411–32. https://doi.org/10.1051/alr:2004050.

Engelsma MY, Culloty SC, Lynch SA, Arzul I, Carnegie RB. Bonamia parasites: a rapidly changing perspective on a genus of important mollusc pathogens. Dis. Aquat. Organ. 2014;110:5–23. https://doi.org/10.3354/dao02741.

Thomas Y, Cassou C, Gernez P, Pouvreau S. Oysters as sentinels of climate variability and climate change in coastal ecosystems. Environ Res Lett. 2018;13:104009. https://doi.org/10.1088/1748-9326/aae254.

Okon EM, Birikorang HN, Minir MB, Kar ZA, Téllez-Isaías G, Khalifa NE, et al. A global analysis of climate change and the impacts on oyster diseases. Sustainability. 2023;15(17):12775. https://doi.org/10.3390/su151712775.

Hoffmann AA, Hercus MJ. Environmental stress as an evolutionary force. Bioscience. 2000;50:217. https://doi.org/10.1641/0006-3568(2000)050[0217:ESAAEF]2.3.CO;2.

Turner RD, Chiu C, Churchyard Gj, Esmail H, Lewinsohn DM, Gandhi NR, et al. Tuberculosis infectiousness and host susceptibility. J Infect Dis. 2017;216:S636–43. https://doi.org/10.1093/infdis/jix361.

Pimentel ZT, Dufault-Thompson K, Russo KT, Scro AK, Smolowitz RM, Gomez-Chiarri M, et al. Microbiome analysis reveals diversity and function of Mollicutes associated with the Eastern oyster, Crassostrea virginica. mSphere. 2021;6(3):e00227–21. https://doi.org/10.1128/msphere.00227-21.

Babayani ND, Vineer HR, Walker JG, Davidson RK. Editorial: Climate and parasite transmission at the livestock-wildlife interface. Front Vet Sci. 2021;8:816303. https://doi.org/10.3389/fvets.2021.816303.

Hines IS, Madanick JM, Smith SA, Kuhn DD, Stevens AM. Analysis of the core bacterial community associated with consumer-ready Eastern oysters (Crassostrea virginica). PloS One. 2023;18(2):e0281747. https://doi.org/10.1371/journal.pone.0281747.

Unzueta-Martínez A, Welch H, Bowen JL. Determining the composition of resident and transient members of the oyster microbiome. Front Microbiol. 2021;12:828692. https://doi.org/10.3389/fmicb.2021.828692.

Chabé M, Lokmer A, Ségurel L. Gut protozoa: friends or foes of the human gut microbiota? Trends Parasitol. 2017;33(12):925–34. https://doi.org/10.1016/j.pt.2017.08.005.

DuPont S, Lokmer A, Corre E, Auguet JC, Petton B, Toulza E, et al. Oyster hemolymph is a complex and dynamic ecosystem hosting bacteria, protists and viruses. Anim Microbiome. 2020;2:12. https://doi.org/10.1186/s42523-020-00032-w.

Guo S, Xiong W, Hang X, Gao Z, Jiao Z, Liu H, et al. Protists as main indicators and determinants of plant performance. Microbiome. 2021;9:64. https://doi.org/10.1186/s40168-021-01025-w.

Borrell YJ, Miralles L, Huu HD, Mohammed-Geba K, Garcia-Vazquez E. DNA in a bottle—Rapid metabarcoding survey for early alerts of invasive species in ports. PloS One. 2017;12(9):e0183347. https://doi.org/10.1371/journal.pone.0183347.

DeMone C, McClure JT, Greenwood S, Fung R, Hwang MH, Feng Z, Shapiro K. A metabarcoding approach for detecting protozoan pathogens in wild oysters from Prince Edward Island, Canada. Int. J Food Microbiol. 2021;360:109315. https://doi.org/10.1016/j.ijfoodmicro.2021.109315.

Reaser JK, Burgiel SW, Kirkey J, Brantley KA, Veatch SD, Burgos-Rodríguez J. The early detection of and rapid response (EDRR) to invasive species: a conceptual framework and federal capacities assessment. Biol Invasions. 2020;22:1–19. https://doi.org/10.1007/s10530-019-02156-w.

MacAulay S, Ellison AR, Kille P, Cable J. Moving towards improved surveillance and earlier diagnosis of aquatic pathogens: From traditional methods to emerging technologies. Rev Aquacult. 2022;14(4):1813–29.

Bass D, Stentiford GD, Wang HC, Koskella B, Tyler CR. The pathobiome in animal and plant diseases. Trends Ecol Evol. 2019;34:996–1008. https://doi.org/10.1016/j.tree.2019.07.012.

Hsiao ST, Chuang SC, Chen KS, Ho PH, Wu CL, Chen CA. DNA barcoding reveals that the common cupped oyster in Taiwan is the Portuguese oyster Crassostrea angulata (Ostreoida; Ostreidae), not C. gigas. Sci Rep. 2016;6:34057. https://doi.org/10.1038/srep34057.

Gollasch S, Minchin D, David M. The transfer of harmful aquatic organisms and pathogens with ballast water and their impacts. In Global Maritime Transport and Ballast Water Management, Springer Netherlands, Dordrecht. 2015;35–58. https://doi.org/10.1007/978-94-017-9367-4_3.

Hsu PC, Lu CY, Hsu TW, Ho CR. Diurnal to seasonal variations in ocean chlorophyll and ocean currents in the north of Taiwan observed by geostationary ocean color imager and coastal radar. Remote Sens. 2020;12:2853. https://doi.org/10.3390/rs12172853.

Hu J, Lan W, Huang B, Chiang KP, Hong H. Low nutrient and high chlorophyll a coastal upwelling system – A case study in the southern Taiwan Strait. Estuar Coast Shelf Sci. 2015;166:170–7. https://doi.org/10.1016/j.ecss.2015.05.020.

Bushnell B, Rood J, Singer E. BBMerge – Accurate paired shotgun read merging via overlap. PloS One. 2017;12:e0185056. https://doi.org/10.1371/journal.pone.0185056.

Edgar RC, Haas BJ, Clemente JC, Quince C, Knight R. UCHIME improves sensitivity and speed of chimera detection. Bioinformatics. 2011;27:2194–200. https://doi.org/10.1093/bioinformatics/btr381.

Kandlikar GS, Gold ZJ, Cowen MC, Meyer RS, Freise AC, Kraft NJB, et al. ranacapa: An R package and Shiny web app to explore environmental DNA data with exploratory statistics and interactive visualizations. F1000Res. 2018;7:1734. https://doi.org/10.12688/f1000research.16680.1.

McMurdie PJ, Holmes S. Phyloseq: An R package for reproducible interactive analysis and graphics of microbiome census data. PloS One. 2013;8:e61217. https://doi.org/10.1371/journal.pone.0061217.

Shen W, Ren H. Taxonkit: a practical and efficient NCBI taxonomy toolkit. J Genet Genomics. 2021;48:844–50. https://doi.org/10.1016/j.jgg.2021.03.006.

Forster D, Dunthorn M, Mahé F, Dolan JR, Audic S, Bass D, et al. Benthic protists: the under-charted majority. FEMS Microbiol Ecol. 2016;92(8):fiw120. https://doi.org/10.1093/femsec/fiw120.

Lokmer A, Goedknegt MA, Thieltges DW, Fiorentino D, Kuenzel S, Baines JF, Wegner KM. Spatial and temporal dynamics of Pacific oyster hemolymph microbiota across multiple scales. Front Microb. 2016;7:1367. https://doi.org/10.3389/fmicb.2016.01367.

Gong J, Shi F, Ma B, Dong J, Pachiadaki M, Zhang X, et al. Depth shapes α- and β‐diversities of microbial eukaryotes in surficial sediments of coastal ecosystems. Environ Microbiol. 2015;17(10):3722–37. https://doi.org/10.1111/1462-2920.12763.

Stoeck T, Bass D, Nebel M, Christen R, Jones MDM, Breiner HW, et al. Multiple marker parallel tag environmental DNA sequencing reveals a highly complex eukaryotic community in marine anoxic water. Mol Ecol. 2010;19:21–31. https://doi.org/10.1111/j.1365-294X.2009.04480.x.

Vaulot D, Geisen S, Mahé F, Bass D. pr2-primers: An 18S rRNA primer database for protists. Mol Ecol Resour. 2022;22(1):168–79. https://doi.org/10.1111/1755-0998.13465.

Kim SH, Bathige SDNK, Jeon HB, Lee D, Choi KS, Kim HJ, Park KI. First report of Perkinsus marinus occurrence associated with wild Pacific oysters Crassostrea gigas from the west coast of Korea. J Invert Pathol. 2024;204.

Wood JL, Andrews JD. Haplosporidium costale (Sporozoa) associated with a disease of Virginia oysters. Science. 1962;136:710–11. https://doi.org/10.1126/science.136.3517.710.b.

Cherif-Feildel M, Lagy C, Quesnelle Y, Bouras H, Trancart S, Houssin M. Detection of the protistan parasite, Haplosporidium costale in Crassostrea gigas oysters from the French coast: A retrospective study. J Invertebr Pathol. 2022;195:107831. https://doi.org/10.1016/j.jip.2022.107831.

Wang Z, Lu X, Liang Y, Wang C. Haplosporidium nelsoni and H. costale in the Pacific oyster Crassostrea gigas from China’s coasts. Dis Aquat Organ. 2010;89:223–28. https://doi.org/10.3354/dao02196.

Zakariah MI, Daud HM, Abdullah MI, Husna NA, Hassan M. Marine gregaraines (Apicomplexa): Their biology, identification and control. Int J Pure Appl Zool. 2021;9(5):9–15.

Lasa A, Auguste M, Lema A, Oliveri C, Borello A, Taviani E, et al. A deep-sea bacterium related to coastal marine pathogens. Environ Microbiol. 201;23(9):5349–63. https://doi.org/10.1111/1462-2920.15629.

Delafont V, Rodier MH, Maisonneuve E, Cateau E. Vermamoeba vermiformis: a free-living amoeba of interest. Microb Ecol. 2018;76:991–1001. https://doi.org/10.1007/s00248-018-1199-8.