Abstract
Environmental gradients can influence morpho-physiological and life-history differences in natural populations. It is unclear, however, to what extent such gradients can also modulate phenotypic differences in other organismal characteristics such as the structure and function of host-associated microbial communities. In this work, we addressed this question by assessing intra-specific variation in the diversity, structure and function of environmental-associated (sediment and water) and animal-associated (skin and gut) microbiota along an environmental gradient of pollution in one of the most urbanized coastal areas in the world. Using the tropical sea cucumber Holothuria leucospilota , we tested the interplay between deterministic (e.g., environmental/host filtering) and stochastic (e.g., random microbial dispersal) processes underpinning host-microbiome interactions and microbial assemblages. Overall, our results indicate that microbial communities are complex and vary in structure and function between the environment and the animal hosts. However, these differences are modulated by the level of pollution across the gradient with marked clines in alpha and beta diversity. Yet, such clines and overall differences showed opposite directions when comparing environmental- and animal-associated microbial communities. In the sea cucumbers, intrinsic characteristics (e.g., body compartments, biochemistry composition, immune systems), may underpin the observed intra-individual differences in the associated microbiomes, and their divergence from the environmental source. Such regulation favours specific microbial functional pathways that may play an important role in the survival and physiology of the animal host, particularly in high polluted areas. These findings suggest that the interplay between both, environmental and host filtering underpins microbial community assembly in H. leucospilota along the pollution gradient in Hong Kong.
Keywords: Microbial community, holothurian, nutrient pollution, environmental gradient, host-microbiome
Introduction
Metazoans harbour diverse and dynamic microbial communities (microbiome) that play essential roles in the ecology and function of their animal host (Apprill, 2017). Symbiotic microorganisms influence host metabolic processes (Egerton et al., 2018; Hakim et al., 2016; Hanning & Diaz-Sanchez, 2015), development (Carrier & Reitzel, 2018), immune responses (de Araujo et al., 2019; Krediet et al., 2013), reproduction (Brucker & Bordenstein, 2013; Sharon et al., 2010), behaviour (Sherwin et al., 2019) and survival (Kešnerová et al., 2020). However, microbial communities and the interactions with their hosts are not static and can vary across time (e.g., host’s development or seasonality, Kešnerová et al., 2020; Shoemaker & Moisander, 2017) and space (e.g., host’s body plan or geography, Griffiths et al., 2019; McKnight et al., 2020). Thus, to understand the ecological dynamics of animal hosts in a changing world, it is becoming more important to understand the mechanisms and drivers underpinning the origin and regulation (functional and structural) of their associated microbial communities (Kohl, 2020).
Host-associated microbiomes are shaped by a diversity of evolutionary and ecological processes that can be explored through the framework used in community ecology. Under this framework, the formation of microbial communities is seen as the outcome of selective processes in which a larger species pool is subjected to a set of biotic and abiotic filters (Hardy et al., 2012). For instance, habitat/environmental filtering is usually considered one of the dominant forces in structuring communities consisting of habitat-specialized species (Cavender-Bares et al., 2009). In this case, the habitat/environment is part of the selective processes as it plays a dual role acting as a source for the hosts’ microbiome and also influencing the composition and dynamics of the established microbial community (Johnke et al., 2020). As a result, host intra- and interspecific variation in microbiome composition is expected to be lower for organisms inhabiting similar environments in comparison to their counterparts living in more divergent conditions (Johnke et al., 2020). Hosts, however, can also influence associated microbiomes in several ways, such as through selective feeding or filtering (Gao et al., 2014; Theis et al., 2016). This may result from evolutionary processes that produce patterns in which closely related host species harbour similar microbiota even if they inhabit different environments (San Juan et al., 2020; Sanders et al., 2015). In any case, environmental and host filtering are aligned with the fundamental idea that deterministic processes dictate the microbial community assembly in natural populations (Näpflin & Schmid-Hempel, 2018; J. Wang et al., 2013; Weigel, 2020; Yan et al., 2016).
Previous studies have also suggested that stochastic processes can influence microbial community assembly (Oliphant et al., 2019; Zhou et al., 2013). These processes include random dispersal potential and colonization chance of microbes (Zhou & Ning, 2017), random arrival sequence to host from the environment during dispersal (Burns et al., 2017), changes in community abundance and survival due to random speciation, extinction and ecological drift (Chase & Myers, 2011). Under this paradigm, species present in the microbial community would be independent and unpredictable, with no specific occurrence pattern or relation with the respective niche (Rosindell et al., 2012). This provides us with a wider perspective when studying processes underpinning microbial community assembly. What is worth mentioning is that deterministic and stochastic processes are not mutually exclusive (Bordenstein & Theis, 2015). Both processes work together in structuring microbial communities although a stronger contribution of a particular process during the assembly may occur depending on the investigated model and ecosystem (Yan et al., 2016).
In marine environments, the interplay between deterministic and stochastic factors is an important driver of inter- and intra-specific differences in organismal characteristics across heterogeneous seascapes. In these systems, gradients of environmental conditions are known to influence geographic variation in attributes such as physiology (Gaitán-Espitia et al., 2014; Gaitán-Espitia, Villanueva, et al., 2017), life-history (Pecuchet et al., 2018), zonation (Mangialajo et al., 2012), behaviour (Cornwell et al., 2019), intra- and inter-specific genetic diversity (Zhong et al., 2024), stress tolerance and phenotypic plasticity (Gaitán-Espitia, Bacigalupe, et al., 2017) in natural populations. As such, it would be expected that environmental gradients would also influence variation and phenotypic differences in other organismal characteristics, including the structure and function of host-associated microbial communities. In fact, gradients in sea surface temperature and salinity are known to influence symbiotic microbial assemblages in benthic marine species (Capistrant-Fossa et al., 2021; Ketchum et al., 2021; Osman et al., 2020) and coastal ecosystems (Bolhuis et al., 2013; Z. Wang et al., 2019; Zäncker et al., 2018; G. Zhang et al., 2021). While these natural environmental gradients may have a fundamental role in modulating ecological patterns of marine microbiomes, anthropogenic-mediated gradients (e.g., nutrient pollution) can potentially induce drastic changes in these patterns by altering ecological dynamics, the origin, and regulation of host-associated microbial communities (Degregori et al., 2021; Milan et al., 2018; Stevick et al., 2021; Ziegler et al., 2019). 
In this study, we aimed to assess to what extent deterministic and/or stochastic processes along an anthropogenic-mediated gradient of pollution influence intra-specific variation in diversity, structure, and function of host-associated microbial communities in marine organisms. Here, our model system was Holothuria leucospilota, a tropical sea cucumber that dominates shallow waters in Hong Kong and the Indo-Pacific region. H. leucospilota is a deposit-feeding species that assimilates organic matter from surface sediments (including bacteria, benthic phytoplankton, meiofauna, and organic detritus), which serves as a constant environmental source for microbe acquisition (Gao et al., 2014; Pagán-Jiménez et al., 2019). However, H. leucospilota is characterised by the capacity to secrete secondary metabolites from the skin, gonads, and guts, which have antibacterial and antifouling properties (Darya et al., 2020), potentially allowing them to regulate the influence of environmental microbial reservoirs. Based on 16S amplicon sequencing, we examined both sea cucumber (skin and intestine) and environment (sediment and water) microbiomes along a pollution gradient in Hong Kong, one of the busiest ports and highly urbanized areas in the world. If environmental filtering is the main driver of microbial community assemblies along the pollution gradient, then similar host and environmental microbiomes will be observed within sites whilst higher microbiome variation would be expected across the gradient. In this context, microbial dispersal patterns and ecological drift might also play an important role in shaping inter-individual microbiome variation within sites (Stothart et al., 2021). Conversely, if host filtering and specificity are the main drivers of microbial community assembly in H. leucospilota , then low inter-individual variation in sea cucumber microbiome composition would be expected along the pollution gradient independently of the variation in the environmental microbial communities. However, intra-individual variation would be expected as sea cucumber skin and guts have different antibacterial properties (Darya et al., 2020).