Fine-scale succession patterns and assembly mechanisms of bacterial community of Litopenaeus vannamei larvae across the developmental cycle

Abstract

Background: Microbiome assembly in early life may have a long-term impact on host health. Larval nursery is a crucial period that determines the success in culture of Litopenaeus vannamei, the most productive shrimp species in world aquaculture industry. However, the succession patterns and assembly mechanisms of larval shrimp bacterial community still lack characterization at a fine temporal scale. Here, using a high-frequency sampling strategy and 16S rRNA gene amplicon sequencing, we investigated dynamics of larval shrimp bacterial community and its relationship with bacterioplankton in the rearing water across the whole developmental cycle in a realistic aquaculture practice.

Results: Alpha-diversity of larval shrimp bacteria showed a U-shaped pattern across the developmental cycle with the stages zoea and mysis as the valley. Correspondingly, the compositions of dominant bacterial taxa at the stages nauplius and early postlarvae were more complex than other stages. Remarkably, Rhodobacteraceae maintained the overwhelming dominance after the mouth opening of larvae (zoea I~early postlarvae). The taxonomic and phylogenetic compositions of larval bacterial community both showed stage-dependent patterns with higher rate of taxonomic turnover, suggesting that taxonomic turnover was mainly driven by temporal switching among closely related taxa (such as Rhodobacteraceae taxa). The assembly of larval bacteria was overall governed by neutral processes (dispersal among individuals and ecological drift) at all the stages, but bacterioplankton also had certain contribution during three sub-stages of zoea, when larval and water bacterial communities were most associated. Furthermore, the positive host selection for Rhodobacteraceae taxa from the rearing water during the zoea stage and its persistent dominance and large predicted contribution to metabolic potentials of organic matters at post-mouth opening stages suggest a crucial role of this family in larval microbiome and thus a potential source of probiotic candidates for shrimp larval nursery.

Conclusions: Our results reveal pronounced succession patterns and dynamic assembly processes of larval shrimp bacterial communities during the developmental cycle, highlighting the importance of the mouth opening stage from the perspective of microbial ecology. We also suggest the possibility and potential timing in microbial management of the rearing water for achieving the beneficial larval microbiota in the nursery practice. Video Abstract.

Keywords: Community assembly; Early life microbiome; Host development; Shrimp larvae; Succession pattern.

Fig. 1

Fine-scale temporal dynamics of α-diversity and evenness indices of larval shrimp bacterial communities. Data present means ± standard deviation

Fig. 2

Dynamics of dominant phyla/proteobacterial classes and families of larval shrimp bacterial communities. The phyla and proteobacterial classes with average relative abundant > 1% and the families with average relative abundant > 2% at least in one sampling time point are shown

Fig. 3

Heatmap showing the dynamics of dominant larval bacterial OTUs across developmental stages. The OTUs with average abundance > 2% at least in one sampling time point are shown. The data of OTU abundance was log₂ transformed. The phylogenetic tree was built using the maximum likelihood method in MEGA 7.0

Fig. 4

Taxonomic and phylogenetic turnover of larval and water bacterial communities with host development. Principal coordinate analysis (PCoA) visualizing compositional variations of larval and water bacterial communities across developmental stages based on Bray-Curtis dissimilarity (a) and weighted UniFrac distance (b). Time-decay in similarity between larval bacterial communities based on Bray-Curtis similarity (c) and phylogenetic similarity based on UniFrac distance (1 - weighted UniFrac distance) (d). The data of similarity values and hours between observations were shown as log₁₀ transformed

Fig. 5

Fit of the neutral models for larval bacterial communities with corresponding bacterioplankton as the source. a The Venn diagrams show the number and proportion of OTUs being unique in shrimp larvae/rearing water and shared by larvae and water. The OTUs that occurred more frequently than predicted by the model are shown in green, while those occurred less frequently than predicted are shown in orange. Blue dashed lines represent 95% confidence intervals around the model prediction and the OTUs fall within the confidence intervals were considered as neutrally distributed. R² values present the goodness of fit of the neutral model, ranging from 0 (no fit) to 1 (perfect fit). The histograms show the cumulative relative abundance of three categories of OTUs (above prediction, below prediction, and neutrally distributed) in larval bacterial communities at each stage. b The taxonomic distribution of three categories of OTUs at three sub-stages of zoea, when the occurrence of the larvae-water shared OTUs in larval bacterial communities fitted the model

Fig. 6

Fit of the neutral models for larval bacterial communities with larval matecommunity as the source. a The OTUs that occurred more frequently than predicted by the model are shown in green, while those occurred less frequently than predicted are shown in orange. Blue dashed lines represent 95% confidence intervals around the model prediction and the OTUs fall within the confidence intervals were considered as neutrally distributed. R² values present the goodness of fit of the neutral model, ranging from 0 (no fit) to 1 (perfect fit). The cumulative relative abundance and taxonomic distribution of three categories of OTUs (above prediction, below prediction, neutrally distributed) in larval bacterial communities at each stage are shown. b The dynamics of R² value of the neutral model, Akaike Information Criterion comparing the fit of the neutral model and the binomial distribution model, and the estimated migration rate (m) of larval bacterial communities across developmental stages