The “white shrimp” Penaeus vannamei is a native species to the west Pacific coast, from Mexico to Peru [1]. Currently, this species represents more than 50% of the cultivated crustaceans worldwide and a global market value of USD 18 billion approximately [2]. The main producing countries are located in the South and Southeast Asia; however, in Latin America, a significant aquaculture industry remains, with significant productions in Ecuador, Mexico, Honduras, and Brazil [1]. Numerous studies have examined the roles of microbiota in nutrition, cell proliferation, and host immunomodulation [3,4,5,6]. The understanding of the microbiota of P. vannamei remains insufficient compared to vertebrates, including terrestrial livestock and finfish species [7]. The modulation of microbiota as an alternative approach to maximize production and disease control using prebiotics, probiotics, and synbiotics requires a comprehensive knowledge of the diversity of the microbiota in the host organism. Recently, advances in next-generation sequencing (NGS) have allowed further research of crustacean microbiota [8,9,10,11,12]. These studies focused on modulating the microbiota of P. vannamei through dietary interventions at laboratory scale. Therefore, our study was focused to characterize the bacterial microbiota composition in the intestinal tract of P. vannamei reared in intensive commercial ponds in Ecuador. This study was performed in the province of El Oro, which corresponds to the area where shrimp commercial cultivation began in 1968 [1]. Understanding the bacterial ecology in the crustacean gut can help to improve both the management of hatcheries for higher productivity and the safety of shrimps as food. Details of the sample site, experimental methodology, and sequence analysis are provided in Online Resource 1. The sequencing of 16S rRNA amplicons using Ion Torrent technology generated a total of 1,005,400 sequences, with an average quality per sequence (Phred score) of 29, equivalent to 99.9% precision in sequence fit and a probability of error of 0.00126. Sequence data have been deposited in the Sequence Read Archive of the National Centre for Biotechnology Information (SRA, NCBI) under the BioProject (PRJNA352369, SAMN05971820, SAMN05971821, SAMN05972042, SAMN05972043, SAMN05972044, SAMN05971813, SAMN05971815, SAMN05971819). After de-replication, quality, chimera, and unique sequence filtering, there were a total of 422,838 high-quality sequences, which were assigned to a total of 1949 operational taxonomic units (OTUs), with 97% grouped by sequence identity. Later, a screening of OTUs assigned to Archaea (114 OTUs and 23,057 sequences) yielded a total of 399,781 sequences assigned to 1835 OTUs. Given that the primers used were designed for the V2–V3 region of bacterial 16S RNA, the Archaea sequences were filtered after this step to avoid errors in statistical and qualitative inferences. However, during the nursery phase, the Archaea domain was represented by the phyla Thaumarchaeota (12%) and Euryarchaeota (0.05%), while in the harvest stage, the microbiota contained Thaumarchaeota (3%) and Euryarchaeota (0.2%, Online Resource 1 Fig. 2).

Sequence analysis revealed that the most abundant phyla in the microbiota of shrimp during the nursery phase were CKC4 (37.3%), Proteobacteria (29.8%), Actinobacteria (11.6%), and Firmicutes (10.1%), while the intestinal microbiota of harvested shrimp comprised mostly Proteobacteria (28.4%), Chloroflexi (19.9%), and Actinobacteria (15.1%, see Fig. 1a). In total, 32 phyla were detected, with 23 phyla in the nursery phase and 32 phyla in the harvest phase. The taxonomic analysis identified a total of 520 genera (306 in nursery; 454 in harvest). Among them, the OTUs with greater relative abundances in the microbiota of nursery phase shrimp were CKC4 uncultured bacterium (37%) and Escherichia-Shigella (18%). Among the microbiota in harvested shrimp, the most abundant genera were uncultured Caldilinea (19%) and Alphaproteobacteria uncultured bacterium (10%, see Fig. 1b). The Chao1 and Robbins indexes indicated that the alpha diversity was higher in the harvested shrimp compared to the microbiota of shrimp in the nursery stage (Figs. 1 and 2a and Online Resource 1). Non-directional beta diversity is a measure of the variation in community structure in response to some environmental/experimental factors [13]. The beta diversity of the bacterial communities associated with P. vannamei, comparing nursery and harvest stages, was investigated through a PCoA showed in Fig. 2b. The first two components explain a total of 58.39% of the variation (first component, 40.88%; second component, 17.51%). This figure also showed that the composition of the nursery’s microbiota is uniform, as the dots are grouped in the graph. In contrast, harvested shrimp’s microbiota showed wider dispersion, indicating differences in their beta diversity. The ANOSIM test [14] results, with R = 0.98 ≈ 1 (p = 0.0219), indicated significant differences between the shrimp microbiota for these two farming phases. Microbiota was compared with linear discriminant analysis effect size (LEfSe) [15] to identify characteristic taxa associated with each phase. The graphical results of the LEfSe (Fig. 3) indicate that the taxa associated with the nursery phase were, in descending order, the genera Staphylococcus, the Staphyloccaceae family, the Pseudomonadales and uncultured Cyanobacteria orders, the Pseudomonadaceae family, and the Frankiales order. In contrast, in the microbiota of the harvested shrimp, the taxa represented were the phylum Verrucomicrobia, class Spartobacteria, Chthoniobacteria family LD29, the Chthoniobacteria order, phylum Chloroflexi, the Caldilinea order, and the Caldilineae family. The present work offers the most detailed description of the microbiota of P. vannamei under commercial production conditions to date. Despite the economic importance of the species, the examination of the intestinal microbiota of the white shrimp has been limited to probiotic effects and laboratory scale assays. Few studies have explored the microbiota composition during the growth of this crustacean [16,17,18,19]. Furthermore, several reports about lab scale P. vannamei microbiota have observed that Proteobacteria is the most abundant phylum, with relative abundance between 68 and 97% [8,9,10,11,12, 17, 18, 20, 21]. As described in Online Resource 1 Table 1, our results showed differences with the composition described in those reports. The differences in microbiota during growth were reflected in our study, where we observed an increase in the diversity of the microbiota of the harvested shrimp compared to that of the nursery shrimp. Huang et al. [17] used NGS to describe the microbiota during lab scale growth (14, 30, 60, and 90 days; Online Resource 1 Table 1) and found that the most prevalent phylum was Proteobacteria, (averaging 43.3%), corresponding to > 70% of the relative abundance in the early phases. This contrasts with our results that showed that the nursery phase microbiota was dominated by the CKC4 phylum (37%). This finding is of interest because little information is available regarding this bacterial group, and this leads to new questions regarding the basis for this association and the consequences (beneficial or detrimental) for the host. It has been suggested that changes in intestinal microbiota may be associated with the severity of shrimp diseases [20, 21]. However, in our study, under intensive commercial farming conditions, the harvest shrimp microbiota showed no more than 28% Proteobacteria. There is a lack of detailed information about the relative abundance at the genus level in shrimp microbiota. Sha et al. [11] studied P. vannamei microbiota in China at the lab scale and reported that the relative abundance of the top 10 genera ranged 0.36–0.81%, whereas the top three genera were Octadecabacter (2.7%), Acinetobacter (2.0%), and Demequina (1.1%). In contrast, we found that harvest shrimp showed high levels of uncultured Caldilinea (19%) and Alphaproteobacteria uncultured bacterium (10%). It is well known that the colonization of microorganisms in host intestine is influenced by factors relative to host and non-host interactions [5, 22]. Such factors include the already present intestinal assemblage [23, 24], host physiological condition [25], growth stage of the host [17, 26, 27], and environmental conditions [20]. Previous investigation developed in laboratory conditions over other crustacean’s species like Homarus gammarus [28], Neocaridina denticulata [29], and Penaeus monodon [30] shows that microbiota highly dissimilar to water/fed microbiota have been developed by time, suggesting that host selection structures the gut microbiota influenced by environmental, physiological, and nutritional factors. A stable microbial community is important to the host’s health but it can be constantly influenced by various environmental factors [31]. For example, salinity is one of these factors, since recent reports have revealed the influence of hyposaline/hypersaline environments on shrimp microbiota composition [12]. Furthermore, other authors have reported the important influence of water and biofloc on shrimp intestinal microbiota [31]. It is worthy to notice that this study was performed at commercial scale, and we tried to homogenize the environmental factors (water, diet, temperature, oxygen, salinity). In agreement with our results, Rungrassame et al. [30] suggest that there is an increasing diversity in the microbiota of P. monodon between 30 and 90 days of culture. Furthermore, Cheung et al. [29] associated this increase in diversity with the sexual maturation of the shrimp Neocaridina denticulata. The published results on the composition of the microbiota in P. vannamei argue that major bacterial groups shifted at different growth stages. The molting process of the host, the differences in diet, and development of the host digestive and reproductive systems might be responsible for the variance of bacterial composition [17, 18]. Therefore, the differences observed between intestinal microbiota of nursery and harvest shrimp could be attributed to the growth stage of the host [17, 26, 27] rather than environmental conditions [20]. The marked difference observed between the microbiota reported in Asia and our results could be explained by the non-host factor interaction, but host factors associated with the culture of shrimp genetic lines may influence this marked differentiation. P. vannamei was introduced in Asia experimentally from 1978 to 1979 but commercially only since 1996 in mainland China and Taiwan, followed by other Asian countries in 2000 [32]. Most Asian countries have legislated against the importation of P. vannamei due to risk of introducing new pathogenic bacterial diseases, keeping isolated shrimp genetic lines. Potential pathogenic Vibrio species are halophilic bacteria that are ubiquitous in marine and coastal environments. Some of them are pathogenic to human and marine animals [33] and have been identified as serious disease problem in shrimp culture ponds all over the world [33,34,35]. In most cases, however, Vibrio genus is considered opportunistic pathogens for shrimps; however, vibrio-related emerging infectious diseases have recently expanded in geographic range, increasing the risk of economic losses in the future [36]. In the present study, the genus Vibrio was detected in the nursery and harvest stages, 2.02 and 0.64% of relative abundance, respectively, whereas the salinity of the ponds was 50/00, very close to the lower tolerance limits for V. parahaemolyticus and V. vulnificus [37, 38]. This result suggests that risk of contamination with potentially pathogenic Vibrio still remains in inland intensive culture facility. Previous studies have reported the presence of coliforms in P. vannamei microbiota. Those studies include data from different geographic locations such as Texas, USA [39], China [40], Brazil [41, 42], Thailand [33], and India [43]. Those reports agree with our observation about the presence of the Escherichia/Shigella genus in shrimp microbiota. Moreover, Rungrassamee et al. [44] reported variations in relative abundance of the Escherichia/Shigella genus during the growth the P. monodon. On the other hand, several publications have been focused in dietary modulation of the microbiota of shrimp and reported some minor changes when specific cultivable strains have been evaluated, mostly Vibrio [45,46,47,48,49,50,51,52]. However, probiotic-induced microbiota changes may enhance the immune responses of shrimp [47], and enhancing immune responses could require an unnecessary expenditure of energy [19]. Non-indigenous probiotics may lead to environmental imbalances, particularly because effluent treatments may have reduced effect on probiotic removal prior to discharge [19]. Our results suggest that the microbiota of P. vannamei under conditions of intensive cultivation in Ecuador diversifies during the farming process. The increased number of phyla and genera at harvest may be important to the host in terms of increased metabolic capabilities and production of valuable substances, such as amino acids, vitamins, enzymes, and specific growth factors [27].

Fig. 1
figure 1

Comparison of relative abundance. a Shrimp microbiota composition (relative to OTUs composition) at phylum level, including seven phyla showing the highest abundance. b Shrimp microbiota composition (relative to OTUs composition) at genus level, including 15 genera showing the highest abundance between nursery (■) and () harvested shrimp microbiota

Full size image
Fig. 2
figure 2

Diversity analysis. a Alpha diversity index Chao1. OTU richness indices are shown for the nursery and harvest intestinal microbiota of shrimps (t stat p value = 0.018 < 0.05). The line in each box plot indicates the median, the box delimits the 25th and 75th percentile, and the whisker is the range. b Principal component analysis (PCoA) normalized distribution plot (ANOSIM test statistic ≈ 0.989, p value = 0.0219 < 0.05, number of permutations 999). corresponds to microbiota of nursery shrimps. corresponds to microbiota of harvest shrimps

Full size image
Fig. 3
figure 3

Graphical summary of LEfSe (Linear discriminant analysis effect size) results. The histogram shows the LDA scores computed for differentially abundant taxa between nursery and harvest shrimps microbiota. The histogram identifies which taxa among all those detected as statistically and biologically differential explain the differences between microbial communities

Full size image