Comparative analysis of morphology and gut microbiome underlying trophic niche differentiation in three stomatopod species

Abstract

Given that predatory strategies and trophic niche in stomatopods vary with raptorial appendage differentiation, a comparative analysis was conducted on three species (Odontodactylus japonicus, Lysiosquilla sulcirostris, and Oratosquilla oratoria) exhibiting distinct habitats and morphologies. The study measured their raptorial appendage morphology, digestive tract morphology, and gut microbiota. The results revealed significant differences in the raptorial appendages and digestive tract morphology among the three stomatopod species. O. japonicus has the significantly largest merus length, merus width, merus height and, L. sulcirostris has the significantly largest propodus length, dactylus length, hepatopancreas length and midgut length, while O. oratoria has the significantly largest hindgut length, cardiac stomach length and pyloric stomach length. Regarding digestive tract morphology, O. japonicus possessed the widest villus and tallest villus, in contrast to O. oratoria, which displayed the smallest values for these features. A total of 8,273 operational taxonomic units (OTUs) were obtained from all 16S rRNA sequences. The OTU numbers for O. oratoria, L. sulcirostris, and O. japonicus were 344, 113, and 82, respectively, with only 17 OTUs shared among all three species. The gut microbiota of O. japonicus was dominated by the phylum Pseudomonadota (88.90%), primarily comprised of the genus Cupriavidus (68.40%). In contrast, the dominant phylum in both L. sulcirostris and O. oratoria was Bacillota, with Mycoplasma being the dominant genus in each, respectively. Furthermore, O. oratoria exhibited the highest gut microbial diversity, whereas L. sulcirostris had the lowest. Functional profiling revealed that the relative abundance of gut microbiota associated with amino acid, lipid, and carbohydrate metabolism was significantly higher in O. japonicus than in both O. oratoria and L. sulcirostris. In conclusion, our findings provide further evidence for feeding habit differentiation and associated regulatory mechanisms among the three stomatopod species. The high gut microbial diversity in O. oratoria may contribute to its adaptability to the dynamic, heterogeneous habitats it occupies. Although the microbial abundance in O. japonicus is intermediate between that of O. oratoria and L. sulcirostris, its microbiota appears to be more efficient in digestive function, particularly in the metabolism of key nutrients. This study systematically elucidates the regulatory mechanisms underlying trophic niche differentiation in stomatopods from integrated morphological and physiological perspectives. Our findings provide a valuable foundation for future research on the evolution of higher taxonomic groups within Stomatopoda based on nutritional ecology.

Introduction

Feeding habits are a central driver of animal evolution [1], particularly for apex carnivores. These species often exist at smaller population densities, rendering them vulnerable to premature extinction driven by dietary pressures [2, 3]. Consequently, some carnivores evolve broader dietary spectra as an adaptive response to the threat of extinction [1, 4, 5]. Beyond mere survival requirements, the distribution of food resources is a key selective pressure evolution of feeding habits [6], where animals tend to maximize energy efficiency by favoring an optimal foraging strategy—prioritizing food that is energy-rich, readily obtainable, and safe [7].

Following ingestion, efficient food utilization relies on both physical and chemical digestive processes [1, 8, 9]. Thus, species-specific dietary habits are often correlated with adaptations in digestive tract morphology and physiology [10, 11]. In fact, previous studies have confirmed that several adaptations are crucial for enhancing digestive efficiency and nutrient absorption in carnivores [12, 13]. These key factors include changes in amino acid requirements [14], a shortened digestive tract [15], variations in digestive enzyme activity [16], and a distinct composition and diversity of the gut microbiota [17]. It is noteworthy that within a shared habitat, species exhibit dietary differentiation to partition available food resources. Conversely, in distinct habitats, even phylogenetically closely related species develop specific diets [18] and corresponding digestive tract morphologies adapted to local conditions, leading to significant physiological divergence [19, 20].

Stomatopods, a group of malacostracan crustaceans belonging to the subclass Hoplocarida, are typically highly aggressive carnivores [21]. The second maxillipeds of stomatopods have evolved into specialized predatory organs- raptorial appendages that function like a “compound bow” [22], capable of striking prey with a rapid, lethal force [23, 24]. Moreover, the high-speed movement of these appendages generates cavitation, with the subsequent implosion of the bubbles delivering a secondary lethal blow [25, 26]. Equipped with highly evolved, remarkably efficient predatory appendages, stomatopods occupy the top of the food chain in benthic and coastal sandy-mud communities [27]. This makes them an ideal model for studying adaptive diversification in the feeding habits of extreme predatory crustaceans. It is noteworthy that there are over 450 described species of stomatopods [22]. Our previous research has already documented significant interspecific morphological differences in their raptorial appendages. The raptorial appendages of stomatopods are ecologically crucial and have evolved into two distinct forms: the “smasher” type and the “spearer” type [28–30]. Stomatopods with “smasher” appendages typically inhabit hard coral reefs, where they feed on shellfish and crustaceans. To break open the hard shells of their prey and compete for limited crevices in the coral, they have developed robust, hammer-shaped dactylopodites [30, 31]. In contrast, stomatopods with “spearer”-type appendages reside in soft sediment-bottom habitats. While these environments offer more abundant living space and reduced competition, they require the animals to excavate their own burrows. Furthermore, to effectively spear fish or soft-bodied mollusks, they have evolved slender, sharp, and barbed dactylopodites [31, 32].

The intestine is a vital organ for digestion and absorption in animals. The presence of gut villi significantly increases the surface area for digestion and nutrient uptake. Gut microbiota also play a critical role in host digestion, absorption, and nutritional metabolism. These microorganisms are capable of synthesizing essential amino acids and unsaturated fatty acids for the host. Furthermore, certain gut microbes can enhance the activity of digestive enzymes-such as amylase, protease, and lipase-thereby aiding the host in nutrient breakdown. Previous research has shown that while crustaceans with different feeding habits possess distinct digestive tract morphologies and microbiomes [33, 34], those with similar diets—even across different families—demonstrate a striking convergence in related morphological and physiological adaptations [33]. Dietary studies on stomatopods have been initiated, yet they are predominantly limited to the widely distributed Oratosquilla oratoria in Chinese coastal waters [35]. In contrast, research into the regulatory mechanisms governing their diet is even scarcer, with our research group having reported solely on the digestive tract characteristics of O. oratoria [36]. This study selected the “smasher” Odontodactylus japonicus and the “spearers” Lysiosquilla sulcirostris and O. oratoria to investigate stomatopod feeding adaptations. The digestive tract phenotypes and gut microbial communities were analyzed by employing multi-level microscopy and 16S rRNA high-throughput sequencing. This research clarifies the regulatory mechanisms behind trophic niche differentiation in stomatopods, thereby establishing a foundation for understanding the evolution of higher stomatopod taxa through the lens of trophic niche.

Materials and methods

Collection of animal specimens and harvesting of gut tissues

Specimens of O. oratoria (avg. length: 107.63 ± 8.85 mm; avg. weight: 21.96 ± 5.41 g), L. sulcirostris (avg. length: 233.47 ± 15.73 mm; avg. weight: 150.72 ± 15.74 g), and O. japonicus (avg. length: 115.67 ± 6.46 mm; avg. weight: 40.56 ± 7.57 g) were manually collected by divers in the Taiwan Strait (Fig. 1). They were transported in an insulated container with sterile seawater to the laboratory and acclimated for 24 h. Subsequently, the animals were placed in a container filled with a seawater and ice mixture (water temperature maintained at 0–4 °C) to induce deep anesthesia within 2 min. The body and raptorial appendages were photographed. Under sterile conditions, the raptorial appendages and gut tracts were dissected for subsequent experiments. The gut samples were longitudinally incised in sterile Petri dishes using a sterile scalpel. All visible chyme within the gut lumen was gently collected with a sterile spatula. The remaining gut tissue was then rinsed three times with sterile phosphate‑buffered saline (PBS) to remove adherent residues, after which the tissue samples were transferred into sterile cryotubes for storage.

Fig. 1.

Sampling sites and raptorial appendage morphology of the three stomatopod species: O. oratoria (top), L. sulcirostris (middle), and O. japonicus (bottom). Note: ML: merus length; PL: propodus length; DL: dactylus length; MW: merus width; MH: merus height

Morphological analysis of raptorial appendages

To quantify the geometric differences in the raptorial appendages of stomatopods, we collected these appendages from 30 individuals each of O. japonicus, L. sulcirostris, and O. oratoria. The following measurements were taken using a vernier caliper: merus length (ML), propodus length (PL), dactylus length (DL), merus width (MW), and merus height (MH) [37]. Simultaneously, the lengths of various digestive system organs were measured, including the cardiac stomach length (CSL), pyloric stomach length (PSL), midgut length (MGL), hindgut length (HGL), and hepatopancreas length (HL) [36]. To account for the influence of body size on the morphology of raptorial appendages and the digestive tract, the measured traits were based on ratios relative to body length. The standardized values were then assessed for normality using the Shapiro–Wilk test and for homogeneity of variances using Levene's test. Finally, one-way analysis of variance (ANOVA) was employed to analyze the morphological differences among the three stomatopod species [37].

Analysis of the midgut tissue structure

Given that the midgut is the primary site of nutrient digestion and absorption [38], this study focused on its tissue structure.The midguts of O. japonicus, L. sulcirostris and O. oratoria (All experimental samples were collected from mature male individuals) were dissected into 5 mm segments and fixed in 4% paraformaldehyde for 24 h. The tissues were then dehydrated through a graded ethanol series (70%, 80%, 90%, and 100%), cleared in xylene, and embedded in paraffin. Serial sections of 4–6 μm thickness were cut from the paraffin blocks using a Leica rotary microtome (Leica Histocore BIOCUT). Finally, the sections were stained with hematoxylin–eosin (HE) and examined under a light microscope (OLYMPUS CX33) for histological observation. The height of the midgut villi and the thickness of the midgut wall were quantified using Image J software (version 1.53a). Differences in these parameters among the three stomatopod species were then assessed for statistical significance using a one-way ANOVA in SPSS software [36].

Midgut tissue samples (< 3mm³) from O. japonicus, L. sulcirostris and O. oratoria were primarily fixed in 2.5% glutaraldehyde at room temperature for 4 h in the dark. The fixative was then replaced by glutaraldehyde, and the samples were stored at 4℃. Subsequently, the samples were washed three times with 0.1 mol/L phosphate buffer (PB), followed by gradient dehydration in an ethanol series (30%, 50%, 70%, 80%, 90%, and 100%). After critical point drying (Quorum K850, East Sussex, UK), the samples were sputter-coated with a platinum film using an ion coater (HITACHI MC1000, Tokyo, Japan). Finally, the samples were observed under a scanning electron microscope (HITACHI SU8100, Tokyo, Japan) at an accelerating voltage of 3 kV to examine the ultrastructure.

DNA extraction, library construction, and sequencing

The experimental animals were temporarily kept for 24 h. The water for rearing was sourced from their natural habitat, with a temperature of 23 °C. No feed was provided during the temporary keeping period. Midgut tissues were aseptically dissected from ten temporarily reared individuals of each species (O. oratoria, O. japonicus, and L. sulcirostris), The tissues were immediately flash-frozen in liquid nitrogen and stored at −80 °C. Genomic DNA was subsequently extracted from the homogenized tissues using the CretMagTM Power Soil DNA Kit (Cretagen Biotechnology, Suzhou, China). The purity and integrity of the extracted DNA were assessed using a NanoDrop2000 spectrophotometer (Thermo Fisher Scientific, USA) and 1% agarose gel electrophoresis.

The V3-V4 hypervariable region of the bacterial 16S rRNA gene was amplified from qualified genomic DNA using barcoded primers 341 F (CCTAYGGGRBGCASCAG) and 806R (GGACTACHVGGGTWTCTAAT) [39]. The PCR products were verified by 1% agarose gel electrophoresis and subsequently purified using the AxyPrep DNA Gel Extraction Kit (Axygen Biosciences, USA).

The purified PCR products were ligated with Y-shaped adapters, and adapter self-ligation fragments were removed using magnetic beads. The library templates were enriched by PCR, and the DNA was denatured with NaOH to generate single-stranded fragments, finalizing the library construction. The resulting DNA libraries were sequenced on the Illumina NovaSeq PE250 platform by Wuhan Wanmo Technology Co., Ltd.

Preprocessing of sequencing data

Raw paired-end reads in fastq format were quality-controlled to generate high-quality data for downstream analysis. Briefly, adapter contamination and low-quality sequences were trimmed using Trimmomatic (v0.36) [40] with a sliding window approach (Number of bases to average across: 50; Average quality required: 20; Minimum length of reads to be kept: 150). Subsequently, reads containing ambiguous bases (N) were removed using PEAR (v0.9.6) [41]. Paired-end reads were merged into consensus sequences using Flash (v1.2.11) [42] and PEAR (v0.9.6) [41] based on their overlap, with the following parameters: minimum overlap of 10 bp, 10% mismatch rate allowed in the overlap region, no mismatches allowed in the barcode, and a maximum of 2 mismatches in the primer region. Chimeric sequences, short sequences(< 120 bp) and mitochondria sequences were filtered out to generate the final set of clean tags, which were then rarefied for subsequent analysis.

OTU clustering and gut microbiota taxonomic annotation

To investigate the gut microbial diversity, clean tags from each species were processed separately. Operational Taxonomic Units (OTUs) were clustered from the clean tags of each species using a 97% similarity threshold via the UPARSE [43] algorithm implemented in VSEARCH (v2.7.1) [44], with analysis conducted in QIIME (v1.8.0) [45]. A Venn diagram [46] was subsequently generated to visualize the shared and unique OTUs among the three stomatopod species.

The taxonomic classification of each OTU was performed using the RDP Classifier [47] against the SILVA database (Release 138.2). The gut microbial composition of the three stomatopod species was visualized using five complementary approaches: (1) a bar plot of taxonomic composition generated with QIIME's qiime taxa barplot command [48]; (2) a phylogenetic tree constructed from aligned OTU sequences (MAFFT v7.505) [49] with Fast Tree (v2.1.11) [50], annotated with abundance information; (3) UPGMA clustering of samples based on Unweighted Unifrac distances, integrated with taxonomic abundance data [51]; (4) a heatmap illustrating interspecific similarities and differences in microbial taxa [52]; and (5) a ternary plot generated with the R (v4.4.1) “grid” package to visualize the relative abundance of species across the three stomatopod species [53].

Analysis of gut microbial diversity

The Alpha diversity of the gut microbiota was assessed using QIIME (v1.8.0) to evaluate differences in microbial community richness and diversity among the three stomatopod species [54]. Several indices were calculated, including Richness, Chao1, ACE, Shannon, Simpson, Invsimpson, Pielou's evenness, Goods-coverage, and Faith's Phylogenetic Diversity (PD-whole-tree). The results were visualized using box plots. Beta diversity was assessed to examine inter- and intraspecific differences in the gut microbiota of stomatopods. Anosim similarity analysis was used for statistical testing [55]. The analysis was conducted using multiple methods: a heatmap based on Weighted Unifrac distance, PCoA based on Jaccard distance, hierarchical clustering (Hcluster), PCA, PLS-DA based on Bray–Curtis distance, and NMDS based on AltGower distance, to comprehensively evaluate the compositional similarities and differences among samples [56].

Analysis of differences in gut microbial composition

To identify differentially abundant microbial taxa, pairwise comparisons among the three stomatopod species were performed using the Wilcoxon rank-sum test [57]. Furthermore, the Kruskal–Wallis test and random forest analysis was applied to identify taxa with significant abundance differences across all three species [58].

Analysis of gut microbial correlations

Microbial co-occurrence networks are a common approach for analyzing community structure [59]. In this study, we used the SparCC algorithm to infer a correlation matrix among gut microbial taxa across the three stomatopod species, aiming to identify co-occurrence or co-exclusion patterns associated with dietary differentiation. The network was constructed using the “igraph” package in R, with a robust correlation coefficient threshold of 0.9 and a significance level of p < 0.01.

Prediction of gut microbial metabolic functions

The functional potential of the gut microbiota was predicted using PICRUSt [53] by mapping the obtained 16S rRNA sequences against database to infer the gene functional profile of the microbial community. Subsequently, differences in predicted metabolic functions among the three stomatopod species were assessed using the Kruskal–Wallis test and random forest analysis [58].

Results

Morphological differences in raptorial appendages and the digestive tract among O. japonicus, L. sulcirostris, and O. oratoria

ANOVA revealed significant differences in all measured ratios of the raptorial appendages (ML/TL, PL/TL, DL/TL, MW/TL, MH/TL; p < 0.001) among the three species. Specifically, O. japonicus exhibited the largest ML/TL, MW/TL, and MH/TL ratios, whereas L. sulcirostris had the largest PL/TL and DL/TL ratios (Fig. 2; Table 1). Similarly, significant differences were found in all digestive tract morphology ratios (HL/TL, MGL/TL, HGL/TL, CSL/TL, PSL/TL; p < 0.001). O. japonicus showed the largest, L. sulcirostris had the largest HL/TL and MGL/TL, and O. oratoria possessed the largest HGL/TL, CSL/TL, and PSL/TL (Table 1).

Fig. 2.

Morphological characteristics of the raptorial appendages and digestive system in O. japonicus, L. sulcirostris, and O. oratoria. Note: The same traits with the same superscript letter on the same line indicate no significant difference between groups (p> 0.05). On the contrary, there is a significant difference (p < 0.05)

Table 1.

Morphological characteristics of the raptorial appendages and the digestive tracts in O. japonicus, L. sulcirostris, and O. oratoria (Mean ± SD)

Morphological indices	Group (mean ± standard deviation)
	Odontodactylus japonicus	Oratosquilla oratoria	Lysiosquilla sulcirostris
ML/TL	0.2209 ± 0.0075a	0.1988 ± 0.0057b	0.1840 ± 0.0078c
PL/TL	0.1702 ± 0.0064a	0.1948 ± 0.0066b	0.2633 ± 0.0118c
DL/TL	0.1612 ± 0.0069a	0.1784 ± 0.0075b	0.2530 ± 0.0103c
MW/TL	0.0804 ± 0.0039a	0.0802 ± 0.0053a	0.0604 ± 0.0026b
MH/TL	0.0527 ± 0.0023a	0.0467 ± 0.0044b	0.0478 ± 0.0024b
CSL/TL	0.0942 ± 0.0137a	0.1286 ± 0.0101b	0.1042 ± 0.0058c
PSL/TL	0.0402 ± 0.0034a	0.0379 ± 0.0048b	0.0290 ± 0.0026c
MGL/TL	0.5550 ± 0.0255a	0.4926 ± 0.0263b	0.5883 ± 0.0277c
HGL/TL	0.0878 ± 0.0063a	0.1352 ± 0.0136b	0.1274 ± 0.0088c
HL/TL	0.7232 ± 0.0347a	0.7084 ± 0.0366a	0.7509 ± 0.0213b

The same traits with the same superscript letter on the same line indicate no significant difference between groups (p > 0.05). On the contrary, there is a significant difference (p < 0.05)

Differences in midgut tissue structure among O. japonicus, L.sulcirostris, and O. oratoria

The midgut wall of all three stomatopod species shared an identical histological composition, comprising from the lumen outward: a mucosal epithelial layer, a submucous laver, a muscular layer, and an outer membrane (Fig. 3A-L). The mucosal epithelium consisted of a single layer of columnar epithelial cells, forming linear longitudinal folds known as midgut villus (Fig. 4A-F). These villi significantly increase the gut surface area for contact with food. The three stomatopod species exhibit significant differences in midgut villus density. Visually, the density in O. japonicus and O. oratoria appears markedly higher than that in L. sulcirostris. While the midgut villus of L. sulcirostris and O. oratoria are relatively uniform in shape, those of O. japonicus display noticeable size variation. The mucosal epithelium consists mainly of columnar cells interspersed with goblet cells. The muscular layer is poorly developed, composed of sparse longitudinal and circular muscle fibers. Significant interspecific variation is also observed in midgut villus length. The villus is longest in O. japonicus (502.63 ± 54.61 μm), followed by L. sulcirostris (179.80 ± 8.02 μm), and shortest in O. oratoria (86.33 ± 7.58 μm). Although no significant difference in villus width exists between O. japonicus and L. sulcirostris, both species possess wider villus than O. oratoria. (Table 2).

Fig. 3.

Midgut tissue morphology in O. japonicus (A-D), L. sulcirostris (E–H) and O. oratoria (I-L). Note: MV: midgut villus, DC: digestive cavity, VL: villus length, VW: villus width, MEL: mucosal epithelial layer, CM: circular muscle, LM: longitudinal muscle, OUM: outer membrane, GC: goblet cells, CC: columnar cells

Fig. 4.

The ultrastructural features of the midgut villus in O. japonicus (A-B), L. sulcirostris (C-D), and O. oratoria (E-F)

Table 2.

Morphometric measurements of gut tissues in O. japonicus, L. sulcirostris, and O. oratoria (Mean ± SD, μm)

Structure	Odontodactylus japonicus	Lysiosquilla sulcirostris	Oratosquilla oratoria
villus length	502.63 ± 54.61a	179.80 ± 8.02b	86.33 ± 7.58c
villus width	61.77 ± 11.38a	54.10 ± 4.88a	24.77 ± 3.23b

The same traits with the same superscript letter on the same line indicate no significant difference between groups (p > 0.05). On the contrary, there is a significant difference (p < 0.05)

Overview of 16S rRNA sequencing data

Quality filtering of Illumina NovaSeq PE250 reads yielded 2,454,867 clean tags, which were clustered into 8,273 OTUs at a 97% similarity threshold. The number of OTUs per sample varied by species: O. oratoria (range: 22–820), L. sulcirostris (54–799), and O. japonicus (142–257). At the individual level, the numbers of unique OTUs in O. japonicus, L. sulcirostris, and O. oratoria ranged from 4 to 82, 1 to 62, and 16 to 30, respectively. Across all 30 samples, the total OTU count was 0 (Fig. 5A). The total number of OTUs identified for each species was 344 (O. oratoria), 113 (L. sulcirostris), and 82 (O. japonicus). A core of 17 OTUs was shared among all three species, while pairwise comparisons revealed 72 shared between O. oratoria and L. sulcirostris, 41 between O. oratoria and O. japonicus, and 20 between L. sulcirostris and O. japonicus (Fig. 5B).

Fig. 5.

Distribution of OTUs abundances in the gut microbiota of O. japonicus (Oj), L. sulcirostris (Ls), and O. oratoria (Oo). Note: A sample-specific OTU abundance; B species-specific OTU abundance

Gut microbial composition of O. japonicus, L. sulcirostris, and O. oratoria

Taxonomic annotation revealed that the gut microbiota of the three stomatopod species was primarily composed of eight phyla, each with a relative abundance exceeding 1%: Bacillota, Pseudomonadota, Patescibacteria, Bacteroidota, Actinomycetota, Cyanobacteria, Verrucomicrobiota, and Fusobacteriota, (Fig. 5A-B). Notably, the gut community of O. japonicus was overwhelmingly dominated by Pseudomonadota (88.90%), whereas Bacillota was the dominant phylum in both L. sulcirostris (45.23%) and O. oratoria (88.87%) (Fig. 6A-B). At the genus level, 20 taxa exhibited a relative abundance greater than 1% across the three stomatopod species: Mycoplasma, Cupriavidus, Sphingomonas, Candidatus_Hepatoplasma, type_III, Incertae_Sedis, Methylobacterium, Hydrotalea, Mesomycoplasma, Candidatus_Obscuribacter, Enhydrobacter, Caulobacter, Mycobacterium, Cetobacterium, Pseudomonas, Halomonas, Paenibacillus, Aquabacterium, Lachnospiraceae_NK4A136_group, Novosphingobium. Among these, Cupriavidus (68.40%) was the dominant genus in O. japonicus, whereas Mycoplasma was the dominant genus in both L. sulcirostris (87.90%) and O. oratoria (17.52%) (Fig. 6C-D).

Fig. 6.

Taxonomic composition of gut microbiota at the phylum (A-B) and genus (C-D) levels in O. japonicus (Oj), L. sulcirostris (Ls), and O. oratoria (Oo)

Gut microbial diversity of O. japonicus, L. sulcirostris, and O. oratoria

Alpha diversity analysis

Alpha diversity of the gut microbiota was assessed using nine indices (Table 3). O. oratoria exhibited the highest microbial richness and diversity, with the greatest values for Richness (46), Chao1 (106.96), ACE (126.32), Shannon (2.14), Simpson (0.62), Invsimpson (7.35), Pielou's evenness (0.55), and Faith's PD (PD-whole-tree, 10.85). In contrast, L. sulcirostris showed the lowest diversity, with minimal values for Richness (13), Chao1 (34.87), Shannon (0.51), Simpson (0.15), Invsimpson (1.83), and Pielou's evenness (0.20). O. japonicus had the lowest ACE (41.90) and PD-whole-tree (4.52) values, while L. sulcirostris had the highest Goods-coverage index (0.96). O. oratoria had the lowest Goods-coverage (0.87). These results indicate that the gut microbiota of O. oratoria has the highest species richness and most balanced community structure, whereas that of L. sulcirostris has the lowest diversity and evenness (Fig. 7).

Table 3.

Alpha diversity indices of O. japonicus, L. sulcirostris, and O. oratoria

Species	Individual	Richness	Chao1	Shannon	Simpson	Invsimpson	Pielou	ACE	Goods-coverage	PD-whole-tree
Odontodactylus japonicus	Oj1	26	39.00	1.49	0.53	2.13	0.46	59.34	0.94	3.78
	Oj2	22	35.75	1.45	0.53	2.11	0.47	40.95	0.95	4.11
	Oj3	20	27.00	1.50	0.55	2.25	0.50	30.01	0.96	3.05
	Oj4	16	17.43	1.16	0.46	1.86	0.42	21.31	0.98	2.72
	Oj5	20	35.00	1.19	0.43	1.74	0.40	34.08	0.95	3.10
	Oj6	19	37.00	1.39	0.53	2.12	0.47	29.96	0.96	3.06
	Oj7	30	56.25	1.83	0.62	2.60	0.54	51.33	0.93	14.17
	Oj8	24	35.00	1.54	0.55	2.25	0.49	40.27	0.95	3.63
	Oj9	23	68.50	1.45	0.54	2.17	0.46	52.93	0.94	3.43
	Oj10	26	43.50	1.32	0.45	1.83	0.40	58.86	0.93	4.18
	Max	30	68.50	1.83	0.62	2.60	0.54	59.34	0.98	14.17
	Min	16	17.43	1.16	0.43	1.74	0.40	21.31	0.93	2.72
	Mean	23	39.44	1.43	0.52	2.11	0.46	41.90	0.95	4.52
Oratosquilla oratoria	Oo1	34	41.33	2.52	0.85	6.56	0.71	48.59	0.95	8.88
	Oo2	55	178.00	1.96	0.56	2.30	0.49	171.81	0.81	18.57
	Oo3	45	178.20	1.58	0.48	1.91	0.42	300.39	0.83	12.14
	Oo4	78	172.23	3.49	0.92	11.99	0.80	208.73	0.77	12.07
	Oo5	82	148.35	3.69	0.94	17.76	0.84	184.02	0.78	10.69
	Oo6	60	126.60	3.01	0.89	8.88	0.74	128.57	0.83	17.51
	Oo7	69	181.88	3.55	0.95	20.21	0.84	167.69	0.80	17.60
	Oo8	22	31.00	1.27	0.44	1.79	0.41	35.01	0.95	7.10
	Oo9	4	4.50	0.11	0.04	1.04	0.08	7.00	0.99	2.69
	Oo10	6	7.50	0.25	0.09	1.10	0.14	11.34	0.99	1.18
	Max	82	181.88	3.69	0.95	20.21	0.84	300.39	0.99	18.57
	Min	4	4.50	0.11	0.04	1.04	0.08	7.00	0.77	2.69
	Mean	46	106.96	2.14	0.62	7.35	0.55	126.32	0.87	10.85
Lysiosquilla sulcirostris	Ls1	3	4.00	0.06	0.02	1.02	0.05	NA	0.99	3.21
	Ls2	1	1.00	0.00	0.00	1.00	NA	NA	1.00	NA
	Ls3	9	16.50	0.32	0.11	1.12	0.14	25.36	0.97	4.40
	Ls4	62	179.14	3.05	0.88	8.50	0.74	159.10	0.81	14.58
	Ls5	6	9.00	0.17	0.05	1.06	0.09	16.00	0.98	1.68
	Ls6	2	2.00	0.03	0.01	1.01	0.04	NA	1.00	0.66
	Ls7	5	11.00	0.12	0.04	1.04	0.07	NA	0.98	1.65
	Ls8	24	54.60	0.83	0.26	1.35	0.26	83.80	0.92	4.33
	Ls9	18	70.50	0.56	0.17	1.21	0.19	94.89	0.93	8.38
	Ls10	1	1.00	0.00	0.00	1.00	NA	NA	1.00	NA
	Max	62	179.14	3.05	0.88	8.50	0.74	159.10	1.00	14.58
	Min	1	1.00	0.00	0.00	1.00	0.04	16.00	0.81	0.66
	Mean	13	34.87	0.51	0.15	1.83	0.20	75.83	0.96	4.86

Fig. 7.

Alpha diversity indices of O. japonicus (Oj), L. sulcirostris (Ls), and O. oratoria (Oo)

Beta diversity analysis

Beta diversity analysis revealed distinct clustering of gut microbial communities among the three stomatopod species. ANOSIM testing confirmed that interspecific differences were statistically significant (p < 0.05) and greater than intraspecific variation (R > 0 for all comparisons) (Table 4, Fig. 8). The tightest clustering was observed in O. japonicus, followed by L. sulcirostris, while O. oratoria exhibited the most dispersed distribution pattern. Correspondingly, the confidence ellipse area was smallest for O. japonicus, intermediate for L. sulcirostris, and largest for O. oratoria, indicating greatest inter-individual variation in the latter species. Heatmap analysis further supported these patterns, showing low, moderate, and high dissimilarity blocks for O. japonicus, L. sulcirostris, and O. oratoria, respectively (Fig. 9A-F). These results demonstrate that feeding habit specialization drives species-specific gut microbiota composition, with O. oratoria showing the highest individual-level variability.

Table 4.

ANOSIM of gut microbial communities among O. japonicus (Oj), L. sulcirostris (Ls), and O. oratoria (Oo)

Group	R	P-value	P-adj-BH
Oj/Oo	0.62	0.001	0.001
Oj/Ls	0.98	0.001	0.001
Oo/Ls	0.49	0.001	0.001

Fig. 8.

ANOSIM of gut microbial communities among O. japonicus (Oj), L. sulcirostris(Ls), and O. oratoria (Oo)

Fig. 9.

Beta diversity analysis of gut microbial communities in O. japonicus (Oj), L. sulcirostris (Ls), and O. oratoria (Oo). Note: A Heatmap (Weighted Unifrac); B PCA (Bray–Curtis); C NMDS (altGower); D Hierarchical clustering (Bray–Curtis); E PCoA (Jaccard); F PLS-DA (Bray–Curtis)

Differential abundance of gut microbiota among O. japonicus, L. sulcirostris, and O. oratoria

Differential abundance analysis using the Wilcoxon test revealed significant differences at the phylum level (Fig. 10). The abundance of Pseudomonadota was significantly higher in O. japonicus compared to both L. sulcirostris and O. oratoria. Conversely, the abundance of Bacillota was significantly lower in O. japonicus compared to both L. sulcirostris and O. oratoria. Additionally, the abundance of Bacillota in O. oratoria was significantly lower than that in L. sulcirostris (Fig. 10A-C). At the genus level, significant differences in microbial abundance were detected among the three species. The abundance of Sphingomonas, Paenibacillus, Mycobacterium, Methylobacterium, Hydrotalea, Enhydrobacter, Cupriavidus, Caulobacter, Candidatus-Obscuribacter, and Bosea was significantly higher in O. japonicus than in O. oratoria. Compared to L. sulcirostris, O. japonicus also had a significantly higher abundance of the aforementioned genera plus Novosphingobium, Pseudomonas and Sediminibacterium. Conversely, O. oratoria exhibited a significantly higher abundance of Mycoplasma than O. japonicus. Notably, the abundance of Mycoplasma was highest in L. sulcirostris, being significantly greater than in both O. japonicus and O. oratoria (Fig. 10D-F).

Fig. 10.

Wilcoxon rank-sum test results showing differentially abundant taxa at the phylum (A-C) and genus (D-F) levels among O. japonicus (Oj), L. sulcirostris (Ls), and O. oratoria (Oo)

Both Random Forest analysis (Fig. 11A) and the Kruskal–Wallis test (Fig. 11B) identified significant differences (p < 0.05) in the relative abundances of several bacterial genera among the three stomatopod species. These discriminatory taxa included Mycoplasma, Cupriavidus, Hydrotalea, Sphingomonas, Paenibacillus, Caulobacter, Methylobacterium-Methylorubrum, Candidatus-Obscuribacter, Enhydrobacter, Candidatus-Hepatoplasma, Mycobacterium, and Bosea.

Fig. 11.

Identification of differentially abundant genera in gut microbiota of O. japonicus (Oj), L. sulcirostris (Ls), and O. oratoria (Oo) Random Forest analysis (A) and Kruskal-Wallis test (B)

Network analysis and functional prediction of gut microbiota in O. japonicus, L. sulcirostris, and O. oratoria

A co-occurrence network was constructed based on OTU data to analyze microbial interactions in the three stomatopod species. The results revealed that both O. japonicus and O. oratoria formed seven distinct modules, whereas L. sulcirostris formed five. Notably, the modular structure of L. sulcirostris exhibited higher connectivity and integration compared to the other two species (Fig. 12).

Fig. 12.

Co-occurrence network analysis of gut microbiota at the OTU level in O. japonicus (Oj), L. sulcirostris (Ls), and O. oratoria (Oo)

Functional prediction of the gut bacterial communities was performed using PICRUSt based on database. The analysis annotated six primary pathways: Metabolism, Environmental Information Processing, Genetic Information Processing, Cellular Processes, Human Diseases, and Organismal Systems. At the secondary pathway level, the top 30 most abundant functions included Biosynthesis of ansamycins, Valine, leucine and isoleucine biosynthesis, Bacterial chemotaxis, and Fatty acid biosynthesis, among others (Fig. 13). Kruskal–Wallis tests revealed significant interspecific differences (p < 0.05) in 95 predicted pathways. The gut microbiota of O. japonicus exhibited significantly higher abundances in pathways related to core nutritional processes compared to L. sulcirostris and O. oratoria. These included amino acid degradation (e.g., valine, leucine, isoleucine, lysine), carbohydrate and energy metabolism (e.g., pyruvate metabolism, TCA cycle), lipid metabolism (e.g., fatty acid metabolism), vitamin and cofactor metabolism (e.g., thiamine metabolism), and nutrient transport (e.g., ABC transporters). Additionally, pathways for degradation of complex compounds were significantly enriched in O. japonicus (e.g., polycyclic aromatic hydrocarbon degradation) (Fig. 14).

Fig. 13.

Functional annotation and abundance profiling of gut microbiota in O. japonicus (Oj), L. sulcirostris (Ls), and O. oratoria (Oo)

Fig. 14.

Kruskal-Wallis test analysis of predicted functional abundance in gut microbiota of O. japonicus (Oj), L. sulcirostris (Ls), and O. oratoria (Oo)

Discussion

Stomatopods are highly trophic level marine crustaceans that play a crucial role in benthic ecosystems by maintaining energy flow and dynamic balance, utilizing their specialized raptorial appendages as primary hunting organs [27, 60]. To efficiently utilize food resources and avoid nutritional niche competition among closely related species [61], the raptorial appendages of stomatopods have diverged into two distinct forms during evolution: “smashers” and “spearers”, potentially resulting in different predatory strategies [28–32]. Nevertheless, systematic studies on this aspect are scarce. We conducted a comprehensive analysis on three stomatopod species, employing morphological measurements of raptorial appendages and digestive tracts, midgut histological observation, and gut microbiome sequencing. This study aims to systematically elucidate the morphological and physiological regulatory mechanisms underlying their dietary differentiation.

The morphological differentiation of the raptorial appendages is readily observable and is substantiated by quantitative measurements of morphological indices. While previous studies have suggested that environmental factors like water flow and substrate are primary drivers of this differentiation [62], the appendage morphology itself directly reflects hunting methods, capabilities, and foraging ranges [22, 23, 63]. Consequently, adaptive variation in appendage morphology to different habitats necessitates divergent hunting strategies, ultimately leading to trophic niche differentiation.

Feeding preferences are closely correlated with morphological differentiation of the digestive tract [64, 65]. Moreover, these morphological differences can serve as diagnostic characters for species identification and classification [66, 67]. As the longest segment of the digestive tract, the midgut secretes both permeable digestive enzymes and a protective mucus layer. This dual secretion protects the gut from mechanical wear during digestion and thereby enables efficient nutrient absorption [68–70]. Previous studies have confirmed that midgut morphology varies significantly among animals in accordance with their dietary types [71]. This study revealed that the midgut of O. japonicus exhibited the greatest villus width and villus length, followed by L. sulcirostris with intermediate values, while O. oratoria showed the lowest measurements. These morphological differences are likely adaptations to the physical form and digestibility of their respective prey [67]. The midgut typically features a thick epithelial layer composed of digestive and regenerative cells, along with a relatively thin muscular layer [72, 73]. Based on this structural principle, the thicker gut epithelium observed in O. japonicus suggests a potentially enhanced capacity for digestive enzyme secretion and the rhythmic regeneration of digestive cells, compared to L. sulcirostris and O. oratoria [74, 75]. It is noteworthy that the epithelial cells of the midgut differentiate into villi, structures that increase the digestive surface area and regulate the filtration of food particles [76]. Longer villi exhibit a greater capacity for filtering solid substances, facilitating adaptation to liquid diets rich in carbohydrates but poor in nitrogen [66], whereas shorter villi are associated with contrasting functions. Consequently, the interspecific differences in midgut villus length among the three stomatopod species may indirectly reflect their divergent feeding preferences.

Given that phylogenetic, dietary, habitat, and nutritional factors directly shape gut microbial communities [9, 15, 77–79], it follows that stomatopods with distinct feeding habits harbor significantly different gut microbiota. In this study, the highly conserved 16S rRNA gene was employed as a marker to characterize the gut microbiota of three stomatopod species [80]. Analysis of the sequencing data revealed that O. oratoria possessed the highest number of OTUs, followed by L. sulcirostris, with O. japonicus howing the lowest OTU count, indicating interspecific differences in microbial diversity. Furthermore, both alpha and beta diversity analyses confirmed that the gut microbiota of O. oratoria exhibited the highest richness and diversity. Given the distinct phylogenetic backgrounds, habitats, and feeding preferences of the three stomatopod species [30,31), their gut microbial communities may represent a deterministic-stochastic balance shaped by different environmental factors and selection pressures [81]. Previous studies indicate that gut microbial diversity is positively correlated with host environmental adaptability [82]. Therefore, the relatively high microbial diversity observed in O. oratoria may be closely linked to its broad distribution range [83], suggesting enhanced ecological flexibility. The more complex co-occurrence network in O. oratoria suggests that its gut microbiota exhibits stronger interactions, greater stability, and higher resistance to disturbance [84]. Furthermore, the low number of shared OTUs (only 17 among the three species), with higher overlap between O. oratoria and L. sulcirostris than with O. japonicus, indicates limited microbial sharing. We hypothesize that habitat-driven geographical isolation promotes host genetic differentiation, which in turn restricts the horizontal transmission of gut microbes. This limitation fosters the co-evolution of host-microbe symbiotic relationships, leading to the formation of species-specific microbial communities and subsequent divergence in nutritional strategies among stomatopods [85–87]. Nevertheless, the observed differences in gut microbiota similarity among the three stomatopod species may be attributable to geographical isolation. Additionally, the shared "spearer" predatory strategy of O. oratoria and L. sulcirostris may also contribute to their higher degree of microbial sharing, as similar feeding habits can select for convergent microbial communities [31].

The core gut microbiota plays a crucial role in shaping interspecific differences in gut physiological functions [88]. Wilcoxon tests revealed significant differences in the core microbiota composition among the three stomatopod species. Specifically, Bacillota dominated the gut communities of O. oratoria and L. sulcirostris, whereas Pseudomonadota was the dominant phylum in O. japonicus. The analysis indicates that genetic variation, environmental microbial composition, and dietary differentiation are the primary factors driving these disparities [89, 90]. Furthermore, Bacillota and Pseudomonadota have been confirmed as dominant bacterial phyla in the intestines of various crustaceans [91–93]. Bacillota are hypothesized to contribute to lipid metabolism by reducing fat deposition and enhancing fatty acid absorption through the promotion of diverse digestive enzymes. Additionally, they can degrade fibers into volatile fatty acids for host utilization [94–96]. In contrast, Pseudomonadota have been demonstrated to play a crucial role in degrading organic macromolecules and participating in amino acid metabolism [97–99]. These findings suggest that O. oratoria and L. sulcirostris may possess strong capabilities for lipid digestion and absorption, whereas O. japonicus exhibits potential for enhanced amino acid metabolism, although further validation is required. Functional analysis revealed significant differences in amino acid, lipid, and carbohydrate metabolism among the three species. This functional divergence in microbial metabolic capacity likely enhances host utilization of dietary resources, promoting trophic niche differentiation and improving adaptability to varying food supplies [100, 101]. Despite having intermediate microbial diversity, O. japonicus showed the highest abundance of metabolic genes for amino acids, lipids, and carbohydrates, implying a potentially superior digestive efficiency. The gut microbiota profile of O. oratoria may reflect a trade-off between environmental adaptation and specialized feeding functions, while the low microbial abundance in L. sulcirostris may limit functional specialization. Consequently, we propose that stomatopods with “smasher” raptorial appendages possess stronger food digestion and absorption capabilities than those with “spearer” appendages, although this hypothesis requires further experimental validation. Although this study did not directly determine the genetic data of the hosts, we will make reasonable inferences based on previous research. The three species of stomatopods exhibit significant differences in morphology and feeding habits, and these phenotypic differences are likely to reflect the long-term genetic differentiation underlying them. This genetic differentiation drives the differentiation of the microbiota composition, and more closely related stomatopods have more similar gut microbial communities.

Conclusion

This study systematically elucidated the morphology of the raptorial appendages and digestive tract, as well as the gut microbial composition, of O. japonicus, L. sulcirostris, and O. oratoria. The raptorial appendages, digestive tract lengths, and midgut tissue structures all exhibited significant interspecific differences among the three stomatopod species. In terms of gut microbiota composition, O. japonicus was dominated by the phylum Pseudomonadota, with Cupriavidus as the predominant genus. In contrast, both L. sulcirostris and O. oratoria showed Bacillota as the dominant bacterial phylum, and Mycoplasma as the most abundant genus. Our findings provide valuable morphological and physiological data for exploring the trophic niche differentiation underlying their specially raptorial appendages. Furthermore, this work contributes foundational knowledge for advancing the nutritional ecology theory of stomatopods. Future studies should incorporate a greater number of species to reconstruct the evolutionary history of Stomatopoda from a trophic perspective.

Authors’ contributions

Conceptualization, J.Z. and F.L.; methodology, F.L.; software, J.Z.; validation, Y.W. (Yilin Wang) and X.D.; formal analysis, J.Z.; investigation, J.Z., F.L. and Y.W. (Yifei Wang); resources, F.L.; data curation, J.Z., F.L. and R.W.; writing—original draft preparation, J.Z.; writing—review and editing, J.Z., F.L.; visualization, J.Z.; supervision, F.L.; project administration, F.L.; funding acquisition, F.L. All authors have read and agreed to the published version of the manuscript.

Funding

This work was supported by the Natural Science Foundation of Shandong Province (ZR2024MD076) and the opening foundation of the Observation and Research Station of Bohai Strait Eco-Corridor, MNR (BH202401).

Data availability

The 16S rRNA reads of three stomatopod species generated during the current study are available in the NCBI SRA under BioProject accession number PRJNA1355003.

Declarations

Ethics approval and consent to participate

The use of animals in this study was approved by the Experimental Animal Ethics Committee of Yantai University. All procedures complied with ethical standards as the species involved are non-protected and the research raised no ethical concerns.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.