https://orcid.org/0000-0001-6608-8314
Figures
Abstract
The Japanese flying squid Todarodes pacificus (Ommastrephidae) is a commercially and ecologically important species; however, there remains much room for investigation in its early life phase, especially its diet in wild environments. After excising the digestive gland (cecum sac) of wild paralarvae of T. pacificus using Laser Microdissection (LMD), the dietary species were estimated via metagenomic analysis. The 16S rRNA analysis predominantly detected Burkholderiales and Xanthomonadales, regardless of mantle length (ML) of T. pacificus and capture area. COI (Cytochrome c oxidase subunit I) analysis detected in various organisms including Discosea, Arthropoda, Nemertea, Porifera, golden algae, and fungi (Ascomycota and Basidiomycota), which were found irregularly. About half of the paralarval cecum sacs were found empty during the histological analysis. We also estimated the trophic position (TP) of wild paralarvae in the same sea region via stable isotope analysis of amino acids. The TP estimated was 3.0 for all larval groups regardless of ML, suggesting that the trophic tendency of paralarvae is carnivorous, likely feeding on herbivorous organisms. Taken together, our results suggest that the paralarvae feed mostly on various kinds of living herbivorous organisms and partly on detritus.
Citation: Adachi K, Chikaraishi Y, Nomura S, Goto-Inoue N, Zaima N, Kuriya Y, et al. (2026) New insights into the feeding behaviour of the Japanese squid Todarodes pacificus paralarvae, and a combined analysis of metagenome and amino acid isotope ratios. PLoS One 21(3): e0340579. https://doi.org/10.1371/journal.pone.0340579
Editor: Claudio D'Iglio, University of Messina, ITALY
Received: August 2, 2025; Accepted: December 23, 2025; Published: March 16, 2026
Copyright: © 2026 Adachi et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: The metagenome data are available in the DDBJ of Japan Sequence Read Archive (DRA) (SUBMISSION: DRA021296, EXPERIMENT: DRX665417- DRX665444). Direct link is are follows. https://ddbj.nig.ac.jp/search/entry/sra-submission/DRA021296, https://ddbj.nig.ac.jp/search/entry/sra-experiment/DRX665417, https://ddbj.nig.ac.jp/search/entry/sra-experiment/DRX665444.
Funding: This work was partially supported by JSPS KAKENHI (grant numbers: 16H04963, 20H03055, and 23K26975) for KA, YT, NM and JY. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors declare no competing interests.
1. Introduction
Cephalopods are among the most important fishery products worldwide and are known to be key links in food webs. They act as intermediaries in the energetic flux between diverse trophic levels and are normally subdominant mesopredators and important preys [1–3]. The early life stage of cephalopods, known as paralarvae, are planktonic and characterized by distinct morphological and ecological traits from those of later developmental phases, such as juveniles and adults [4,5].
The epipelagic Japanese flying squid (Todarodes pacificus Steenstrup, 1880, Cephalopoda: Ommastrephidae) is the predominant species in the aquatic bodies west of Japan [6]. Adults of T. pacificus are basically predatory, and their ontogenetic dietary shifts from crustaceans (e.g., copepods) in juveniles (<50 mm mantle length (ML)) to fish and squids in the larger individuals (>150 mm), as proposed based on the stomach content analysis [7]. Stable isotope analysis (δ13C and δ15N of bulk tissue) also suggested ontogenetic dietary shifts and their seasonal variations [7]. However, information on the diet of early paralarvae (ML, 1–3 mm) is still limited owing to technical difficulties in treatment. The size of digestive organs within the mantle of such paralarvae is too small for visual observation under the microscope [8]. In addition, scant trophic information has been available because the amount of digestive content is too small (< 0.25 mm in stage 34) for a traditional stable isotope analysis [8].
Recently, metabarcoding analysis after Laser Microdissection (LMD) of digestive organs has allowed us to identify the comprehensive diet of Ommastrephidae squid paralarvae [9,10]. Equipped with a focused laser beam, LMD is a highly selective process for the precise separation of samples, and is a microscope-controlled manipulation technique. After the extraction of DNA from LMD samples of the digestive organ, metagenome analysis using Next-Generation Sequencing (NGS) has emerged to identify the diet of juveniles of marine species, such as fish or cephalopods. In the juveniles of flying squids (family Ommastrephidae), the gut contents are assemblages of fungi, plants, algae, and other animals from marine and terrestrial origins, as well as eukaryotic and prokaryotic microorganisms, suggesting that the diet of the juveniles consists of detritus materials [9]. However, as the metagenomic analysis shows the species composition in the digestive tract remains for only a snapshot term of the juvenile phase, the species composition observed is not always comparable to the diet that is actually assimilated for the integrated term of the juvenile phase, suggesting that metagenome analysis alone is insufficient for identifying the trophic tendency of juvenile flying squids.
The analysis of stable isotope ratios of amino acids is a powerful tool used to accurately determine the positions of heterotrophic organisms in the trophic level hierarchy of ecological food webs [11–15] and to identify the basal resources for the organisms in their food chains [15]. The nitrogen isotope ratios (δ15N) of the trophic amino acids (e.g., glutamic acid) predictably increase by 4–9‰ at each step in trophic elevation, which is attributed to the isotopic fractionation during the deamination of these amino acids during metabolism [11]. In contrast, the change in the δ15N value of the source amino acids (e.g., phenylalanine) is always limited to 0–1‰ at each trophic level, because deamination is not a dominant process for these amino acids in the metabolism [11]. Using these values, the trophic position (TP) of organisms can be inferred as integrated information on the trophic tendencies of organisms in food webs [11,12]. However, unlike metagenomic analysis, the TP inferred through stable isotope analysis does not include any information that can be used to directly identify the potential diet species of the organisms.
Thus, combining metagenomic and amino acid isotope analyses provides a powerful and complementary approach for elucidating the diet of early Ommastrephidae paralarvae. This integrated framework is expected to yield a synergistic understanding of dietary composition and trophic structure. In this study, we conducted metagenomic analysis of the digestive organs of T. pacificus paralarvae—following LMD treatment—along with trophic position estimation based on the stable nitrogen isotope ratios of amino acids.
2. Materials and methods
2.1. Metagenome analysis
2.1.1. Samples.
Paralarvae were collected from two locations. The first location was in the northern part of the East China Sea, where paralarvae were collected between February 4 and 25, 2015 (St. 1–3 in Fig 1), with a bongo net that was diagonally towed from a depth of 100 m for recovery. The collected paralarvae were fixed in 10% neutral formalin solution and then stored in 55% ethanol at 4°C. Individuals collected from Stations 1 and 2 were divided into 12 groups according to their ML for metagenomic analysis. The second location was in the Sea of Japan between October 16 and 30, 2016 (St. 4–6 in Fig 1). The collection and fixation of paralarvae were performed using the same procedure as that for the samples from the East China Sea. For metagenomic analysis, the samples collected from Stations 1 and 2 in the East China Sea (10 individuals) were divided into three groups, A, B, and C, according to the ML. Samples collected from the Sea of Japan were formed the Group D for metagenomic analysis, regardless of ML. Details of the sample locations and samples are summarized in Table 1. The samples were collected with permission from the Fisheries Resource Assessment of the Fisheries Agency of Japan, following the guidelines for the Care and Welfare of Cephalopods in Research [16]. All experimental procedures were conducted in accordance with the Animal Experimentation at the Fisheries Resources Institute of the Japan Fisheries Research and Education Agency (FRA) and the experimental guidelines set by Hokkaido University.
The details are summarized in Table 1.
2.1.2. Laser Microdissection (LMD).
To prepare the larvae for sectioning, the samples were washed in PBS for 30 min and replaced with 10% glucose for 1 h, 20% glucose for 24 h, and then with 30% glucose for another 24 h at room temperature (around 20°C), and finally, placed in Optimal Cutting Temperature (OCT) compounds (Sakura finetek, USA) and stored at 5°C until sectioning. Thereafter, the paralarvae were transferred to plastic embedding containers, embedded in the OCT compound, and kept at −80°C for 5 min to create frozen blocks. The frozen sections of 10-μm thickness were prepared with a cryostat at −16°C. After the bulging cecum sac was checked under a microscope, sectioning was continued, and the sections were affixed to polyethylene naphthalate (PEN) slides (Leica, Germany). Fifteen sections were affixed to separate PEN slide, producing approximately 45–60 sections per individual. These sections were stored at −80°C. Frozen sections containing only the OCT compound were used as negative controls. Gloves were worn during all procedures and equipment was cleaned with 70% ethanol to avoid contamination. Before using the LMD 6500 (Leica), the workbench and equipment were cleaned with 70% ethanol. The cecum sac was carefully excised so as to not include any other organs (Fig 2a, b), collected in a PCR tube containing 20 μL of sterile distilled water (Millipore, USA), and stored at −30°C.
a. Sagittal section of Todarodes pacificus paralarvae on a PEN slide stained with Toluidine Blue O. (B) Vertical section (3 μm) of the paralarvae stained with Hematoxylin–Eosin. a. Paralarvae with swollen cecum sacs. b. Paralarvae with empty cecum sacs.
2.1.3. DNA extraction.
To remove the film from the PEN slides, the PCR tubes were sonicated for 1 min. A parafilm was wrapped around the opening of the tube to prevent water from entering. Subsequently, DNA extraction was altered for each test. In the first test, DNA extraction was performed in the following order: the LMD sample was transferred to a master tube and placed in a freeze-dryer. Two iron balls were placed in the master tube in a cell grinder (20 rpm, 5 min). Then, 500 μL of DNA extraction buffer (10 mM Tris-HCl, pH 8.0 including 10 mM EDTA and 1% SDS) was added and vortexed, and the solution was transferred to a 1.5-mL tube. Five microliters of proteinase K (800 U/mL; Qiagen, Germany) were added and incubated at 55°C for 2–3 h. An equal volume of PCI was added, vortexed, and centrifuged at 11,000 rpm for 10 min at room temperature. The upper layer was removed; 50 μL of 0.3 M NaCl, 1 μL of glycogen solution, and 1 mL of 100% ethanol were added; and the mixture was centrifuged at 11,000 for 10 min at 4°C. All materials, except the precipitate, were removed and washed with 70% ethanol. The precipitate was dried and 15 μL of sterile distilled water was added. The concentration of the extracted DNA was measured using Synergy H1 (BioTek, USA) and QuantiFluor dsDNA system (Promega, USA). All pipettes were autoclaved, and filter tips (Watson) were used. All reagents required for extraction were commercially available for molecular biology experiments.
2.1.4. PCR and Sequencing.
PCR was performed using the 16S rRNA and cytochrome c oxidase subunit I (COI). 16S rRNA: The V4 region was amplified using a two-step PCR. AMPure XP (Beckman Coulter, USA) was used to purify the first PCR amplicon. COI: The library was prepared using a two-step tailed PCR method, which included LNA primers to inhibit the amplification of the CO1 gene from T. pacificus. The library was prepared using the Synergy H1 and QuantiFluor dsDNA systems (Promega). A Fragment Analyzer and dsDNA 915 Reagent Kit (Advanced Analytical Technologies) were used to evaluate the quality of the prepared library. NGS was performed using MiSeq (Illumina, San Diego, CA, USA) at 2 × 300 bp. The details of the PCR conditions, sequencing results, and measurement table for the FASTQ file are summarized in Supporting Information 1. These data are available in the DDBJ of Japan Sequence Read Archive (DRA) (SUBMISSION: DRA021296, EXPERIMENT: DRX665417- DRX665444).
2.1.5. Bioinformatic analysis.
The FASTQ barcode splitter in the FASTX-Toolkit was used to extract the sequences whose initial sequence reads matched those of the primers. Only sequences whose initial sequence reads exactly matched those of the primers used were extracted. We then removed sequences with quality values less than 20 using Sickle tools and discarded sequences with more than 40 bases in length and their pair sequences. The paired-end merge script FLASH was used to merge the sequences that underwent quality filtering. For the 16S rRNA V4 region, 1) the fragment length after merging was 260 nucleotides, 2) reads fragment was 230 nucleotides long, and 3) minimum overlap length was 10 nucleotides. For COI, the conditions for merging were as follows: 1) the fragment length after merging was 320 nucleotides. Other conditions were identical to those of 16S rRNA analysis. OTU creation was conducted under the condition of 97% sequence homology using USearch. A BLAST search was performed using the representative sequences of the created OTUs against the nucleotide dataset to conduct phylogenetic inference.
2.2. Stable isotope analysis for amino acids
2.2.1. Sample preparation.
The paralarvae collected at Station 3 were used (Fig 1). The larvae were divided into four groups of five individuals each according to the ML (I, 0.72–0.82 mm; II, 1.2–1.8 mm; III, 2.2–2.8 mm; and IV, 3.3–3.8 mm) (Table 1). The mantle was cut from the larvae using a razor to avoid damaging internal organs. The cut mantles from each group were mixed to prepare four samples. Each sample was preserved in 55% ethanol and stored at room temperature until further analysis. Details of the sample locations and samples are summarized in Table 1. These samples were collected with the permission and used in accordance with the experimental guidelines, as described in the metagenomic analysis experiment.
2.2.2. Determination of trophic position of paralarvae.
Measurements were performed according to the methods described by our previous reports [11,12]. The outline was as follows: Proteins in the samples were hydrolyzed into amino acids using 12 M hydrochloric acid (100°C for 12–24 h). The hydrolyzed solution was defatted using dichloromethane/n-hexane (2/3, v/v). Isopropyl esterification was performed using thionyl chloride/isopropanol (1/4, v/v, 110°C for 2 h), followed by acylation using pivaloyl chloride/dichloromethane (1/4, v/v, 110°C for 2 h). Finally, pivaloyl/isopropyl ester derivatives of the amino acids were obtained through liquid–liquid extraction using water–dichloromethane/n-hexane. The stable nitrogen isotope ratios (δ15N) of amino acids in each sample were measured by gas chromatography/isotope ratio mass spectrometry (GC/IRMS) using a 7890B gas chromatograph (Agilent Technologies, USA) and a Delta V isotope ratio mass spectrometer (Thermo Fisher Scientific, USA). Measurements were performed thrice for each sample, and the average δ15N values of glutamic acid and phenylalanine were used to estimate the trophic position (TP) of the samples [17]. The TP was calculated using the following formula:
3. Results
3.1. Laser microdissection (LMD) and Histology
For histological analysis under LMD, the cecum sacs were excised to purify the DNA (Fig 2A). The cecum sacs were observed to be swollen in 50–60% of the individuals at each station (Fig 2B). Under microscopic observation, yolk was not observed even in the smallest samples (Group A), suggesting that all samples were post-yolk absorbed and that the effects of parental nutrition can be ignored.
3.2. Metagenome analysis for 16S rRNA (procaryote)
The most represented group was Proteobacteria (86% of the reads), with the dominant orders being Burkholderiales (beta proteobacteria, 48.1%) and Xanthomonadales (gamma proteobacteria, 21.6%) (Fig 3). Upon examining each sample, we found that these bacteria were dominant in all individuals, and that no significant difference based on the ML or capture area was observed. The details of the results are summarized in Supporting Information 2.
The x-axis represents the sample id summarized in Table 1. The y-axis represents the percentage of bacteria.
3.3. Metagenome analysis for COI (eucaryote)
Discoseae (30.8%), Arthropoda (28.7%), Ascomycota (23.4%), and Basidiomycota (4.6%) accounted for 87% of the total (Fig 4). Discosea, a class of Amoebozoa, was most commonly observed in Groups A (ML, 0.82–1.1 mm) and B (1.3–1.9 mm) from the East China Sea and Group D (2.6–3.8 mm) from the Sea of Japan. The abundance of Amoebozoa could be mainly ascribed to Cochliopodium arabianum at the OTU level. Acanthamoeba castellanii was also detected in Groups A and D. In Arthropoda, sporadic profiles were obtained regardless of the individuals, ML, or sampling points (e.g., Samples 3, 4, 7, 9, and 10 in large numbers). In Samples 3 and 9, this value was mainly due to the midge (Harnischia japonica). The data in sample numbers 4, 7, and 10 were mainly due to the genera Insects and Arachnida (Ecliptopera, Haplothrips, Fujientomon, and Dermatophagoides). Ribbon worms (Nemertea; Cerebratulus), marine sponges (Porifera; Desmacella), and golden algae (Poterioochromonas and Pedospumella) were also found occasionally. Notably, more than a certain amount (>5%) of Ascomycota was found in almost all individuals, but there was no universality of OTU levels among individuals. For Basidiomycota, less but significant higher number was detected in Group D (2.6–3.8 mm).
The x-axis represents the sample id summarized in Table 1. The y-axis represents the percentage of eukaryotic organism.
Using blocking primers, amplification of T. pacificus COI was mostly suppressed. We checked the validity of the primers used for PCR using the copepod genome as a template, indicating the absence of copepods in the cecum sacs of the paralarvae. The details of the analyzed results are summarized in Supporting Information 3.
3.4. Stable isotope analysis for amino acids
The trophic position of the paralarvae in all groups was approximately 3.0, regardless of the ML (TP, I: 2.9, II: 2.9, III: 3.0, IV: 3.0; Fig 5). This indicated that the diet of the paralarvae was herbivorous, with a TP of 2.0. The δ15N value of Phe, which strongly reflects the δ15N value of diets, in Group I (ML < 1 mm) was significantly higher than that of other groups, which may be caused by that the depth of habitat of Group I different from that of the other group [18]. The δ15N values of Ala were higher than those of Glu at ML of 1–2 and 3–4 mm, suggesting that diets experience corrosion under hypoxic conditions within the food chain to the paralarvae [19]. These results are summarized in Supporting Information 4.
The x-axis represents the group name summarized in Table 1. The y-axis represents TP.
4. Discussion
In this study, we performed a combination analysis of metagenome and amino acid stable isotope ratios to identify the diet of the paralarvae of T. pacificus. Fig 6 is a schematic summarizing the present study. In the metagenomic analysis, the diet composition varied depending on the individual. Stable isotope analysis showed that the TP of paralarvae was approximately 3.0, regardless of ML, indicating that the trophic position of the paralarvae diet was 2.0 on average. Collectively, we suggest that paralarvae (TP = 3.0) sporadically feed on a variety of organisms near herbivorous species (TP = 2.0) in the trophic food web.
A) Results of metagenomic analysis. Burkholderiales and Xanthomonadales were detected predominantly by 16S rRNA analysis, regardless of T. pacificus ML and sampling area. Diverse organisms, including Discosea, Arthropoda, Nemertea, Porifera, golden algae, and fungi (Ascomycota and Basidiomycota), were irregularly detected by cytochrome c oxidase subunit I analysis. These results enable dietary preference prediction of the paralarvae. B) Stable isotope analysis of amino acids. The trophic position (TP) was estimated at 3.0 for paralarvae, regardless of ML, indicating a carnivorous trophic tendency, with herbivores being the primary prey.
Here, we provide three explanations for the results of this study. The most direct interpretation is that paralarvae (TP = 3.0) feed strictly on living herbivorous organisms (TP = 2.0). In a previous study, the estimated TP of nine fish (e.g., Pseudopleuronectes yokohamae) was 3.0 ± 0.1, which were secondary consumers, based on the same isotope technique [20]. The present data (TP = 3.0) clearly followed those of that previous report, suggesting that the paralarvae fed on herbivores. In the same report, the TP of adult T. pacificus was estimated to be 3.6, indicating that the ontogenetic dietary shift of TP occurred between paralarvae and adult [20]. In several papers, the apparent TP of detritus (marine snow or POM = total phytoplankton + decomposers) was estimated to be 1.4–1.5 [21–23]. If the diet of squid paralarvae is detritus, the TP of paralarvae should be evaluated as 2.4–2.5. For instance, in a study of eel larvae in the western Pacific [21], the apparent TP of the potential diet (e.g., marine snow) in the western Pacific was 1.4, based on the TP estimated for eel larvae (TP = 2.4) in the food web. The present results (TP = 3.0) were 0.6 unit higher than that of eel larvae. We emphasize here that the difference in the TP between the paralarvae and eel larvae (i.e., 0.6) is significant and quite robust, because the biomass at the basis of food web pyramid (TP = 1.0–2.0) is considerably large compared to the biomass of high TP organisms (TP > 3.0) and, because these numbers are taken as a weighted average, unlikely to fluctuate [23].
This explanation requires that the prey be alive. For organisms listed from metagenomic analyses, the life cycle, size, and mobility need to be considered. Discosea is typically a heterotrophic plankton which was universally observed in this study. In recent years, it has become evident that protozoans play significant roles as herbivores in ecosystems [24]. Most of them are free-living in soil, freshwater, on plants, and in brackish or sea water, and they basically play an important role as primary grazers on bacteria and are important for nutrient cycling as they are the prey of larger animals [25]. There is no disadvantage of paralarval feeding on Discosea, either in terms of low motility or size (15–36 μm in diameter). Our results are consistent with this finding. Notably, OTU for C. arabianum disappeared in Group C, suggesting an ontogenetic shift in juveniles of the East China Sea.
Although it is difficult to think that the paralarvae graze on adult living animals, as listed from the metagenome analysis, such as Arthropoda, it is still possible that the paralarvae diet includes the larval planktonic phases, including eggs, of these animals. Porifera sponge are sessile organisms but their larvae are planktonic amphiblastula [26]. Adult nemertea may attain a length of more than 1 m and a width of several centimeters; however, the size of their planktonic larvae is only several micrometers long [27]. Sporadic arthropod data were mainly ascribed to insects and arachnids. These may look terrestrial, but the midge (Harnischia) is known to be highly adaptable and lives in the sea [28]. Adult chironomids have degenerated mouths and do not feed, indicating that their trophic level does not shift between the juvenile and adult phases. The herbivorous juvenile (red worms) are an important food resource for aquatic organisms, especially fish [29]. It is possible that chironomid larvae or adults are preyed upon by squid larvae. It is reasonable to assume that the golden algae Synurophyceae (Poterioochromonas malhamensis: Samples 6, 9, 11, and 14) and Chrysophyceae (Pedospumella sinomuralis: Sample 9) remained in the digestive glands of the herbivores mentioned above.
About half of the paralarvae cecum sacs were empty (Fig 2B), suggesting that they were hungry and may eat what was available on the spot ad hoc. We should consider the basic ecology of the paralarvae. Previous experiments in aquaria have reported that the paralarvae repeatedly rise and sink and spin on the surface just after hatching. The swimming is basically vertical and its speed is 0.43 ± 0.15 cm/s [30,31]. In addition, T. pacificus paralarva is rhynchoteuthion, which is characterized by the fusion of both tentacles that split into two independent ones for prey capture in adult [32,33]. All samples in this study were confirmed to have fused tentacles by microscopic observation, suggesting that they cannot utilize the tentacles freely, like adults. The function of fused tentacles in prey capture is a key point. Thus, the challenge for the first explanation is to determine how paralarvae capture living zooplankton in motion. The size of the prey organism must also be considered.
The second explanation is that the paralarvae feed on detritus derived from carcass animals detected in metagenome analysis (Arthropoda, Nemertea, Porifera, Mollusca, etc.) and decomposers (Burkholderiales, Xanthomonadales, Ascomycota, and Basidiomycota) detected in the metagenome analysis. In 16S rRNA analysis in the present study, Burkholderiales (beta-proteobacteria) and Xanthomonadales (gamma-proteobacteria) accounted for approximately 86% of the OTUs. The Burkholderiaceae family includes multiple genera that have been reported to degrade aromatic compounds in soil and water [34]. Xanthomonadaceae are hydrocarbon degraders [35]. In addition, Ascomycota and Basidiomycota recognized in the COI analysis are two major phyla in which many species have been identified in the ocean as marine yeasts. In terrestrial environments, Ascomycetes and Basidiomycetes are the most abundant fungi in soils and contribute jointly or sequentially to plant residue decomposition [36,37]. It is possible that these fungi also work as decomposers of animal carcasses in marine environments, which are included in the diet of the paralarvae. This explanation supports the previous study of Ommastrephidae paralarvae, including Dosidicus gigas, Sthenoteuthis oualanensis, and so forth, such that the gut contents analyzed through metagenome were mainly composed of a mixture of continental and exclusively marine animal DNA in combination with single-cell organisms (cyanobacteria, diatoms, and ciliophorans) and other organisms often associated with organic material degradation [9]. Thus, we can ignore how the larvae capture zooplankton during their motion. However, robust data from the stable isotope (TP = 3.0 in paralarvae) could not easily support this explanation. If the paralarvae feed on detritus, the TP of the paralarvae should be approximately 2.4, which is similar to that of eel larvae (TP = 2.4).
The third explanation is the intermediate between the first and second explanations, but the most convincing interpretation. The paralarvae mostly fed on living organisms detected in the metagenome analysis and partly on detritus. This reasonably explains both the metagenomic and stable isotope results. In the first explanation, the question is whether larvae can catch living organisms despite their low swimming and prey-capturing abilities of their fused tentacles [32,33]. It is possible that the paralarvae catch Discosea and live animals if the prey organisms are very small and slow-moving, but they cannot explain all the feeding [25]. In second explanation, the challenge is the robustness of the stable isotope data (TP = 3.0) and the inconsistency of the idea (apparent TP of detrivores is 2.4) [21–23]. The glycolysis pathway is performed anaerobically, and the TCA cycle is carried out aerobically. Ala is metabolized (decomposed) by the former and the latter, whereas Glu is metabolized (decomposed) only by the latter. As the decomposition proceeds, the isotope ratios (δ15N) increase; therefore, organic matter decomposed in an oxygen-poor environment has the δ15N values of Ala higher than those of Glu. In the present study, such a trend was found in samples with ML of 1–2 and 3–4 mm, suggesting the contribution of non-oxidative corrosion. Based on this idea at this stage, the contribution of detritus was estimated to be 18%, because the difference in the δ15N value between Ala and Glu (i.e., 1.4‰) accounts for approximately 18% (=1.4‰/8.0‰ × 100) excess of the degradation of Ala compared to that of Glu in the diets of paralarvae [19,23]. We note that although the sample size appears small, the TP estimation based on the δ15N values of amino acids is independent of natural variation in the δ15N values of inorganic nitrogen (e.g., ammonia and nitrate) and dietary resources (e.g., phytoplankton). Moreover, the measurement error of the δ15N values is 0.4–0.7‰, accounting for 0.05–0.1 units of error in the TP estimate. Thus, in contrast to the traditional methods (e.g., isotope analysis of bulk tissues), the TP of organisms can be estimated within 0.1–0.2 units even when the sample size is limited [12].
O’Dor et al proposed the “suspension feeding” using paralarvae of Illex illecebrosus (Ommastrephidae) hatched after five days [38]. They reported that mucus on the body surface is transferred to mice by ciliary motion and cleaning behavior of the mantle, which may be a critical bridge between yolk absorption and the minimum development required for effective predation. Vidal & Haimovichi (1998) supported this, but mucus enriched with microorganisms may be important in the diet of the small rhynchoteuthion Illex argentinus [39]. To the best of our knowledge, these are the first landmarks for the feeding of Ommastrephidae.
We believe that the samples examined in this study were collected after the aforementioned event because microscopic observation did not show yolk or similar mucus. In addition, the present data did not consider the assimilation process of nutrition from cecum sacs, which contain a mixture of various organisms, although TP reflected these elements after assimilation. Further experiments must be conducted to fill these gaps in knowledge. We can prepare the paralarvae using artificial insemination or from the egg mass in an aquarium, which can be maintained 6–10 days after hatch just after yolk absorption [40–42]. Preparation of prey may be challenging; however, living planktonic organisms of various sizes and motilities can be captured by sampling. The composition of prey can be determined by metagenome analysis. Artificial detritus can be produced according to previous reports [43,44]. Feeding under different conditions and experiments based on them are possible under the controlled environment of a laboratory, and will be necessary in the future to validate the present explanation.
5. Conclusion
Metagenomic analysis of the digestive gland of wild T. pacific paralarvae revealed various organisms, including bacteria, animals, algae, and fungi. The stable isotope analysis of amino acids in the muscle estimated 3.0 for the trophic position of paralarvae, suggesting that the paralarvae fed mostly on herbivorous organisms and partly on detritus. To substantiate this theory, further studies involving feeding trials with larvae and other candidates in the laboratory are required.
Supporting information
S1 File. PCR conditions.
The reaction conditions, primer sequences, and PCR results corresponding to the 16SrRNA and COI sequences are summarized in the “16S” and ‘COI’ tabs within the Excel file, respectively. The relationship between sample numbers and fastq file names is summarized in the “measurement table for fastq” tab.
https://doi.org/10.1371/journal.pone.0340579.s001
(XLSX)
S2 File. 16S rRNA summary.
The relationship between sample numbers and the number of reads assigned to each classification item is summarized into tabs from kingdom level to OTU level. The Sheet1 tab displays the raw data shown in Fig. 3., indicating the percentage at the order level.
https://doi.org/10.1371/journal.pone.0340579.s002
(XLSX)
S3 File. COI summary.
The “OTU number” tab shows the relationship between sample numbers and the number of reads assigned to each classification from kingdom to OTU level. The “percentage” tab shows the percentage of the number of reads assigned to each classification.
https://doi.org/10.1371/journal.pone.0340579.s003
(XLSX)
S4 File. Amino acid.
The table shows the δ15N values (‰, vs Air) for all amino acids measured by stable isotope analysis for amino acids. Trophic position (TP) was calculated using the formula in the main text.
https://doi.org/10.1371/journal.pone.0340579.s004
(XLSX)
Acknowledgments
We thank Yasunori Sakurai of the Hakodate Cephalopod Research Center for providing useful suggestions at the beginning of the research. We thank Suguru Okamoto, Norio Yamashita, and Toshiki Kaga of the Fisheries Resources Institute, Japan Fisheries Research, and Education Agency for providing the samples. We appreciate the Division of Biological Research, Science Research Center, Kochi University, for the use of research instruments and technical assistance.
References
- 1. Gasalla MA, Rodrigues AR, Postuma FA. The trophic role of the squid Loligo plei as a keystone species in the South Brazil Bight ecosystem. ICES Journal of Marine Science. 2010;67(7):1413–24.
- 2. Villanueva R, Perricone V, Fiorito G. Cephalopods as predators: A short journey among behavioral flexibilities, adaptions, and feeding habits. Front Physiol. 2017;8:598. pmid:28861006
- 3. Barrett CJ, Bensbai J, Broadhurst MK, Bustamante P, Clark R, Cooke GM, et al. Cuttlefish conservation: a global review of methods to ameliorate unwanted fishing mortality and other anthropogenic threats to sustainability. ICES Journal of Marine Science. 2022;79(10):2579–96.
- 4. Young RE, Harman RF. Larva’, ‘paralarva’ and ‘subadult’ in cephalopod terminology. Malacologia. 1988;29(1):201–7.
- 5. Vidal EAG, Shea EK. Cephalopod ontogeny and life cycle patterns. Front Mar Sci. 2023;10.
- 6.
Sakurai Y, Yamashita N, Yamamoto J, Uchikawa K, Takahara H. Todarodes pacificus, Japanese common squid. In: Rosa R, G P, O’Dor R, editors. Advances in Squid Biology, Ecology and Fisheries Part II. Nova Science Publishers, Inc. 2013:249–71.
- 7. Shin D, Park TH, Lee C-I, Jo JH, Choi CG, Kang S, et al. Feeding Ecology of Common Squid Todarodes pacificus in the South Sea of Korea Determined through Stable Isotope and Stomach Content Analyses. Water. 2022;14(19):3159.
- 8. Watanabe K, Sakurai Y, Segawa S, Okutani T. Development of the ommastrephid squid Todarodes pacificus, from fertilized egg to rhynchoteuthion paralarvae. Amer Malacological Bull. 1996;13:73–88.
- 9. Fernández-Álvarez FÁ, Machordom A, García-Jiménez R, Salinas-Zavala CA, Villanueva R. Predatory flying squids are detritivores during their early planktonic life. Sci Rep. 2018;8(1):3440. pmid:29467371
- 10. Novotny A, Zamora-Terol S, Winder M. DNA metabarcoding reveals trophic niche diversity of micro and mesozooplankton species. Proc Biol Sci. 2021;288(1953):20210908. pmid:34130506
- 11. Chikaraishi Y, Kashiyama Y, Ogawa N, Kitazato H, Ohkouchi N. Metabolic control of nitrogen isotope composition of amino acids in macroalgae and gastropods: implications for aquatic food web studies. Mar Ecol Prog Ser. 2007;342:85–90.
- 12. Chikaraishi Y, Steffan SA, Ogawa NO, Ishikawa NF, Sasaki Y, Tsuchiya M, et al. High-resolution food webs based on nitrogen isotopic composition of amino acids. Ecol Evol. 2014;4(12):2423–49. pmid:25360278
- 13. Nielsen JM, Popp BN, Winder M. Meta-analysis of amino acid stable nitrogen isotope ratios for estimating trophic position in marine organisms. Oecologia. 2015;178(3):631–42. pmid:25843809
- 14. McMahon KW, McCarthy MD. Embracing variability in amino acid δ15N fractionation: mechanisms, implications, and applications for trophic ecology. Ecosphere. 2016;7(12).
- 15. Vane K, Cobain MRD, Larsen T. The power and pitfalls of amino acid carbon stable isotopes for tracing origin and use of basal resources in food webs. Ecological Monographs. 2025;95(1).
- 16. Fiorito G, Affuso A, Basil J, Cole A, de Girolamo P, D’Angelo L, et al. Guidelines for the Care and Welfare of Cephalopods in Research -A consensus based on an initiative by CephRes, FELASA and the Boyd Group. Lab Anim. 2015;49(2 Suppl):1–90. pmid:26354955
- 17. Chikaraishi Y, Ogawa NO, Kashiyama Y, Takano Y, Suga H, Tomitani A, et al. Determination of aquatic food‐web structure based on compound‐specific nitrogen isotopic composition of amino acids. Limnology & Ocean Methods. 2009;7(11):740–50.
- 18. Yamamoto J, Shimura T, Uji R, Masuda S, Watanabe S, Sakurai Y. Vertical distribution of Todarodes pacificus (Cephalopoda: Ommastrephidae) paralarvae near the Oki Islands, southwestern Sea of Japan. Mar Biol. 2007;153(1):7–13.
- 19. Takizawa Y, Choi H, Shibuya M, Park N, Dharampal PS, Steffan SA, et al. Fractionation of nitrogen isotopes in the enzymatic deamination of alanine: an insight into the quantitative evaluation of anaerobic metabolism in ecosystems. Prog Earth Planet Sci. 2025;12(1).
- 20. Xing D, Choi B, Takizawa Y, Fan R, Sugaya S, Tsuchiya M, et al. Trophic hierarchy of coastal marine fish communities viewed via compound-specific isotope analysis of amino acids. Mar Ecol Prog Ser. 2020;652:137–44.
- 21. Miller MJ, Chikaraishi Y, Ogawa NO, Yamada Y, Tsukamoto K, Ohkouchi N. A low trophic position of Japanese eel larvae indicates feeding on marine snow. Biol Lett. 2013;9(1):20120826. pmid:23134783
- 22. Batista FC, Ravelo AC, Crusius J, Casso MA, McCarthy MD. Compound specific amino acid δ 15 N in marine sediments: A new approach for studies of the marine nitrogen cycle. Geochimica et Cosmochimica Acta. 2014;142:553–69.
- 23. Goto AS, Miura K, Korenaga T, Hasegawa T, Ohkouchi N, Chikaraishi Y. Fractionation of stable nitrogen isotopes (<sup>15</sup>N/<sup>14</sup>N) during enzymatic deamination of glutamic acid: Implications for mass and energy transfers in the biosphere. Geochem J. 2018;52(3):273–80.
- 24. Strom SL, Macri EL, Olson MB. Microzooplankton grazing in the coastal Gulf of Alaska: Variations in top‐down control of phytoplankton. Limnology & Oceanography. 2007;52(4):1480–94.
- 25. Schilde C. Amoebozoa. Reference Module in Life Sciences. 2019.
- 26. Maldonado M. The ecology of the sponge larva. Can J Zool. 2006;84(2):175–94.
- 27. Henry JJ, Martindale MQ. Conservation of the spiralian developmental program: cell lineage of the nemertean, Cerebratulus lacteus. Dev Biol. 1998;201(2):253–69. pmid:9740663
- 28. Tang H, Cheng Q, Krosch MN, Cranston PS. Maritime midge radiations in the Pacific Ocean (Diptera: Chironomidae). Systematic Entomology. 2022;48(1):111–26.
- 29.
Armitage PDC, Pinder LCV. The Chironomidae: Biology and Ecology of Non-Biting Midges. London: Chapman & Hall. 1995.
- 30. Bower JR, Sakurai Y. Laboratory observations on Todarodes pacificus (Cephalopoda: Ommastrephidae) egg masses. American Malacological Bulletin. 1996;13(1–2):65–71.
- 31. Jun Yamamoto1 SM, Shin’ichi F, Yasunori S. Effect of temperature on swimming behavior of paralarvae of the Japanese common squid Todarodes pacificus (in Japanese). Bull Jpn Soc Fish Oceanogr. 2012;76(1):18–23.
- 32. Shigeno S, Kidokoro H, Goto T, Tsuchiya K, Segawa S. Early Ontogeny of the Japanese Common Squid Todarodes pacificus (Cephalopoda, Ommastrephidae) with Special Reference to its Characteristic Morphology and Ecological Significance. Zoological Science. 2001;18(7):1011–26.
- 33. Shea EK. Ontogeny of the fused tentacles in three species of ommastrephid squids (Cephalopoda, Ommastrephidae). Invertebrate Biology. 2005;124(1):25–38.
- 34. Morya R, Salvachúa D, Thakur IS. Burkholderia: An Untapped but Promising Bacterial Genus for the Conversion of Aromatic Compounds. Trends Biotechnol. 2020;38(9):963–75. pmid:32818444
- 35.
Gutierrez T. Aerobic hydrocarbon-degrading gammaproteobacteria: Xanthomonadales. 2017.
- 36. Grinhut T, Hadar Y, Chen Y. Degradation and transformation of humic substances by saprotrophic fungi: processes and mechanisms. Fungal Biology Reviews. 2007;21(4):179–89.
- 37. Fukasawa Y. Ecological impacts of fungal wood decay types: A review of current knowledge and future research directions. Ecological Research. 2021;36(6):910–31.
- 38. O’Dor RK, Helm P, Balch N. Can rhynchoteuthion suspension feed?. Vie Milieu. 1985;35:264–71.
- 39. Vidal EAG, Haimovici M. Feeding and the possible role of the proboscis and mucus cover in the ingestion of microorganisms by Rhynchoteuthion paralarvae (Cephalopoda: Ommastrephidae). Bulletin of Marine Science. 1998;63(2):305–16.
- 40. Adachi K, Yabumoto M, Yoo H, Morioka K, Ikeda Y, Ueta Y, et al. Salt soluble without jelly-like component from the oviducal gland induces chorionic expansion in the ova of the Japanese common squid Todarodes pacificus. Invertebrate Reproduction & Development. 2016;60(4):281–9.
- 41. Puneeta P, Vijai D, Yamamoto J, Adachi K, Kato Y, Sakurai Y. Structure and properties of the egg mass of the ommastrephid squid Todarodes pacificus. PLoS One. 2017;12(8):e0182261. pmid:28767686
- 42. Yamamoto J, Adachi K, Bower JR, Matsui H, Nakaya M, Ohtani R, et al. Close-up observations on the spawning behavior of a captive Japanese flying squid (Todarodes pacificus). Sci Rep. 2019;9(1):19739. pmid:31875026
- 43. Shanks AL, Edmondson EW. Laboratory-made artificial marine snow: a biological model of the real thing. Mar Biol. 1989;101(4):463–70.
- 44. Costa V, Chemello R, Iaciofano D, Lo Brutto S, Rossi F. Seagrass detritus as marine macroinvertebrates attractor. Ninth International Symposium “Monitoring of Mediterranean Coastal Areas: Problems and Measurement Techniques”. Firenze University Press. 2022:619–26.