Administration of 1-kestose to beluga reduces intestinal Turicibacter and collagenase gene levels, and blood creatinine levels

Abstract

In recent years, an increased emphasis on enhancing the care and health management of captive marine mammals has been observed. Belugas (White Whale, Delphinapterus leucas), belonging to the family Monodontidae, are of considerable importance and often the centerpiece of aquarium collections worldwide. This study aimed to investigate the effects of the administration of prebiotics on the gut microbiota and overall health of the beluga. Prebiotic 1-kestose, a fructooligosaccharide comprising sucrose and fructose, was administered to three belugas, alongside their regular vitamin supplements for a duration of 8 weeks. 16S rRNA gene amplicon sequencing of intestinal DNA revealed that the relative abundance of the genus Turicibacter, a potentially pathogenic bacteria, significantly reduced after 1-kestose administration when compared to that at baseline (P=0.050). In addition, a quantitative PCR analysis revealed that the levels of collagenase gene, a putative virulence factor gene of Turicibacter, significantly reduced after 1-kestose administration (P=0.050). Blood creatinine levels that were initially above the normal value also reduced after 1-kestose administration (P=0.023). Therefore, this study demonstrated the potential of 1-kestose to improve the health and welfare of aquarium belugas.

INTRODUCTION

In recent years, the welfare of marine mammals has increasingly come into focus, with enhanced efforts directed towards the care and health management of these captive animals [3]. Among the marine species, belugas (White Whale, Delphinapterus leucas) from the Monodontidae family are particularly significant, often serving as the main attractions in aquariums globally. Belugas are the most populous toothed whales in the Arctic regions and exhibit an omnivorous diet [1]. They consume a wide array of food sources including Arctic cod (Boreogadus saida), redfish (Sebastes marinus), halibut (Reinhardtius hippoglossoides), shrimp (Pandalus borealis), saffron cod (Eleginus gracilis), rainbow smelt (Osmerus mordax), and various species of Pacific salmon (Oncorhynchus spp.) [1].

The gut microbiota of animals is increasingly acknowledged for its substantial impact on overall health, encompassing digestion, immunity, and disease resistance [16]. Despite this, research on the gut microbiota of belugas and its potential manipulation for health enhancement remains relatively sparse. A study conducted on two belugas housed at the Marine and Waterpark of the Atlantis Hotel in Sanya, China, provided insights into their gut microbiota. The analysis revealed that the predominant bacterial families present were Lachnospiraceae, Ruminococcaceae, and Peptostreptococcaceae [1].

The burgeoning field of prebiotics, which entails the administration of specific beneficial substances, has shown promise in enhancing the gut health of various animals [18, 20]. Notably, the prebiotic 1-kestose, a fructooligosaccharide with a degree of polymerization of three, has proven effective in boosting the populations of beneficial microbes such as Bifidobacteria, Lactobacillales, and a butyrate-producing Clostridium cluster XIVa both in vitro and in vivo [8, 13, 18, 19, 28]. For instance, in Japanese eels (Anguilla japonica), administering a combination of 1-kestose with heat-killed Lactiplantibacillus plantarum FM8 significantly enhances the relative abundance of Romboutsia, increases intestinal acetate concentrations, and improves feed efficiency [10]. Similarly, in Magellanic penguins (Spheniscus magellanicus), this treatment notably boosts the activity of gut Lactobacillaceae, reduces the levels of Clostridium perfringens and its alpha-toxin-encoding gene, and diminishes the inflammatory response [9].

This study presents a novel administrative trial exploring the applications of 1-kestose to enhance the health and welfare of belugas in an aquarium. This research may benefit these iconic marine mammals and also expand our understanding of the potential use of 1-kestose in marine mammal management.

MATERIALS AND METHODS

Ethics statement

This study, involving healthy belugas at the Port of Nagoya Public Aquarium, proceeded without an Institutional Animal Care and Use Committee (IACUC) due to the facility’s lack of such a committee. Nonetheless, we strictly adhered to the Japanese Association of Zoos and Aquariums (JAZA) guidelines to ensure ethical treatment and welfare of the animals. Additionally, the study was minimally invasive and did not affect the welfare of belugas, maintaining our commitment to uphold the highest standards of animal care and ethical research practices. It is also important to note that the samples analyzed in this study, including feces and blood, were collected as part of the animals’ regular health examinations and not specifically for the purposes of this research, further minimizing any additional stress or impact on the animals involved.

Study design

This study was conducted at the Port of Nagoya Public Aquarium and involved three beluga whales (N=3). The baseline characteristics of these belugas are provided in Table 1. Belugas No. 1 and No. 2, both born in captivity, are aged 15 and 10 years, respectively. Beluga No. 3, who is wild-born, is estimated to be 15 years old. The research aimed to assess the impact of an 8-week prebiotic administration on the gut microbiota and plasma biochemistry, using pre-treatment measurements as a baseline.

Table 1. Baseline characteristics of belugas and their daily food and calorie intake during the 8-week study period.

	No.1	No.2	No.3
Sex	Female	Male	Male
Age	15	10	15 (estimated)
Body Weight (kg)	557	505	1,223
Food intake (kg/day)	9.1–14.3	8.7–13.3	12.0–19.9
Calorie intake (kcal/day)	17,419–26,179	16,453–24,587	16,906–36,187

Breeding conditions, diet, and administration

The belugas are accommodated in an indoor facility featuring three interconnected elliptical pools, with a combined water volume of 2,078 cubic meters. Water and air temperatures are maintained at 13.0–18.0°C and 9.5–30.0°C, respectively. To accommodate their breeding cycle, the daylight duration varies throughout the year, ranging from approximately 10 to 17.5 hr. The diet of the belugas included Atka mackerel, herring, sand lance, and rainbow smelt, with daily individual portions ranging from 8.7 to 19.9 kg (16,453 to 36,187 kcal), as shown in Table 1. These portions were distributed across three to four feeding sessions per day. Additionally, one vitamin supplement tablet per 2.5 kg of food was given for health maintenance (Mazuri^® Marine Mammal Supplement with Vitamins A & C, Mazuri, IN, USA), following a dosing protocol that was uniformly applied to other cetaceans at the aquarium. During the 8-week study period, each beluga was given 25 capsules of 1-kestose (iKes75; WELLNEO SUGAR, Tokyo, Japan; 75% purity, 400 mg per tablet) twice daily.

Sample collection

For fecal sample collection, a Terumo syringe catheter tip type (50 mL, SS-50CZ, Terumo Corp., Tokyo, Japan) was used with a two-hole Safed^® Neraton catheter (SF-ND2010, 40 cm, Fr. 20, 6.7 mm, Terumo Corp.). After observing the fecal discharge, the catheter was inserted 40 cm into the rectum from the anus, and gentle negative pressure was applied while withdrawing to collect fecal samples. Samples were taken prior to the commencement of the trial and at the conclusion of week 8, then immediately stored at −20°C for subsequent analysis.

16S rRNA gene sequencing

Genomic DNA was extracted from fecal samples using the QIAamp DNA Stool Mini Kit (QIAGEN, Hilden, Germany). The process began with thawing the frozen fecal samples on ice. Each sample, measuring approximately 100 mg, was then suspended in 1 mL of InhibitEX buffer provided in the kit. Disruption of the samples was carried out using zirconia beads on a FastPrep FP100A instrument (MP Biomedicals, Santa Ana, CA, USA). Subsequent DNA extraction from the bead-treated suspensions was performed using the Magtration System 12GC with GC series MagDEA DNA 200 (Precision System Science, Matsudo, Japan), following the manufacturer’s guidelines.

For amplifying the V3–V4 region of the prokaryotic 16S rRNA genes, primers Pro341F and Pro805R were used as described in previous literature [26]. The sequencing was conducted at Bioengineering Lab Co., Ltd. (Sagamihara, Japan). We employed paired-end sequencing (2 × 300 bp) on the Illumina MiSeq platform using the MiSeq Reagent Kit ver. 3 (Illumina, San Diego, CA, USA).

Bioinformatics analysis

The bioinformatics analysis was conducted using the EzBioCloud webpage (https://www.ezbiocloud.net/) [29]. For this, the paired-end compressed FASTQ 16S sequencing files were uploaded and processed using the default settings, and the results were visualized. Taxonomic classification was carried out using the most recent EzBioCloud database (version 20230823) as a reference. To display the relative abundance of microbial genera in the samples, a genus-level bar plot was created.

Diversity within each sample was evaluated using alpha diversity indices, specifically the Shannon and Chao1 indices, calculated at the genus level. Beta diversity analysis, which measures the dissimilarity between microbial communities from different samples, was also conducted at the genus level. This analysis incorporated principal coordinate analysis (PCoA) alongside permutational multivariate analysis of variance (PERMANOVA) based on Jensen-Shannon distance matrices to graphically represent the findings.

Lastly, to identify significantly different microbial taxa between sample groups at the genus level, linear discriminant analysis effect size (LefSe) was utilized. This analysis applied cut-off values of P<0.05 and LDA score >4.0 to ensure statistical significance and relevance.

Quantitative (q) PCR analysis of collagenase gene

In this study, we developed a method using quantitative PCR (qPCR) to measure the levels of the collagenase gene in Turicibacter. A primer set was specifically designed to amplify a fragment of the collagenase gene, which is common among various Turicibacter strains as recorded in GenBank (https://www.ncbi.nlm.nih.gov/genbank/, accessed on 7 November 2023, Supplementary Fig. 1). The primer sequences were derived from a conserved region within the collagenase gene of Turicibacter. Specifically, for the Turicibacter sanguinis strain MOL361, the forward primer (5ʹ-CGTGAAACAACAAAAGAAGAAAT-3ʹ) is designed to anneal between bases 472 and 494, and the reverse primer (5ʹ-TTTGACTAAATCATATTCCCAACG-3ʹ) between bases 637 and 660.

The qPCR analysis of both the collagenase and the 16S rRNA genes in intestinal DNA solutions was conducted using the QuantStudio 3 (Thermo Fisher Scientific, Waltham, MA, USA). For each DNA template, the reaction mixture was prepared with PowerTrack SYBR Green Master Mix (Thermo Fisher Scientific), following the manufacturer’s guidelines. The specific primer set mentioned was employed to amplify the collagenase genes, and primers F_Bact 1369 and R_Prok1492 were used for the amplification of 16S rRNA genes, as previously described [11]. The PCR cycle included an initial denaturation at 95°C for 2 min, followed by 40 cycles of denaturation at 95°C for 10 sec, annealing at 60°C for 15 sec, extension at 72°C for 15 sec, and a final extension at 72°C for 1 min. The collagenase gene levels were quantified by multiplying the number of collagenase gene copies by eight and dividing by the number of total 16S rRNA gene copies, reflecting the eight copies of the 16S rRNA gene found in the Turicibacter sanguinis strain MOL361 according to GenBank accession number CP053187.1.

Plasma biochemistry

For this component of the study, a total of 10 mL of blood was extracted from the superficial fluke vein of each subject using a 21-gauge needle. The blood was drawn into two 5-mL syringes (Ds 5mL, Nipro, Settsu, Japan). Immediately after collection, 3 mL of the drawn blood was transferred into two 1.5-mL Fuji heparin tubes (Tube 1.5 HE, Fujifilm, Tokyo, Japan) for stabilization. These tubes were then rapidly centrifuged at speeds ranging from 2,700 to 4,500 × g, adjusted based on the number of tubes and the type of rotor in use, for a duration of 5 min. Following centrifugation, the plasma was separated and preserved for subsequent biochemical analysis. The analysis was performed using an automatic biochemical analyzer (FDC 7000V, Fujifilm Medical, Tokyo, Japan).

Determination of short-chain fatty acids (SCFA) and lactate concentration

The carboxyl groups SCFAs (acetate, propionate, and butyrate) in the intestinal contents were labeled with 2-nitrophenyl hydrazide using a Short- and Long-Chain Fatty Acid Analysis Kit (YMC, Kyoto, Japan) with a modified manufacturer’s protocol, and analyzed using high-performance liquid chromatography (HPLC), as described previously [25]. Briefly, 20–50 mg of the intestinal contents was weighed into a 1.5-mL Eppendorf tube. Approximately 1 mL of phosphate-buffered saline (PBS) was added to the tube on ice, and the tube was vortexed for 30 sec. The tube was centrifuged at 15,000 rpm at 4°C for 2 min. The supernatant was used for labeling. Approximately 50 µL of supernatant was mixed with 50 µL of PBS, 200 µL of 2 mM caproic acid (Fujifilm-Wako, Osaka, Japan) as an internal standard, 200 µL of 20 mM 2-nitrophenylhydrazin (in water), and 200 µL of 0.25 M N-(3-dimethylaminopropyl)-N-ethylcarbodiimide hydrochloride HCl (dissolved in ethanol, with an equal volume of 3% pyridine in ethanol). The mixture was heated at 60°C for 20 min. Thereafter, 200 µL of 15% (w/v) potassium hydroxide was added, and the mixture was incubated at 60°C for 15 min. The reaction mixture was added to 4 mL of a 3.8:0.4 (v:v) mixture of 0.03 M PBS (pH 6.4):0.5 M hydrochloric acid and filtered through a 0.45-μm filter (PTFE, Puradisc™ 13 mm Syringe Filters, Whatman, Kent, England). The SCFA derivatives were extracted using diethyl ether (5 mL). The diethyl ether layer was evaporated to dryness under a nitrogen stream at 20–25° C. The residue was dissolved in 200 µL of methanol, and 10 µL of each sample was analyzed using HPLC to quantify the major SCFAs. Chromatographic separation was performed using a Nexera XL series HPLC instrument (Shimadzu, Kyoto, Japan) and YMC-Pack FA HPLC column (6.0 mm I.D. × 250 mm; YMC). The solvent system used for the elution from the YMC-Pack FA column comprised 30:16:54 (v/v) acetonitrile (Fujifilm-Wako), methyl alcohol (Fujifilm Wako), and ultrapure water. The elution was performed for 20 min with a 1.2 mL/min flow rate, and the column temperature was maintained at 50°C. The eluents were monitored at 400 nm to detect derivatized 2-nitrophenylhydrazin.

Statistical analysis

Statistical analyses of the microflora and qPCR results were performed using the Mann–Whitney test, and analysis of the creatinine, calcium, SCFA, and body weight data was performed using the Wilcoxon matched-pairs signed-rank test in GraphPad Prism v.9.5.1 (GraphPad Software, San Diego, CA, USA). Spearman’s rank correlation coefficient (R) and statistical significance were calculated using GraphPad Prism software. Statistical significance was set at P<0.05.

RESULTS

Impact of 1-kestose on performance of the study subjects

During the experimental phase, as well as the periods preceding and following the treatment, the belugas exhibited no undesirable clinical symptoms like diarrhea or constipation. Furthermore, there were no observed negative effects related to the consumption of feed supplemented with 1-kestose. Changes in body weight for belugas No. 1, 2, and 3 before and after the 1-kestose intervention were as follows: from 557 kg to 561 kg, from 505 kg to 535 kg, and from 1,223 kg to 1,160 kg, respectively. These changes were not statistically significant (P=0.7604).

Intestinal microbiota analysis

The sequencing of the 16S rRNA gene provided 7,096, 2,504, and 5,338 valid reads from intestinal DNA at baseline, and 2,133, 2,116, and 4,980 valid reads at week 8 post-chimeric sequence removal. The 16S rRNA gene amplicon sequencing datasets were deposited in the NCBI Sequence Read Archive (accession number PRJNA1032236). Comparisons of alpha diversity indices at baseline and week 8 revealed no significant differences in diversity (Shannon index) or richness (Chao1 index), with P-values of 0.275 and 0.513, respectively. Beta diversity analysis, represented by PCoA plots based on an unweighted metric (Fig. 1), showed grouping of bacterial communities but no significant differences in gut microbiota composition between the baseline and week 8 (P=0.101), as confirmed by PERMANOVA.

Fig. 1.

Beta diversity plots before (Baseline) and after 1-kestose administration (Week 8) to belugas.

Analysis of the taxonomic composition and relative abundance of the microbiota at baseline and week 8 at the genus level revealed notable changes, as illustrated in Fig. 2. Initially, Turicibacter was the predominant genus in the two aquarium-born belugas, representing over 55% of the microbiota. In the wild-born beluga, Turicibacter accounted for 21%, with Romboutsia being more prevalent at approximately 29%. After 8 weeks of 1-kestose administration, there was a significant reduction in the relative abundance of Turicibacter. Mycobacterium became the most abundant genus among the aquarium-born belugas, exceeding 69%, and also appeared in the wild-born beluga with 12% occupancy, although Romboutsia remained the dominant genus at 83% occupancy in the latter. LEfSe analysis identified statistically significant differences in taxa between baseline and week 8 (Table 2). Notably, the relative abundances of Turicibacter, Bacteroides, and Actinobacillus_g1 significantly decreased by week 8 compared to baseline (P=0.050). Specifically, Turicibacter showed a dramatic reduction from 47.2% to 0.659% (P=0.050, Fig. 3A).

Fig. 2.

Relative abundances of the bacterial genera comprising the intestinal microbiota of belugas at baseline and week 8.

Table 2. Results of the linear discriminant analysis (LDA) effect size analysis at the genus level for distinguishing the week 8 group (Week 8) from the baseline group (Baseline), using criteria of P-value <0.05 and |LDA effect size| >4.0.

Taxon name	LDA effect size	P-value	Taxonomic relative abundance
			Baseline	Week 8
Turicibacter	5.36	0.0495	47.2	0.659
Bacteroides	4.55	0.0495	5.39	0.0583
Actinobacillus_g1	4.21	0.0495	3.72	0.155

Fig. 3.

Comparison of the relative abundance of Turicibacter (A) and collagenase gene levels (B) at baseline and week 8. Correlation between the relative abundance of Turicibacter and collagenase gene levels (C) Plots represent individual beluga, and bars represent mean ± SEM. *P<0.05.

Quantitative analysis of the collagenase gene using qPCR

Bacterial collagenases are recognized as critical virulence factors [7]. Analysis of the genomic data from the type strain of Turicibacter sanguinis strain MOL361 (GenBank: CP053187.1) identified a gene encoding a protein with annotations suggesting a collagenase-like protease (locus_tag: HLK68_13020). A primer set was developed based on the consensus sequence of collagenase gene homologs. This primer set was used in qPCR to quantify the levels of bacterial collagenase genes in the intestinal DNA of belugas. By week 8, there was a significant reduction in collagenase gene levels compared to the baseline (Fig. 3B). Additionally, the levels of the collagenase genes showed a strong correlation with the relative abundance of Turicibacter in the DNA samples (R²=0.887, P=0.005, Fig. 3C).

Creatinine and calcium content in blood

Plasma biochemical parameters were determined before and after 1-kestose administration (Supplementary Table 1). The administration resulted in a significant decrease in the total levels of creatinine and calcium (both P=0.023) when compared to the baseline (Fig. 4).

Fig. 4.

Blood creatinine (A) and calcium (B) levels after 8 weeks of 1-kestose administration (Week 8) compared to baseline. Blue, green, and red circles represent belugas No. 1, 2, and 3, respectively. *P<0.05.

SCFA content in the intestinal contents

Commensal bacteria metabolize carbohydrates to produce SCFAs, such as acetate, propionate, and butyrate, which are crucial for gut metabolic regulation [15]. Consequently, the concentrations of these SCFAs were measured. There were no significant differences in the SCFA concentrations at week 8 compared to those at baseline (Fig. 5). However, it is noteworthy that in the wild-born beluga, the levels of each SCFA showed an increase at week 8 relative to the baseline measurements.

Fig. 5.

Acetate (A), propionate (B), and butyrate (C) levels after 8 weeks of 1-kestose administration (Week 8) compared to baseline. Blue, green, and red circles represent belugas No. 1, 2, and 3, respectively. ns=not significant.

DISCUSSION

This study aimed to evaluate the effects of 1-kestose administration on the gut microbiome and overall health of belugas. We selected 1-kestose as a prebiotic based on its proven efficacy in enhancing beneficial bacterial populations, as outlined in the introduction. Furthermore, its widespread availability and practical ease of use render it an excellent choice for potential future applications.

Analysis of the gut microbiota revealed no significant differences in alpha or beta diversity, likely due to the inherently small sample size, as the number of belugas available in aquarium settings is limited. However, the administration of 1-kestose significantly reduced the relative abundance of Turicibacter, a genus known for its pathogenic properties and association with host inflammation and dysbiosis [5, 6, 21]. Turicibacter bacteria produce collagenases, which are metalloproteinases that break down the extracellular matrix of animal cells and are closely linked to bacterial pathogenesis [7]. Consequently, we quantified the levels of the collagenase gene in Turicibacter. The results showed that 1-kestose administration led to a significant decrease in both collagenase gene levels and Turicibacter abundance, which may positively impact the health of belugas.

Blood tests were conducted to assess the impact of 1-kestose administration on the health of the belugas. Following the administration of 1-kestose, there was a significant reduction in the total levels of creatinine and calcium compared to baseline values. Typically, the normal blood creatinine level in belugas ranges from 1.2–1.6 [17], but the initial creatinine levels for the belugas in this study were considerably higher, ranging from 1.90–2.57. Elevated creatinine is associated with dehydration and renal disease [27], and an increase above normal levels is generally considered detrimental to beluga health. This indicates that 1-kestose administration may aid in reducing high blood creatinine levels in belugas, thereby supporting their overall health. Additionally, the reduction in calcium levels by 1-kestose administration, which were near the upper end of the normal range (9.1–10.6) for belugas, may also be favorable to their health.

Despite the limitations imposed by a small sample size, which restricts the scope of our conclusions, this case study provides two key insights: Firstly, there was a discernible difference in the gut microbiota between the aquarium-born belugas and their wild-born counterpart. The composition of the intestinal microbiota is shaped by various factors including diet, environmental conditions, and early life exposures, and tends to stabilize over time [22]. Consequently, it is expected that the microbiota of the belugas born in the aquarium, which were not exposed to the diverse marine diet of Arctic cod and other sea foods like the wild-born beluga, would differ notably. In our previous research, the administration of 1-kestose combined with heat-killed L. plantarum FM8 significantly enhanced the gut microbiota of young penguins compared to adult penguins [9]. In humans, while the intestinal microbiota of adults remains stable [23], it undergoes significant changes during infancy [14]. These observations imply that aquarium-born belugas might sustain a healthier gut environment and higher levels of SCFAs if 1-kestose is administered from an early age.

The second insight from this study is the significant shift in the dominant genera within the gut microbiota following 1-kestose administration. In aquarium-born belugas, Turicibacter, initially the most abundant genus, was largely replaced by Mycobacterium. While most Mycobacterium species are non-pathogenic heterotrophic bacteria, some have evolved to become pathogenic in animals [4]. Mycobacterium marinum, a pathogen found in both freshwater and marine environments, is known to cause mycobacteriosis in belugas [2]. However, the absence of mycobacteriosis symptoms during and surrounding the study period suggests that the strains of Mycobacterium that increased due to 1-kestose administration were likely non-pathogenic.

Conversely, in the wild-born beluga, Turicibacter was replaced by Romboutsia after 1-kestose administration. Previous studies have shown that the administration of 1-kestose with heat-killed L. plantarum FM8 significantly boosts the relative abundance of Romboutsia and increases intestinal acetate concentrations in Japanese eel [10]. The observed increase in acetate could be attributed to the metabolic activity of Romboutsia, which is known to produce acetate when exposed to fructooligosaccharides, including 1-kestose [12]. SCFAs, such as acetate, are crucial in combating pathogens by inducing toxic acid stress in the gut environment [24]. This suggests that the notable reduction in Turicibacter abundance in the wild-born beluga may be linked to a decrease in intestinal pH resulting from the SCFA production by the proliferating Romboutsia.

In conclusion, our study provides evidence that administering 1-kestose to belugas significantly impacts their health by reducing levels of the potentially pathogenic genus Turicibacter and its virulence-associated collagenase gene levels in the intestine, as well as normalizing elevated blood creatinine levels. These findings advance our understanding of 1-kestose’s role in managing the health of marine mammals, highlighting its potential benefits for the health and welfare of belugas in aquarium settings. It is important to note that, due to the constraints of working with a limited beluga population, this study was conducted without a control group, relying instead on a before-and-after comparison of 1-kestose administration. This approach makes it challenging to definitively attribute observed changes solely to the intervention, rather than natural progression or other environmental and dietary factors. However, the controlled environment maintained for the belugas at this aquarium over the past five years, including stable diet and vitamin supplementation, minimizes the impact of external variables other than the prebiotic administration. While the results of this study are promising, there is a clear need for larger-scale studies with control groups to further validate the effectiveness of 1-kestose in enhancing health outcomes for marine mammals in captivity. The preliminary findings presented here provide a foundation for future research and highlight the importance of innovative approaches in the healthcare of marine mammals.

Conflicts of Interest

Tadashi Fujii, Hideaki Takahashi, Yoshiki Hirooka, and Takumi Tochio are members of BIOSIS Lab. Co., Ltd. Nobuhiro Kondo is affiliated with WELLNEO SUGAR Co., Ltd. The other authors declare no conflicts of interest.