41
42
2. Materials and methods (1) Cage culture
A total of 500 adult M. nudus (45–55 mm TD) were collected by scuba diving from a barren at depths of 3–4 m off Nojima Island, Shizugawa Bay, Miyagi Prefecture (38°40΄N, 141°30΄E), on 11 December 2014. Immediately after collection, the sea urchins were kept in five cages (n = 100 each) suspended horizontally along a straight line at a depth of approximately 2 m at a wavesheltered site off Areshima Island in the bay for 155 days until 15 May 2015. This site is the same as Section 1. The horizontal distance between each cage was approximately 3 m and the vertical distance between each cage and the seafloor ranged 7.5–9.3 m. The polyethylene cages were cuboid (90 × 87 × 45 cm) with 3 cm meshes. Fresh S. japonica cultivated were fed to the urchins ad libitum every 7–10 days. From observation of the amount of uneaten kelp, I regulated the amount of kelp fed to the sea urchins. The kelp was cultivated by longline method. Water temperature during culture was measured every 15 min by two data loggers (HOBO UA002–64, Onset) attached to the upper surface of the cages at both ends of the culture line. Daily water temperature was calculated as an average of 96 data points over 24 h.
During the cage culture, the ambient photoperiods of daily light and dark length changed from 9.5 h:14.5 h to 11.5 h:12.5 h (National Astronomical Observatory of Japan 2016).
(2) Laboratory culture
Eighteen adult M. nudus were collected from the same site on the same day of cage culture. After collection, they were kept in a cool box with moist urethane mats immersed in seawater, then transported to the laboratory in Sendai (38°28΄N, 140°87΄E) for about 2 h. Each sea urchin was held individually in a 3L aquarium (26.5 cm length (upper), 22
43
cm length (bottom), 14.5 cm height, and 10.5 cm width) of a circulation rearing system (ZHab Mini System, Aquatic Habitats, Apopka, FL) with 18 aquaria on three shelves. A fluorescent was set above six aquaria on each shelf. The feeding experiment was started immediately after transportation. Using artificial seawater (Live Sea Salt, Delphis, Itami, Japan), the turnover rate of seawater in each aquarium was regulated at two to three times per hour. Water temperature and salinity were controlled almost the same as that of the cage culture in Shizugawa Bay (realtime monitoring data of oceanic environment by Tohoku EcosystemAssociated Marine Science). Seawater salinity, DO, pH, and NH3 ion concentration were measured every 4 days using a COND METER (ES51, HORIBA, Kyoto, Japan), DO METER (OM51, HORIBA, Kyoto, Japan), and pH/ION METER (D
53, HORIBA, Kyoto, Japan), respectively. The seawater was kept at 31–33 psu salinity, greater than 6.64 mgO/L, 7.51–8.17 pH, and less than 500 mg/L NH4 by substituting a third of the rearing seawater every 4 days. Photon flux density of the fluorescents at the surface of aquaria was at 167.2 ± 8.5 (SE) mmol photon/m2/s, measured three times using a quantum sensor (LI250A, LICOR, Lincoln, NE). Of 18 sea urchins, nine individuals were fed fresh S. japonica ad libitum as an experimental treatment and the other nine individuals were starved as a negative controlled treatment. The aquaria of both treatments were selected randomly. The laboratory experiment was conducted for 84 days until March 5, 2015. Seawater temperature in the aquaria was monitored every 1 h by a data logger (RTR52A, T&D, Nagano, Japan). Daily water temperature was calculated as an average of 24 data points over 24 h. One urchin fed with S. japonica and two starved urchins died during the experiment.
(3) Sea urchin treatments
44
At the start of experiments, 40 sea urchins (45–55 mm TD) were collected from the barren in the same manner as that described for the cultures. Because the gonads removed from one urchin were insufficient for use of all measurement and analysis items, 20 urchins were used for measurements of gonad color and the other 20 urchins were used for gonad hardness and moisture. The other measurement and analysis items TD, BW, gonad index, age determination, histological observation, and FAA content) were performed using 40 urchins. At the end of the cage and laboratory experiments, six cultured urchins were collected randomly from each of five cages (n = 30). In addition to these cultured urchins, 15 laboratoryreared urchins that survived were used for all measurement and analysis items. Except for the gonad color measurement, all measurement and analysis items were conducted one time for each urchin. Overall, four treatment groups of M. nudus were assessed: at the start of the rearing experiments, at the end of cage culture (C), at the end of laboratory feeding (LF), and at the end of laboratory starvation (LS). The measurements and analyses conducted were same as those in Section 1. No gonad hardness of LF and LS could be measured because of the small number (n = 7 and 8) and the markedly small size.
(4) Statistical analysis
Statistical analysis was conducted using JMP 10 software (SAS Institute Inc.). Data were tested for normality (Shapiro–Wilk Wtest) and homogeneity of variances (Levene’s test). Some of the data were logtransformed to ensure normality and homogeneity of variances. Differences in TD; BW; gonad index; L*, a*, and b* values; ΔE*ab; and FAA contents at the start and end of the experiments among the four treatments were analyzed with oneway ANOVA with Bonferroni correction (α = 0.0019). Tukey’s multiple
45
comparison test was performed as a post hoc test. The difference in gonad hardness between the start and end of cage culture was analyzed with Student’s ttest.
3. Results
(1) Water temperature
Figure 9 shows the changes in daily water temperatures. Water temperature of the laboratory culture almost matched with that of the cage culture in Shizugawa Bay. The water temperature at the start of cultures was 11.9°C, it decreased to 5.1°C in midMarch, and then became erratic. From late March, it increased to a peak of 14.3°C in early May, and then sharply decreased to 8.9°C in midMay.
(2) Urchin body size and age
Table 8 shows TD and BW weight of M. nudus in each treatment. There were significant differences in TD and body weight among treatments (Table 9). The TD and BW of the sea urchins in treatment C were significantly larger than those in the other treatments (p < 0.01). There was no significant difference in BW between the sea urchins at the start of the experiment and in LF (p = 0.33). The BW of the sea urchin in LS was significantly less than that in the other treatments (p < 0.05). All urchins used for experiments were 3 y of age.
(3) Gonad development and gonad index
Gonad developmental stages of M. nudus by sex in each treatment are shown in Table 10. Sixtyfive percent of gonads at the start of the experiment were in the recovering stage, with small numbers of primary spermatocytes or previtellogenic oocytes along the acinal
46
Dec. Jan. Feb. Mar. Apr. May
Cage culture Lab culture
Figure 9. Daily water temperatures at the upper surface of cultured cages (n = 96) and in the main tank of the circulation rearing system (n = 24).
47
Table 8. Test diameters (TD) and body weight (BW) of Mesocentrotus nudus by treatment (mean ± SE).
Start, C, LF, and LS indicate M. nudus at the start of the rearing experiments, at the end of cage culture, at the end of laboratory feeding, and at the end of laboratory starvation, respectively. Lowercase letters indicate significant differences among treatments (p <
0.05).
Treatment N TD (mm) BW (g) Start 40 47.5 ± 0.3b 46.2 ± 0.6b C 30 52.9 ± 0.4a 70.8 ± 1.2a LF 8 47.6 ± 0.5b 49.8 ± 2.6b LS 7 48.0 ± 0.6b 38.4 ± 0.6c
48
Table 9. Results of oneway analysis of variance for test diameter, body weight, gonad index, color, moisture, and free amino acid (FAA) contents in gonads of Mesocentrotus nudus among treatments.
df MS F P Test diameter Treatment 3 180.53 57.19 < 0.001
Error 80 3.16
Body weight Treatment 3 4115.61 135.93 < 0.001 Error 80 30.64
Gonad index Treatment 3 1.63 176.11 < 0.001 Error 80 0.01
L* Treatment 3 0.03 18.52 < 0.001 Error 60 <0.01
a* Treatment 3 16.89 4.88 0.004
Error 60 3.46
b* Treatment 3 99.68 3.62 0.018
Error 60 27.52
∆E*ab Treatment 3 0.59 11.76 < 0.001 Error 60 0.05
Moisture Treatment 3 33.41 4.58 0.006 Error 60 7.30
Total FAA Treatment 3 6634900 61.71 < 0.001 Error 80 107518
Aspartic Acid Treatment 3 2.69 2.04 0.115 Error 80 1.32
Glutamic Acid Treatment 3 0.70 23.40 < 0.001 Error 80 0.03
Alanine Treatment 3 144092 38.50 < 0.001 Error 80 3743
Glycine Treatment 3 49692.4 3.12 0.031 Error 80 15926.0
Proline Treatment 3 3520.1 15.68 < 0.001
49
Error 80 28522.7
Serine Treatment 3 2.11 47.01 < 0.001 Error 80 0.04
Threonine Treatment 3 4.93 74.66 < 0.001 Error 80 0.07
Arginine Treatment 3 90401.2 13.44 < 0.001 Error 80 6724.8
Histidine Treatment 3 5832.0 71.08 < 0.001 Error 80 82.05
Isoleucine Treatment 3 2.73 60.98 < 0.001 Error 80 0.04
Leucine Treatment 3 3.40 85.76 < 0.001 Error 80 0.04
Lysine Treatment 3 0.28 11.63 < 0.001 Error 80 0.02
Methionine Treatment 3 8288.60 48.10 < 0.001 Error 80 170.24
Phenylalanine Treatment 3 2.08 47.16 < 0.001 Error 80 0.04
Tyrosine Treatment 3 83581.9 104.05 < 0.001 Error 80 803.3
Valine Treatment 3 71081.9 79.13 < 0.001 Error 80 898.3
Taurine Treatment 3 1.99 70.05 < 0.001 Error 80 0.03
Ornithine Treatment 3 2342.3 116.07 < 0.001 Error 80 20.18
α-aminobutyric acid Treatment 3 93.42 22.01 < 0.001 Error 80 4.24
Significances (α < 0.0019 due to the Bonferroni correction) are shown as bold.
(Continued)
50
Table 10. Gonad developmental stage of Mesocentrotus nudus by treatment and sex.
Start, C, LF, and LS indicate M. nudus at the start of the rearing experiments, at the end of cage culture, at the end of laboratory feeding, and at the end of laboratory starvation, respectively.
Treatment
Recovering Growing
Female Male Percent
(%) Female Male Percent (%)
Start 8 18 65.0 2 12 35.5
C 3 2 16.7 13 12 83.3
LF 3 4 87.5 0 1 12.5
LS 2 4 85.7 1 0 14.3
51
wall and with NP filling the lumen. Approximately 83% of gonads in C were at the growing stage, with increasing numbers of spermatocytes or early vitellogenic oocytes along the acinal wall and with NP filling the lumen. By contrast, most gonads in LF and LS were in the recovering stage.
Figure 10A shows gonad indices of M. nudus in each treatment. Gonad indices differed significantly among treatments (Table 9). The gonad indices in treatments C and LF were 16.9 and 14.0, respectively, with no significant difference between them, but significantly higher than those at the start of the experiment and LS (p < 0.01). The gonad indices in treatment LS were significantly lower than those at the start of the experiment (p < 0.01).
(4) Gonad Moisture and Gonad Hardness
Gonad moisture and gonad hardness for each treatment are shown in Figure 10B, C.
There were no significant differences in gonad moistures among treatments (Table 9).
Gonad hardness at the start of the experiment and in treatment C was 0.45 ± 0.03 N (mean
± SE) and 0.20 ± 0.01 N, respectively. There was a significant difference between treatments (F = 101.1, p < 0.0001).
(5) Gonad Color
L*, a*, and b* values of gonads from each treatment are shown in Figure 11A–C.
There were significant differences in L* value among treatments (Table 9). There was no significant difference in a* and b* values among treatments (Table 9). L* values of gonads in treatment LS were significantly lower than those in the other treatments (p <
0.01). There were not significant differences in L* values among gonads from the start of
52
Figure 10. Gonad indices (A), Gonad moisture (B), and gonad hardness (C) (mean ± standard error) of Mesocentrotus nudus. Start, C, LF, and LS indicate M. nudus at the start of the rearing experiments (n = 40 for gonad indices, n=20 for gonad moisture and hardness), at the end of cage culture (n = 30), at the end of laboratory feeding (n = 8), and at the end of laboratory starvation (n = 7), respectively. A, B and C indicate significant differences among treatments (p < 0.05).
B
A A
C
A
B
C
n.d.
Start C LF LS
A
B
n.d.
53
Figure 11. L* (A), a* (B), b* (C) and ∆Eab* (D) values (means ± SE) of Mesocentrotus nudus gonads.
Start, C, LF, and LS indicate M. nudus at the start of the rearing experiments (n = 20), at the end of cage culture (n = 30), at the end of laboratory feeding (n = 8), and at the end of laboratory starvation (n = 7), respectively. A and B indicate significant differences among treatments (p < 0.05). (From Figure 3, Takagi et al. 2018)
Start C LF LS
A A A
B
B
A A
A
A
B
C
D
54
the experiment, C, and LF. ΔE*ab values of gonads in each treatment are shown in Figure 3D. There were significant differences in ΔE*ab values among treatments (Table 9). The values of gonads from C were significantly lower than those in the other treatments (p < 0.05). There were no significant differences in values among the start of the experiment, LF, and LS.
(6) Gonad FAA contents
Total FAA and individual FAA contents in gonads of each treatment are shown in Figures 12 and 13. There were significant differences in total FAA and 17 individual FAA contents, except for Asp and Gly in gonads among treatments (Table 9). Total FAA contents in gonads from C and LF were significantly higher than those from the start of the experiment and LS (p < 0.01). There were no significant differences in total FAA contents between C and LF and between the start of the experiment and LS. Of the umami FAA, Glu contents in gonads from C were significantly higher than those in other treatments (p < 0.01). There was no significant difference in Glu contents between the start of the experiment and LF. The contents of Ala, Pro, and Ser in gonads from treatments C and LF were significantly higher than those from the start of the experiment and LS (p < 0.01), and no significant differences were detected between each of them. Of the bittertasting FAA, the contents of Tyr and Val in gonads from LF were significantly higher than those in other treatments (p < 0.01) and those from C were significantly higher than those from the start of the experiment and LS (p < 0.01). Although the contents of His, Ile, Leu, Met, and Phe in the gonads from C and LF did not differ significantly, they were significantly higher than those from the start of the experiment and LS (p < 0.05).
55
Start C LF LS
A A
B B
Figure 12. Total FAA contents (means ± SE) in Mesocentrotus nudus gonads. Start, C, LF, and LS indicate M. nudus at the start of the rearing experiments (n = 40), at the end of cage culture (n = 30), at the end of LF (n = 8), and at the end of LS (n = 7), respectively.
A and B indicate significant differences among treatments (p < 0.05).
56
A B B
C
-ABA
A B B
C A A
B B
A A B
A A
B B
A A
B C
A
C
A A
B B
A A B
B C
A A
C A
A B C
C B
A A
B C
A A B
B C B
B A
B
C D
A B
C D
A A
B B B A
C C
Start C LF LS
C
A A
Figure 13. Free amino acid contents (means ± SE) in Mesocentrotus nudus gonads. Start, C, LF, and LS indicate M. nudus at the start of the rearing experiments (n = 40), at the end of cage culture (n = 30), at the end of laboratory feeding (n = 8), and at the end of laboratory starvation (n = 7), respectively.
A, B, C and D indicate significant differences among treatments (p < 0.05). Free amino acids, aspartic acid (Asp); glutamic acid (Glu); alanine (Ala);
glycine (Gly); proline (Pro); serine (Ser); threonine (Thr); arginine (Arg); histidine (His); isoleucine (Ile); leucine (Leu); lysine (Lys); methionine (Met);
phenylalanine (Phe); tyrosine (Tyr); valine (Val); taurine (Tau); ornithine (Orn); and α aminobutyric acid (α ABA).
57
The contents of taurine in gonads from the start of the experiment and LS were significantly higher than those in treatments C and LF (p < 0.01).
4. Discussion
A yearround rearing experiment of adult M. nudus with ca. 40 mm diameter fed S.
japonica var. religiosa ad libitum revealed that the highest growth rate occurred in April (Agatsuma 1997). Through S. japonica var. religiosa feeding ad libitum, the body growth of adult M. nudus with 50.9 mm diameters from a barren was recorded for 40 days from early April to midMay (Agatsuma 1997). A significantly larger diameter of the sea urchins of treatment C ended in May than those of treatment LF ended in March indicates that the urchins of treatment C grew from April when water temperature increased, possibly, because of an increase in digested kelp (Agatsuma 1997).
In June, most gonads of adult M. nudus on algal turf at depths of 40–50 m off Okoppe, Aomori Prefecture (Odagiri et al. 1984), and on a crustose coralline barren (Sano et al.
2001; Section1of Chapter 2) are in the recovering stage. By contrast, the gonads of sea urchins on the beds of U. pinnatifida and S. japonica shift from the recovering stage to the growing stage in May (Agatsuma 1997). Basically, enlarged gonads filling NP in lumens are the most commercially valuable (Walker and Lesser 1998; Unuma and Walker 2009; 2010; Walker et al. 2015). Therefore, high percentage of growing stage gonad from treatment C in May suggests the possibility of sea urchin harvest before the fishing season.
Gonad hardness is affected by gonad developmental stage (McBride et al. 2004). In Section 1 of Chapter 2, I suggested that gonads of M. nudus in the growing stage are the most preferable hardness, harder than that in the maturation stage and softer than that in the recovering stage. Increased gonad growth in treatment C compared with the start of
58
the experiment indicates improvement in texture. In the present study, the percentage of growing stage in treatment C was higher than that of urchins at the start of the experiment and gonad hardness of treatment C was lower than that at the start of the experiment. The averaged gonad hardness of urchins, which were cultured during April to June with S.
japonica feeding, was about 0.23 N (Section 1 of Chapter 2). This value was higher than that of treatment C and lower than that at the start of the experiment in the present study.
In addition, the growing and recovering gonads were included equally in the cultured urchin in the previous study, of which growing gonads were less than those of treatment C and more than that at the start of experiments in the present study. Some studies have indicated an inverse relationship between gonad texture and/or hardness, and moisture contents of gonads (McBride et al. 2004; Pearce et al. 2004), or have found no correlation between them (Pearce et al. 2002b; Azad et al. 2011; Section 1 of Chapter 2). In the present study, the correlation between moisture and hardness of gonads was not detected.
These findings suggest that gonad hardness changes with the developments of gonad NP and reproductive cells.
Smallsized gonads of M. nudus from barrens exhibit a brown color (Agatsuma et al.
2005). Results of Section 1 indicated low L* values of brown color gonads in urchins from a barren. Likewise, significant low L* was detected in treatment LS in the present study. Significantly, lower ΔE*ab value of treatment C compared with that of other treatments indicates a great improvement in gonad color. Gonad color is derived from carotenoids biosynthesized from dietary βcarotene (Shpigel et al. 2005; Tsushima 2007). Borisovets et al. (2002) showed that contents of βcarotene and βechinenone (mg/100 g wet tissue) in testis of M. nudus at the growing stage are greater than those at the recovering stage. This suggests that the difference in gonad color among treatments
59
would be due to differences in gonad developmental stage or βcarotene contents in S.
japonica.
The ΔE*ab values at the start of the experiment and of LF were slightly higher than those of M. nudus (female: 7.3; male: 7.4) of which color was evaluated as the same level of urchins from fishing ground (Section 1 of Chapter 2). According to Section 1, the ΔE*ab
values of the smallsized recovering gonads from a barren in February and June were high (February, female: 16.0, male: 17.2; June, female: 9.9, male: 12.1). The species M. nudus from an E. bicyclis kelp bed in February had recovering gonads with greater than 15 gonad indices and low ΔE*ab values (female: 5.5; male: 5.7). To date, there have been no studies conducted on the changes in gonad color from the spent stage to the recovering stage. Carotenoids play important roles for egg production and development, larval survival, and biological functions (reviewed by Tsushima 2007). In particular, bcarotene and bechinenone facilitate phagocytosis (Kawakami et al. 1998). The desirable color of the gonads in December just after spawning (Agatsuma 2013) may be caused by rapid accumulation of carotenoids. In addition, in M. nudus, the phagocytic cells produce reactive oxygen species when they engulf foreign materials (Ito et al. 1992). These suggest that the carotenoids in the recovering gonads in December, when organic matter level decrease in all body compartments under foodlimited conditions (Guillou et al.
2000), may play a role in biological defense (Tsushima 2007).
Gonads from treatment C had significantly higher contents of umamitasting Glu and lower contents of bittertasting Tyr and Val compared with LF gonads. Each FAA content in the gonads of treatment C was similar to that of M. nudus cultured by feeding of fresh S. japonica cultivated from April to June which were highly comparable with those from urchins in an Eisenia kelp bed in a sensory evaluation (Section 1 of Chapter 2).
60
Sea urchin gonad taste is affected by the feed. Bittertasting Val, Lys, Ile, Leu, and His increase in the gonads of M. nudus and H. crassispina fed fish meat (Hoshikawa et al. 1998; Osako et al. 2006). In the gonads of M. nudus fed the red alga Pyropia yezoensis, Gly decreases, by contrast, Val and Lys increase (Inomata et al. 2016). Phillips et al.
(2009) reported that gonads of E. chloroticus fed a diet containing Gly and Glu had the sweetest taste, whereas gonads of urchins fed a diet containing Val and Met had the highest sulfur odor and the least sweet taste. Li et al. (2007) reported that fronds of cultivated S. japonica elongated from December to April and the width and weight of fronds increased after April when maturation began. Nitrogen content in the fronds remained stable until May and increased sharply after June. Carbon content in the fronds increased gradually from January to May and increased sharply from June to July. Free amino acid contents in the fronds also changed through phenology of cultivated S.
japonica. Oishi and Kunisaki (1970) reported that the Free Glu showed the greatest content in fronds, followed by Ala and Pro. These contents increased from April and reached a maximum in June. In particular, the content of Glu sharply increased from May to June. It is likely that some constituents in the frond, which increase toward summer, affect the composition of FAA in the gonads. These past studies suggest that the difference in FAA content in gonads between C and LF treatments would have been affected by variations in the nutrient composition of cultivated S. japonica. Low nutrient concentrations have a negative impact on maturation of S. japonica sporophytes (Akaike et al. 1998; Mizuta et al. 1998). Growth and production of Laminariales sporophytes are promoted by nutrient enrichment (North and Zimmerman 1984; Dean and Jacobsen 1986;
Agatsuma et al. 2014; Gao et al. 2016; 2017) and the addition of nutrients to gametophytes (Deysher and Dean 1986; Gao et al. 2013; Agatsuma et al. 2014). Maturation of the kelp
61
U. pinnatifida sporophytes is accelerated by nitrate fertilization of the gametophytes (Gao et al. 2013). These findings suggest a possibility of further taste improvement before the fishing season by feeding on mature S. japonica.
In the gonads at the start of the experiment and in treatment LS, taurine contents were greater than those at C and LF, and Gly contents were not dissimilar to those in C and LF, although they had significantly lower total FAA contents. Inomata et al. (2016) obtained similar results from the gonads of M. nudus starved during April to June. FAAs are generally present in marine invertebrates, and taurine and Gly play an important role in osmotic regulation (Simpson et al. 1959; Lange 1964; Yancy et al. 1982). Highly concentrated FAA in starved urchin gonads may play an important role in survival.
In the present study, size, color, and hardness of M. nudus gonads were improved by feeding on fresh S. japonica cultivated from December to May before the fishing season.
Laboratory feeding of the kelp increased gonad size until March. Gonad taste was improved by the feeding from December to May compared with that from December to March but was insufficient compared to that from April to June (Section 1 of Chapter 2).
62