Changes in Carbon, Nitrogen, and Oxygen Stable Isotope Ratios and Mercury Concentrations in Killer Whales (Orcinus orca) during and after Lactation

Abstract

The changes in the stable isotope ratios of carbon (δ¹³C), nitrogen (δ¹⁵N), oxygen (δ¹⁸O), and mercury (Hg) concentrations in muscle and liver tissues during and after lactation were studied in killer whales stranded along the coast of Hokkaido, in the northern area of Japan (n = 16). Calf muscles displayed δ¹³C- and δ¹⁵N-enriched peaks and a δ¹⁸O-depleted peak during lactation. The δ¹³C- and δ¹⁵N-enriched peaks appear to reflect the extensive nursing of ¹³C- and ¹⁵N-enriched milk and the onset of weaning, whereas the δ¹⁸O-depleted peak may be attributable to the extensive nursing of ¹⁸O-depleted milk and the onset of weaning. The δ¹³C and δ¹⁵N values tended to gradually increase after the weaning, whereas the δ¹⁸O values tended to decrease. The δ¹³C and δ¹⁵N levels in calves were similar between liver and muscle samples, whereas those in mature animals were higher in liver than in muscle samples. The isotopic turnover rates of C and N may be similar between the liver and muscle tissues in calves, which are rapidly growing animals. The Hg concentrations in muscle tissues were slightly higher in small calves than in large calves, probably due to the Hg transfer across placenta. The Hg concentrations in liver and muscle samples increased with increasing body length, and those in two liver samples from mature animals exceeded the high-risk threshold for marine mammal health effects (82 μg/wet g).

1. Introduction

Cetaceans, especially large whales, give birth and suckle in inaccessible oceans, making it difficult to observe their reproduction [1]. Currently, stable isotope ratios of nitrogen (δ¹⁵N) and carbon (δ¹³C) are exclusively used to study the reproduction of cetaceans. The δ¹⁵N levels in calves are higher than in their lactating mothers owing to the suckling of ¹⁵N-enriched milk [2,3,4,5]. The δ¹⁵N-enriched peaks were found in calf muscles during the lactation period in common minke whales (Balaenoptera acutorostrata [6]), humpback whales (Megaptera novaeangliae [7]), bowhead whales (Balaena mysticetus [1]), and Dall’s porpoises (Phocoenoides dalli [1]) owing to the extensive nursing and the onset of weaning. The δ¹³C-enriched small peaks due to lactation are also found in bowhead whale and Dall’s porpoise muscles [1], whereas the δ¹³C peak attributable to lactation was not found in the muscles of common minke whales [6] and humpback whales [7]. Cetacean milk generally contains high concentrations of ¹³C-depleted lipids and moderate concentrations of ¹³C-enriched proteins [8,9], and the lipid concentrations vary during the lactation period [8,10].

The stable isotope ratio of oxygen (δ¹⁸O) in teeth and bones has been increasingly used to discriminate, verify, and identify the habitat of marine animals because this value strongly reflects the geographic and climatic conditions of habitats [11,12,13,14,15,16,17]. Drinking water is the main source of oxygen in the tissues of terrestrial animals [18]. However, cetaceans do not drink water, and they obtain water from ingested food, in which the δ¹⁸O levels are related to environmental water [19]. Small increases in δ¹⁸O values, probably due to breastfeeding, have been reported in the tooth enamel of ancient humans, in addition to the δ¹⁵N enrichment [9,20]. However, the changes in δ¹⁸O values attributable to the lactation have not been reported in cetacean muscles.

Mercury is a global pollutant emitted to the atmosphere from both natural and anthropogenic sources. The mercury accumulation in marine animals generally increases with increasing trophic levels (i.e., increasing δ¹⁵N values) [21,22,23,24,25] and in an age- and/or body length (BL)-dependent manner, with particularly high levels in the liver of odontocetes [26,27,28]. Methylmercury passes through the placenta and causes fetal Hg poisoning [29].

Killer whales (Orcinus orca) are large apex predators; the ages when 90% of the life span has been realized are 34 ± 0.3 for males and 52 ± 0.5 for females [30]. They are the most widespread cetaceans in the ocean, from polar water to tropical seas, although they appear to prefer high latitudes and coastal waters rather than tropical offshore and deep-sea waters [17,31]. There are at least three ecotypes of killer whales in the North Pacific [32], which are genetically distinct and have different dietary preferences: the ‘transient’ ecotype, which feeds on marine mammals, and the ‘resident’ and ‘offshore’ ecotypes, which feed on fish [33].

In February 2005, at the shoreline of Aidomari, Hokkaido prefecture, in the northern area of Japan, a pod of up to twelve killer whales became trapped on drifting ice floes and died after approximately one week (Figure 1). They were the transit ecotype and had eaten mainly seals and squid as determined from the digestive residues in their stomachs [27]. We obtained muscle and liver samples from nine dead killer whales including three calves and three lactating females (Aidomari killer whale (AKW) samples). We thereafter obtained three killer whale samples stranded in 2010 and 2015 from the Stranding Network of Hokkaido (SNH10055, SNH10057, and SNH12015, see

Table 1. Stable isotope ratios of carbon, nitrogen and oxygen and mercury concentrations in muscle and liver of killer whales stranded along the coast of Hokkaido, Japan. Data from SNH16006, SNH16035, SNH16008 and SNH17011 were added to previous data (Endo et al. [34]).

				Muscle				Liver
		Body Length (m)	Age (y)	δ13C (‰)	δ15N (‰)	δ18O (‰)	Hg (μg/ wet g)	δ13C (‰)	δ15N (‰)	δ18O (‰)	Hg (μg/ wet g)
SNH16006 *	Calf male	2.2	<1	−18.2	17.8	16.3	0.22	ND	ND	ND	ND
SNH16035 *	Calf female	2.3	<1	−18.3	16.9	15.9	0.21	−18.3	17.2	16.0	0.48
AKW3 **	Calf male a	2.7	<1	−16.7	18.1	12.4	0.10	−16.6	18.6	12.6	0.30
AKW8 **	Calf female b	2.7	<1	−16.9	18.1	12.3	0.08	−16.7	18.7	13.4	0.50
AKW7 **	Calf female c	3.0	<1	−16.8	18.2	11.9	0.07	−16.6	18.7	13.3	0.30
SNH12015 **	Weaning calf d	3.8	3	−19.3	15.1	15.5	0.52	−19.3	15.4	17.6	2.36
AKW2 **	Mature female e	5.6	13	−17.2	16.0	13.1	1.27	−16.8	18.2	13.0	57.4
SNH16008 *	Mature female	5.7	ND	−17.5	15.8	14.2	1.68	ND	ND	ND	ND
SNH10055 **	Mature female	5.8	ND	−18.2	16.5	15.0	2.45	−18.0	18.0	14.4	107.6
AKW6 **	Mature female f	6.0	17	−17.1	16.4	12.7	1.06	−16.8	19.2	13.7	38.0
AKW9 **	Mature female g	6.5	29	−17.2	16.5	12.0	1.26	−16.7	19.2	13.4	62.4
AKW4 **	Mature female	6.6	24	−17.1	16.8	12.7	1.25	−15.9	19.1	11.7	55.4
AKW5 **	Mature female	6.9	59	−17.0	16.4	12.5	1.30	−16.3	18.6	12.0	97.8
SNH10057 **	Mature male	7.0	ND	−18.0	15.9	15.0	3.24	−17.1	17.3	17.2	53.3
SNH17011 *	Mature male	7.1	ND	−17.7	16.3	13.5	2.03	−17.0	18.5	13.7	41.0
AKW1 **	Mature male	7.7	34	−16.9	17.0	13.1	1.46	−16.5	18.8	12.7	35.1

N.D.; not detected; ^a Suckling, no mother among AKW. ^b Suckling, calf of AKW9. ^c Suckling, calf of AKW6. ^d Corrected from the juvenile female and age 2 (Endo et al. [34]) to the weaning calf and age 3. ^e Lactating, no calf among AKW. ^f Lactating, mother of AKW7. ^g Lactating, mother of AKW8; * Current study ** Endo et al. [34].

), and reported the analytical data of δ¹³C, δ¹⁵N, and δ¹⁸O values and Hg concentrations in the combined killer whale samples [34]. In this previous study, we could not find the δ¹⁵N-enriched peak and apparent trends in the δ¹³C and δ¹⁸O values during the lactation period because of the small number of calf samples, but these data suggested that the δ¹³C and δ¹⁵N levels were higher in liver tissues than in muscle tissues from mature animals but were similar in these tissues from calves.

In the present study, we quantified the δ¹³C, δ¹⁵N, and δ¹⁸O values and the Hg concentrations in other killer whales stranded along the Hokkaido coast in 2016 and 2017 (SNH samples of two calves and two mature animals) after the previous study [34] for verification and further exploration. In addition, we quantified the δ¹⁸O values in the remaining muscle samples of common minke whales in which we had quantified the δ¹³C and δ¹⁵N values before [6] to compare with the change of δ¹⁸O values in killer whales. On the basis of these results, we studied the BL at the onset of weaning using the calculated δ¹⁵N-enriched peak in calf muscles; investigated whether the δ¹³C- and δ¹⁸O-enriched peaks are found in calf muscles; and compared the difference in δ¹³C values, δ¹⁵N value, and δ¹⁸O values between liver and muscle tissues (δ¹³C_liver-muscle, δ¹⁵N_liver-muscle, and δ¹⁸O_liver-muscle, respectively) in calves with those in mature animals. Finally, we discussed placental transport of mercury to neonates, and compared hepatic mercury levels in mature killer whales with those previously reported in other odontocetes.

2. Materials and Methods

2.1. Samples of Killer and Common Minke Whales

Muscle samples from killer whale calves (SNH 16006 and SNH 16035), a mature male animal (SNH 16008), and a mature female animal (SNH 17011) stranded on the coast of Hokkaido, Japan in 2016 and 2017 were obtained from SNH (http://www.kujira110.com/ accessed on 30 March 2024). Liver samples from one calf (SNH 16006) and both mature whales (SNH 16008 and SNH 17011) were also obtained from SNH. These samples were quantified with the δ¹³C, δ¹⁵N, and δ¹⁸O values and the mercury concentrations. All killer whale data from the previous and current studies are summarized in Table 1, and stranding locations from both studies are presented in Figure 1. The ecotypes of all SNH whales in both studies are unidentified.

The calves with BLs of 2.2 m (SNH 16006) and 2.3 m (SNH 16035) were considered neonates because the BLs of newborns are reported to be 2.1 m [35] and 2.3 m [36]. Three calves with BLs of 2.7 m (AKW 3 and AKW 8) and 3.0 m (AKW 7) are estimated to be approximately 3 months old [34]. In two calves, milk was found in their stomachs, whereas the stomach of the third calf was empty. We believe that the three calves died immediately before or at the start of weaning, as the shortest BL of calves in which solid materials are found in the stomach is 2.6 m [37]. We believe that the animal with BL of 3.8 m (SNH 12015) is approximately 3 years old [38], and it was categorized into the late stage of weaning because the weaning is reported to completely cease at BL of 4.3 m [37] and age of 4 years [39], although our previous study misstated its age as 2 years [34].

Muscle samples from common minke whales (n = 20), in which the δ¹³C and δ¹⁵N values and the Hg concentrations had been previously quantified ([6],

Table 2. Stable isotope ratio of oxygen in muscle of common minke whales stranded in Hokkaido, Japan.

Samples	BL	δ18O	δ13C *	δ15N *	Hg *
	(m)	(‰)	(‰)	(‰)	(μg/wet g)
1	2.6 (C)	13.0	−9.1	11.2	0.03
2	3.0 (C)	13.8	−19.1	12.1	0.01
3	3.9 (C)	12.2	−18.8	13.7	0.04
4	4.1 (C)	13.4	−18.5	13.0	0.01
5	4.1 (C)	13.2	−19.0	13.4	0.02
6	4.4 (C)	12.9	−19.6	12.0	0.06
7	4.5 (C)	13.1	−19.7	13.8	0.02
8	4.6 (C)	11.9	−20.0	12.7	0.01
9	4.7 (C)	13.0	−19.6	12.4	0.07
10	4.8 (C)	12.0	−19.0	12.8	0.02
11	4.9 (C)	12.0	−19.2	11.4	0.06
12	5.0 (C)	12.8	−18.3	12.9	0.03
13	7.2 (M)	11.9	−19.0	11.5	0.12
14	7.4 (M)	12.8	−18.4	10.8	0.11
15	7.6 (F)	12.5	−18.7	12.3	0.14
16	7.7 (M)	12.8	−18.6	11.8	0.13
17	7.7 (F)	13.0	−18.9	12.2	0.15
18	7.9 (F)	11.8	−18.9	12.4	0.07
19	9.5 (M)	12.0	−18.9	11.5	0.17
20	10.2 (F)	11.8	−19.2	12.4	0.18

* The δ¹³C and δ¹⁵N values and Hg concentration in muscle were quoted from the previous report (Endo et al. [6]). C; calf, M; male mature animal, F; female mature animal.

), were quantified the δ¹⁸O values.

The samples from killer whales and common minke whales were stored at −20 °C until chemical analysis.

2.2. Chemical Analyses

Lipids were removed from dried muscle and liver samples by chloroform/methanol extraction [40]. The extraction was performed at least three times and repeated until the color of the extraction solvent became clear. Samples were then analyzed for C, N, and O isotopes.

The ¹³C and ¹⁵N levels in the muscle and liver samples from killer whales, the ¹⁸O levels in the muscle and liver samples from the killer whales, as well as in the muscle samples from common minke whales, were analyzed by an isotope-ratio mass spectrometry (Delta V PLUS or Delta V Advantage, Thermo Fisher Scientific, Tokyo, Japan) as reported previously [6,7,41].

As reported previously [6], total mercury (Hg) concentrations in the muscle and liver samples were quantified using a flameless atomic absorption spectrophotometer (Hiranuma Sangyo Co., Ltd., HG-310, Ibaraki, Japan) after digestion of samples by a mixture of HNO₃, H₂SO₄, and HClO₄. The Hg concentrations in muscle and liver samples were expressed on a wet weight basis, and the determination limit of Hg was approximately 0.01 μg/wet g.

2.3. Statistical Analyses

We investigated whether the relationship between BL and analytical data (δ¹³C, δ¹⁵N, and δ¹⁸O values and Hg concentrations) could be fitted by a linear or exponential function using JMP (SAS Institute Japan Ltd., Tokyo, Japan, version 14.3). The data were analyzed by Student’s t-test and presented as the mean ± SD. The level of significance was set at p < 0.05.

3. Results

3.1. Changes in δ¹³C, δ¹⁵N, δ¹⁸O Values and Hg Concentrations during and after Lactation in Killer Whales

Mature males had longer BLs (7.0–7.7 m, n = 3) than mature females (5.6–6.9 m, n = 7) (Table 1), and these ranges were consistent with the reported BLs of killer whales of mature males (6–8 m) and mature females (5–7 m), respectively [42]. The relationships of BL with δ¹³C, δ¹⁵N, or δ¹⁸O values in muscle samples from calves and mature animals are shown in Figure 2.

The δ¹⁵N values from calves (n = 6) were fitted to a quadratic function (F = 8.0816, R² = 0.84345, p = 0.0619) although not significantly. The calculated δ¹⁵N-enriched peak (convex upward) was located at BL of 2.7 m and δ¹⁵N value of 18.2‰. Similarly, the δ¹³C values from calves fitted to a significantly quadratic function (F = 43.7615, R² = 0.9668, p = 0.0060), with the δ¹³C-enrichedpeak at BL of 2.9 m and δ¹³C value of −16.7‰. The δ¹⁸O values from calves were significantly fitted to a quadratic function (F = 91.3808, R² = 0.98385, p = 0.0021), but this function exhibited a convex downward peak (δ¹⁸O-depleted peak) at BL of 3.0 m and δ¹⁸O value of 11.7‰.

As observed for muscle samples shown in Figure 2, the δ¹⁵N, δ¹³C, and δ¹⁸O values in the liver samples from calves (n = 5) were significantly fitted to quadratic functions. As data are not shown in figure, the δ¹⁵N-enriched peak was located at BL of 2.9 m and δ¹⁵N value of 18.8‰ (F = 705.41, R² = 0.9986, p = 0.0014), and the δ¹³C-enriched peak was located at BL of 2.9 m and δ¹³C value of −16.5‰ (F = 143.64, R² = 0.9930, p = 0.0069). Meanwhile, the δ¹⁸O-depleted peak was located at BL of 2.9 m and δ¹⁸O value of 12.9‰ (F = 23.406, R² = 0.9590, p = 0.0410). These BLs of the δ¹⁵N, δ¹³C, and δ¹⁸O peaks calculated in liver samples (2.9 m) were extremely close to those calculated in muscle samples (2.7–3.0 m).

The δ¹³C and δ¹⁵N values for muscle samples in the largest calf at the late stage of weaning (SNH 12015) were the lowest among all killer whales, whereas the δ¹⁸O value in the largest calf was higher than that in all mature whales.

The δ¹³C and δ¹⁵N values for muscle samples in mature males increased with increasing BL, whereas no or slight increases in these values were found in both lactating and non-lactating females. The δ¹³C and δ¹⁵N levels in lactating (n = 3) and non-lactating females (n = 4) were similar, being −17.2 ± 0.1 and −17.5 ± 0.5‰, respectively, for the δ¹³C values and 16.3 ± 0.3 and 16.4 ± 0.5‰, respectively, for the δ¹⁵N values. Conversely, the δ¹⁸O levels in mature males, lactating females, and non-lactating females tended to decrease with increasing BL, and the δ¹⁸O level was slightly lower in lactating females (12.6 ± 0.6‰, n = 3) than in non-lactating females (13.6 ± 1.2‰, n = 4) and mature males (13.9 ± 1.0‰, n = 3).

The Hg concentrations in muscle samples increased with increasing BL (F = 26.7529, R² = 0.6564, p = 0.001), although large variation was found in mature males (n = 3) and the Hg concentrations were slightly higher in small calves (0.22 and 0.21 μg/wet g in SNH 16006 and SNH 16035 samples, respectively) than in large calves (0.10, 0.07, and 0.08 μg/wet g in AKW 3, AKW 7, and AKW 8 samples, respectively). The Hg concentrations in lactating females (1.40 ± 0.24 μg/wet g, n = 3) were similar to these in non-lactating females (1.52 ± 0.62 μg/wet g, n = 4). Conversely, the increasing trend of Hg concentrations in liver samples due to increasing BL was unclear because of the large variation in Hg concentrations in the mature animals (60.9 ± 25.7 μg/wet g, 35.1–107.6 μg/wet g, n = 9) (not shown in figure). All Hg concentrations exceeded low-risk threshold for marine mammal health effects (16 μg/wet g), and two Hg concentrations exceeded high-risk threshold (83 μg/wet g) [43].

BLs, δ¹³C, δ¹⁵N, and δ¹⁸O values of AKW samples from mature animals (transient type, n = 6) were compared with those of SNH samples from mature animals (type is unidentified, n = 4). The BLs were 6.1 ± 0.1 m and 6.1 ± 0.8 m, the δ¹³C values were −17.1 ± 0.1‰ and −17.8 ± 0.3‰, the δ¹⁵N values were 16.5 ± 0.3‰ and 16.1 ± 0.3‰, and the δ¹⁸O values were 12.7 ± 0.4‰ and 14.4 ± 0.7‰, which were similar except for the δ¹⁸O values. On the other hand, the Hg concentrations in muscle and liver samples of AKW were 1.27 ± 0.13 μg/wet g (n = 6) and 12.75 ± 0.78 μg/wet g (n = 6), respectively, slightly lower than those in muscle samples (2.35 ± 0.67 μg/wet g, n = 4) and liver samples (15.12 ± 1.85 μg/wet g, n = 3) of SNH.

3.2. Comparison of δ¹³C_liver-muscle, δ¹⁵N_liver-muscle, and δ¹⁸O_liver-muscle Levels in Calves and Mature Animals

We compared the δ¹³C, δ¹⁵N, and δ¹⁸O levels in muscle samples with those levels in the liver samples, respectively (

Table 3. Stable isotope ratios of carbon, nitrogen, and oxygen in muscle and liver samples of calves and mature animals from killer whales.

		δ13C (‰)	δ15N (‰)	δ18O (‰)
Calves	Muscle (n = 6)	−17.7 ± 1.1	17.4 ± 1.2	14.1 ± 2.1
	Liver (n = 5)	−17.5 ± 1.2	17.7 ± 1.4	14.6 ± 2.1
Mature animals	Muscle (n =10)	−17.4 ± 0.4	16.4 ± 0.4	13.4 ± 1.0
	Liver (n = 9)	−16.8 ± 0.6	18.5 ± 0.6 *	13.5 ± 1.6

* Significantly different from muscle samples (p < 0.01).

).

In mature animals, the δ¹³C and δ¹⁵N levels were higher in liver samples (−16.8 ± 0.6 and 18.5 ± 0.6‰, respectively) than in muscle samples (−17.4 ± 0.4 and 16.4 ± 0.4‰, respectively), but only δ¹⁵N levels were significantly different between tissues. On the other hand, no differences in δ¹³C and δ¹⁵N levels were observed between muscle and liver samples from calves. On the contrary, the δ¹⁸O levels were similar between liver and muscle samples from both calves and mature whales, although these levels were slightly lower in mature animals than in calves.

We investigated the relationship between BL and the differences in δ¹³C, δ¹⁵N, or δ¹⁸O values between liver and muscle samples (Figure 3). The δ¹⁵N_liver-muscle values were significantly larger (t₁₂ = 7.525, p < 0.01) in mature animals (2.12 ± 0.48) than in calves (0.44 ± 0.13), and the δ¹³C_liver-muscle values were also significantly larger (t₁₂ = 3.296, p < 0.01) in mature animals (0.59 ± 0.32) than in calves (0.10 ± 0.10). Although the δ¹⁵N_liver-muscle and δ¹³C_liver-muscle values in calves were small, these values likely had small peaks, as in the case of the δ¹⁵N and δ¹³C values (Figure 2). The trends of δ¹⁵N_liver-muscle and δ¹³C_liver-muscle values associated with increasing BL in mature animals were unclear. By contrast, the δ¹⁸O_liver-muscle values in calves (0.98 ± 0.84) increased linearly with increasing BL (F = 13.7243, R² = 0.821, p = 0.0342), whereas no particular trend was observed for the δ¹⁸O_liver-muscle values in mature animals (0.28 ± 1.07).

3.3. Changes in δ¹⁸O Values during Lactation in Common Minke Whales

We quantified δ¹⁸O values in muscle samples from common minke whales (Figure 4) to compare the change in δ¹⁸O values in killer whales (Figure 2). We show the δ¹⁵N-enriched peak at BL of 4.0 m and the variation in δ¹³C values in common minke whale calves in Figure 4, which was quoted from the previous report [6]. No sex-related differences were found in BL, δ¹³C, δ¹⁵N, and δ¹⁸O values and Hg concentrations between calves and mature animals.

The δ¹⁸O values were 12.8 ± 0.6‰ (n =12) in common minke whale calves, and these values tended to decrease throughout the lactation period (F = 3.7185, R² = 0.27106, p = 0.0826), as in the case of killer whale calves excluding the largest calf (Figure 2). By contrast, no particular changes were found in the δ¹⁸O values from mature whales (12.3 ± 0.5‰, n = 8). The δ¹⁸O-enriched peak likely exists between the large calf and small mature animal.

4. Discussion

We studied the changes in δ¹³C, δ¹⁵N, δ¹⁸O values, and Hg concentrations in muscle and liver tissues during and after lactation of killer whales.

The δ¹⁵N-enriched peak due to lactation was found in the current study of killer whale muscles (Figure 2). The increase in δ¹⁵N values across the peak could represent the extensive nursing of δ¹⁵N-enriched milk, and the decrease in δ¹⁵N values from the peak could represent the onset of weaning [1,2,4,5,6,7]. The BLs at the onset of weaning calculated from the δ¹⁵N peak of 2.7 m in muscle samples (Figure 2) and 2.9 m in liver samples were consistent with the actual BLs at the onset weaning of 2.6 and 3.3 m reported in wild killer whales [37]. Meanwhile, the BL and the age at the complete cessation of weaning were reported to be approximately 4.3 m [37] and 4 years [39], respectively. Unfortunately, we could not estimate this BL because our analysis included only one weaning calf sample (SNH12015) and no juvenile samples.

As with the δ¹⁵N-enriched peak, the δ¹³C-enriched peaks attributable to lactation were found in muscle (Figure 2) and liver samples, with both peaks occurring at the BL of 2.9 m. The δ¹³C-enriched peak during the lactation period has been observed in human scalp hairs [2] and bowhead whale and Dall’s porpoise muscles [1]; this peak was not found in common minke whale muscles (Figure 4) and humpback whale muscles [7]. Cetacean milk contains high concentrations of δ¹³C-depleted lipids and moderate concentrations of δ¹³C-enriched proteins, and lipid concentrations vary among species and during the lactation period [8,10]. The δ¹³C-enriched peak found in killer whales (Figure 2), bowhead whales, and Dall’s porpoises [1] might result from the suckling of milk containing δ¹³C-enriched proteins and relatively lower concentrations of δ¹³C-depleted lipids, whereas the lack of δ¹³C peak or no trend of δ¹³C change found in common minke and humpback whales may be due to the large variation of δ¹³C-depleted lipid concentrations in milk [4,14].

The BLs of the δ¹⁸O-depleted peak of 3.0 and 2.9 m calculated from muscle samples (Figure 2) and liver samples, respectively, were extremely close to the BLs of the δ¹⁵N- and δ¹³C-enriched peaks. Thus, the BLs of the δ¹⁸O-depleted peak could be related to the onset of weaning. A similar δ¹⁸O-depleted peak is likely to exist between the large calf and small mature animal (juvenile) of common minke whales (Figure 4) and humpback whales [7]. Suckling appears to decrease the δ¹⁸O levels of calf tissues, and the onset of weaning may increase the δ¹⁸O levels, seemingly because of lower δ¹⁸O levels in milk than in calf muscles and in weaning foods. Interestingly, a large enrichment in δ¹⁸O values with wide variability was reported in the deciduous teeth enamel formed during the first 6 months of breastfeeding in humans [44].

No studies have been reported on δ¹⁸O levels in the milk of marine mammals. Available information on the δ¹⁸O levels in milk has been obtained from terrestrial animals, for which the main source of oxygen in milk is drinking water, and cow’s milk water is enriched in δ¹⁸O by about 3‰ compared to drinking water (−9.2 to −0.02‰) [45]. The δ¹⁸O values in the seawaters the stranded killer whales inhabited were reported to range from −0.1 to −0.01‰ in the Okhotsk Sea [46] and from −2.0 to 0.1‰ in the North Pacific Oceans around Japan [47]. Based on these reports, we estimate the δ¹⁸O values in killer whale milk to be around 1–3‰, which supports our assumption that the δ¹⁸O level in milk is lower than in muscle of calves. Analyses of δ¹⁸O levels in milk and weaning foods are needed to clarify the δ¹⁸O-depleted peak and δ¹⁸O-enriched peak in killer whales (Figure 2).

The isotopic half-life (T_1/2) for C and N in terrestrial mammalian muscles (1–3 months) is markedly longer than that of the more metabolically active tissue of the liver (3–7 days) [48,49], and T_1/2 could decrease generally with a decrease in animal body mass and with increases in metabolic and growth rates [50,51]. Thus, we believe that the isotopic turnover rates of N and C in muscle tissue are much faster in calves than in mature animals, and therefore there is little time lag between the actual onset weaning and that calculated from the calf muscles (Figure 2). We previously estimated the BL at the onset of weaning in common minke whales [6] and humpback whales [7] using the δ¹⁵N-enriched peaks in their muscle samples, which are close to their actual BLs at the onset of weaning. Curve fitting of δ¹⁵N values in calf muscle samples appears to be a powerful method for estimating the onset of weaning.

By contrast, little information is available on the turnover rate of O in marine mammals. The origins of O in milk could be mainly derived from water [45], whereas those of C and N in milk could be mainly derived from proteins, lipids, and carbohydrates [8,9]. The different patterns of the δ¹⁸O values and the δ¹³C and δ¹⁵N values in calves (Figure 2 and Figure 4) could be attributable to their different origins in milk.

It is known that the δ¹⁵N levels are generally higher in liver tissue than in muscle tissue from mature animals [18,52,53,54], and the higher values in liver tissue are thought to reflect the higher metabolic turnover rates in liver tissue. Consistent with this, the δ¹⁵N levels as well as the δ¹³C levels were higher in liver tissue than in muscle tissue of mature animals (Table 2). Conversely, δ¹⁵N and δ¹³C values were similar between calf liver and muscle tissues, suggesting that the turnover rates of N and C between these tissues are similar in calves, which are rapidly growing animals. By contrast, no difference in δ¹⁸O values was found between liver and muscle tissues in both calves and mature animals, whereas the δ¹⁸O value in both tissues was slightly higher in calves than in mature animals and the δ¹⁸O_liver-muscle values in calves increased with increasing BL (Figure 3). These differences between the δ¹⁸O values and the δ¹³C and δ¹⁵N values could be attributed to the different origins of O and C and N and the different evolution of their turnover rates with growth. Further studies are necessary to investigate the different patterns of the δ¹³C and δ¹⁵N values and the δ¹⁸O value between liver and muscle tissues and between the calves and mature animals.

The dentin growth layer in teeth can be used to investigate the continuous annual changes in the values of δ¹³C and δ¹⁵N, although this analysis cannot investigate any changes occurring within the first year such as the δ¹⁵N- and δ¹³C-enriched peaks and δ¹⁸O-depleted peak in muscle tissues (Figure 2) and liver samples. Newsome et al. [39] analyzed the dentin growth layers of killer whale teeth and reported an approximate 2.5‰ decrease in δ¹⁵N value due to the weaning, which occurs mainly during the first 3 years, and an approximate 1.5‰ of increase in δ¹⁵N value thereafter due to the ontogenetic increase in trophic levels. The decrease (2.5‰) and subsequent increase in δ¹⁵N values (1.5‰) reported in the dentin growth layer corresponded to the present finding of a 3.0‰ increase in δ¹⁵N values (range of maximum and minimum) in calves and 2.5‰ increase in δ¹⁵N value from the largest calf to the largest mature animal (Figure 2). We reported previously the peaks of δ¹⁵N and δ¹³C values during lactation period and the ontogenetic slight increase in δ¹⁵N of mature animals in the muscle samples from Dall’s porpoises and bowhead whales [1].

The Hg is preferentially accumulated in the liver of marine mammals [27,28,55]. In agreement, the Hg concentrations in liver samples from mature killer whales were 60.9 ± 25.6 μg/wet g (n = 9), and two Hg concentrations exceeded 83 μg/wet g, which is the high-risk threshold for marine mammal health effects [43]. Available information on the Hg concentrations in livers and δ¹⁵N values in muscles of marine mammals stranded or incidentally caught off the coast of Hokkaido, which may be preyed upon by the transient ecotype of killer whales, are summarized in

Table 4. Comparison of mercury concentrations in liver and δ¹⁵N values in muscle of cetaceans inhabiting waters around Hokkaido.

Species	Hg concentration in Liver (μg/wet g)	δ15N in Muscle (‰)
Killer whale (mature animals) of this study	60.9 ± 25.6 (n = 9) a	16.4 ± 0.4 (n = 10) a
all animals of this study	39.4 ± 36.0 (n = 16) a	16.7 ± 0.9 (n = 16) a
Dall’s porpoises (Phocoenoides dalli)	8.62 ± 10.60 (n = 52) b	13.1 ± 0.9 (n = 56) b
Harbor porpoises (Phocoena phocoena)	6.80 ± 14.9 (n = 45) b	13.2 ± 1.0 (n = 49) b
Spotted seals (Phoca largha)	1.93 and 0.34 b	14.5 and 15.1 b
Harbor seals (Phoca vitulina)	2.96 ± 3.26 (n = 32) b	15.6 ± 0.52 (n = 32) b
Common minke whales (Balaenoptera acutorostrata)	0.103 ± 0.100 (n = 8) b	12.3 ± 0.8 (n = 20) c

^a Current study, ^b Our unpublished data, ^c Endo et al. [6].

. The Hg concentrations and δ¹⁵N values in killer whales were markedly higher than those in the other mammals, particularly common minke whales (baleen whale).

δ¹⁵N and Hg levels were compared between two ecotypes of killer whales off Norway, fish-eaters and seal-eaters. Skin δ¹⁵N levels were markedly higher in the seal-eaters (12.6 ± 0.3‰) than in the fish-eaters (11.7 ± 0.2‰) [56], and the Hg concentrations in the skin were about twice higher in the seal-eaters than in the fish-eaters (Andvik et al. 2020). In the current study, the δ¹⁵N levels as well as the δ¹³C and BL levels in mature animals from SNH (ecotype unidentified) approximated those from AKW samples (transient ecotype), with slightly higher Hg concentrations in muscle and liver tissues. This suggests that the SNH samples are likely to include the samples taken from transient ecotypes.

The Hg concentrations in muscle samples from small calves (SNH 16006 and SNH 16033) were slightly higher than those of large calves (AKW 3, AKW 7, and AKW 8). Similar phenomena have been reported in the calves of striped dolphins [57,58] and humpback whales [7]. The higher Hg concentrations in small calves (neonates) could be explained by the transfer of methylmercury across the placenta, and the lower Hg concentrations in large calves can be explained by the suckling of milk containing trace concentrations of Hg [57,59] and the growth dilution effect [57].

The marine mammal-eating type of killer whale could have higher Hg contamination and δ¹⁵N levels than the fish-eating type [21,56]. However, these comparisons have not been investigated between both types of calves. In the current study, the Hg concentrations in muscle samples were higher in small calves (SNH samples, ecotype unidentified) than in large calves (AKW samples, transient ecotype), whereas the δ¹⁵N levels were lower in small calves than in large calves. Further studies focusing on the Hg load across placenta and the δ¹⁵N levels are needed to increase the number of ecotype-identified calves.

5. Conclusions

The δ¹⁵N- and δ¹³C-enriched peaks were found in muscle and liver samples from killer whale calves because of extensive nursing and subsequent weaning.

The δ¹⁸O-depleted peaks were found in muscle and liver samples from calves, probably because of the consumption of δ¹⁸O-depleted milk and weaning foods. The decrease in δ¹⁸O values during the lactation period was found in common minke whale calves, similar to killer whale calves.

The δ¹³C and δ¹⁵N values were higher in liver samples than in muscle samples of mature killer whales. By contrast, the δ¹³C and δ¹⁵N values in calves were similar between liver and muscle tissues, probably because of the higher metabolic rate of calves, which are rapidly growing animals.

The Hg concentrations in muscle tissues were slightly higher in small calves than in large calves, probably due to the Hg transfer across placenta. The Hg concentrations in two liver samples from mature animals exceeded the high-risk threshold for marine mammal health effects (82 μg/wet g).