**1. Introduction**

Viviparity and lactation are the most important traits in mammals. The developing fetus is connected to the placenta via an umbilical cord and obtains nutrients through it. Milk is the most complete and natural food of mammals that offspring can consume during the early stages of their life, as it ensures proper nutrition and development. Two types of reproductive taxa have been proposed in mammals based on whether a pregnant and lactating female can catabolize the energy stored in the body to grow fetuses and produce milk [1–4]. Mammals who catabolize accumulated energy resources are called capital breeders, whereas income breeders use energy resources gained concurrently; mysticetes and true seals are generally classified as capital breeders, whereas odontocetes and fur seals are generally classified as income breeders.

Cetaceans give birth and suckle in inaccessible oceans, making it difficult to observe their reproduction, especially in large mysticete whales. Studies focused on the biochemistry and ecology of large whales have been mainly conducted using preserved samples from commercial whaling in the past, but there is little information on reproduction because of the restricted whaling for lactating females and calves. Owing to the international ban on whaling, new samples from large cetaceans are not easy to collect [5–7]. Currently, most studies on lactation and mother-to-offspring relations in marine mammals are conducted using pinnipeds as they give birth and nurse pups on land or ice in accessible areas; they are easy to observe and sample [4, 6–9].

Stable isotope ratios of nitrogen (δ15N) and carbon (δ13C) are used exclusively to study the ecology and biochemistry of marine mammals, and most samples used in these studies are the muscle, liver, blood components, hair, whiskers, bones, skin, and teeth. However, the integrated terms of dietary information are different among these tissues because of different turnover rates [10, 11]. Some studies analyzed two tissues at different turnover rates from individual animals, that is, red blood cells and plasma samples, bone and muscle samples, and hair and plasma samples, to estimate the past and recent information on dietary ecology and biochemistry [12–16].

As the contamination of mercury (Hg) and cadmium (Cd) in milk is very low [17], these burdens in offspring are usually low. However, after weaning, Hg and Cd concentrations in some mammals tend to increase sharply, reflecting the feeding on fish and cephalopods [18–22]. In contrast, adult marine mammals located at high trophic positions are extremely contaminated with Hg [23, 24]. As Hg passes through the placenta, there is great concern regarding the neurotoxicological effects of Hg on infants of not only human populations but also marine mammals [25, 26]. As fetal samples of marine mammals are difficult to obtain, the Hg exposure of the fetus is indirectly estimated using the Hg concentrations in lanugo hair [16, 27–29] and in the red blood cells of neonates [16].

This chapter first describes the different reproduction strategies between mysticetes and odontocetes and between seals and fur seals, and then reviews the studies on lactation in marine mammals using the δ15N and δ13C signatures and Hg concentration. Next, we discuss our data on the ontogenetic profiles (focused on lactation) of δ15N and δ13C signatures and Hg concentration in muscle samples of stranded common minke whale (*Balaenoptera acutorostrata*; MW) (mysticete), Dall's porpoise (*Phocoenoides dalli*; DP) (odontocete), and Kurill harbor seals (*Phoca vitulina stejnegeri,* HS) (pinniped) analyzed in our laboratory. Finally, we compare the δ15N and δ13C values in muscle samples of stranded calves from MW, DP, and killer whale (*Orcinus orca*; KW), and suggest that these values in calves could be excellent proxies for maternal forging habits and trophic levels.

### **2. Lactation in marine mammals**

#### **2.1 Lactation and milk composition**

Most baleen whales (mysticetes) are characterized by long migrations between the feeding and breeding grounds, and the storage of adequate energy for pregnancy and lactation, giving birth and lactating but fasting or eating relatively little [6]. Most baleen whale species have relatively brief lactation (nursing and weaning) periods of 5–7 months, except for bowhead whale (*Balaena mysticetus*, BW) [5, 6]. They produce

#### *Typical Changes in Carbon and Nitrogen Stable Isotope Ratios and Mercury Concentration… DOI: http://dx.doi.org/10.5772/intechopen.103067*

milk relatively low in water (40–50%), high in fat (lipids) (30–50%), and moderately high in protein (9–15%), relying on energy stored before parturition to support the rapid fattening and growth of their calves over a brief lactating period [6]. In contrast, most toothed whales, dolphins, and porpoises (odontocetes) have much more extensive lactations typically lasting 1–3 years, during which the mothers feed. Their milk is higher in water (60–70%) and lower in fat (10–30%) than that of mysticetes but has similar levels of protein (8–11%) [6]. Cetaceans with brief lactation periods tend to produce high-fat milk, investing heavily in offspring for a brief period compared to those with prolonged lactation periods, which tend to produce low-fat milk. The prolonged weaning period in odontocetes may be related to the learning of predatory and social skills, whereas mysticetes, which feed on swarms of zooplankton and schools of small fish, do not need such learning.

The mysticete-odontocete contrast is similar to the contrast between the true seals (family Phocidae) and the fur seal (family Otariidae) [4, 6–8, 30], although both families are piscivores. Most phocid mothers fast during the lactation period rely on energy stored before parturition, which is generally less than 2 months in duration, and pups are usually weaned abruptly. In contrast, otariid mothers feed throughout lactation (except for the initial perinatal period), which may last 1–2 years, and most otariid species are weaned gradually. Furthermore, the milk of phocids is more energy-rich (higher fat concentration) than that of otariids. The different reproduction strategies between mysticetes and odontocetes and between phocids and otariids can be explained by the different breeding strategies as capital and income breeders [1–4].

Weaning involves the transition from nutritional dependence on milk to solid food, and milk and solid food can often be found in the stomachs of gradually weaned offspring. Body lengths (BL) in weaned cetaceans have been reported to be correlated with maternal BL [31].

#### **2.2 Carbon and nitrogen stable isotope ratios**

Stable isotope analyses of carbon (δ13C) and nitrogen (δ15N) are useful tools for obtaining information on feeding ecology. The δ15N value shows a stepwise increase with increases in the trophic level through a food chain, whereas the δ13C value is used to estimate the relative contribution to the diet of potential primary sources [11, 32, 33]. Significant increases of <sup>3</sup>‰ in the <sup>δ</sup>15N value have been shown to occur between predators and their prey, whereas the trophic fractionation for δ13C values is smaller than that for δ15N values, averaging approximately 1‰ [11, 32–35] (**Figure 1**). The δ13C and δ15N profiles are used in studies of habitat preferences such as pelagic *vs*. benthic, and nearshore *vs*. offshore *vs*. estuarine [33], as well as the geographical differences of inhabitants [11, 32, 36, 37]. Furthermore, mother-to-offspring transfer of nutrients owing to lactation in marine mammals has been preferentially investigated using δ13C and δ15N signatures. Higher trophic levels (δ15N values) in nursing offspring than those in their mothers are widely observed in terrestrial and marine mammals [3, 11, 32, 33, 38–40], as the trophic level of milk is higher than that of foods which the mother feeds on (**Figure 1**). In contrast, the δ13C value of offspring does not show a clear-cut pattern because of the large variation in 13C-depleted lipids (fat) concentration in the milk they suckled [3, 33, 40].

As the δ13C values of lipids in mammals are typically 5–7% lower than those of tissue proteins or carbohydrates [11], most studies have analyzed lipid-extracted tissues to remove the potential confounding effect of variation in tissue lipid content on the δ13C values [11, 32, 41]. Consistent with this, the δ13C values in lipid-extracted

#### **Figure 1.**

*Diagram showing the relationships among lactating female, milk and offspring and between pregnant and fetus by stable isotope ratios of carbon and nitrogen. The dashed ellipse indicates the lipid extracted milk.*

milk (shown as a dotted ellipse of milk protein in **Figure 1**) were approximately 5–6% higher than those in milk containing lipids [3, 9, 42, 43].

#### **2.3 Isotopic discrimination and half-life in tissues**

As mentioned above, the enrichment of δ15N values between whole predator and prey (Δ15Npredator-prey) is <sup>3</sup>‰ [3, 11, 32–35]. However, it is practically impossible to measure the whole body δ15N value in large animals; the δ15N value in a tissue is usually substituted for the δ15N value of the whole body, despite the δ15N value in each tissue is different. For instance, the highest δ15N value among seven tissues of fin whales (*Balaenoptera physalus*; FW) was the brain (11.4 0.36%), the middle was muscle (9.88 0.58%), and the lowest was the bone protein (9.19 0.71%) of with a difference of 2.2% [44]. The following is the order of <sup>δ</sup>15N values in tissues that are widely observed [34, 35, 39, 44–46].

*brain > plasma (serum) > liver > muscle = hair (fur) = skin = baleen > red blood cells.*

Thus, the Δ15Npredator-prey values calculated for each tissue are different. Among these tissues, the δ15N value of muscle tissue may closely reflect that of the whole body, as the mass of muscle tissue is the highest in the body (for example, 30% in humans) [11, 32].

#### *Typical Changes in Carbon and Nitrogen Stable Isotope Ratios and Mercury Concentration… DOI: http://dx.doi.org/10.5772/intechopen.103067*

Tissue turnover rates are important for understanding the timeframe of dietary information preserved by the δ15N and δ13C signatures in a given tissue. Tissue turnover should be closely linked to protein turnover, as the tissues typically used for isotopic analyses (hair, muscle, plasma/serum, red blood cells, bone collagen, etc.) primarily comprise proteins, with protein turnover being the most rapid in the liver and plasma/serum, followed by that in the muscle, and slow in red blood cells and bone collagen [11, 32, 47].

The tissue turnover rate in animals generally decreases with body size; for example, the time until 50% replacement of red blood cells was 35 days in humans, but 114 days in beef cattles [11]. Furthermore, the tissue turnover rate in animals is rapid in growing animals, as also in metabolically active tissues [48, 49]. Therefore, as we previously implied, the turnover rate of muscle in MW calves may be markedly faster than that in MS adults, with little time lag between the feeding shift from milk to solid food and the change in δ15N signature in the calf muscle [40].

Unlike the blood, liver, and kidney, the hair, baleen plates, and whiskers do not undergo turnover; past information is preserved without being erased. Hair and baleen plates are reported to grow continuously, at a rate of approximately 1 cm/month for humans [11] and 12.9–20 cm/year for Balaenopteridae species [50], respectively; thus, sampling close to the roots of hair, baleen plates, and whiskers will record the more recent information (δ13C and δ15N values), whereas sampling farther along those will record the information in the past [1, 51–54]. In contrast, the dentin growth layer in teeth allows us to investigate the annual changes in δ13C and δ15N signatures in beluga whales (*Delphinapterus leucas*) [55], KW, and sperm whales (*Physeter microcephalus*) [33]. Studies using the dentin layer have reported a decrease in the δ15N profile because of weaning, with little information on nursing that occurs in a brief period of less than one year. Aubail et al. [56] analyzed not only the δ13C and δ15N values but also the Hg concentration in the dentin layers of ringed seals (*Phoca hispida*) and reported a high Hg concentration owing to the placental transfer of Hg and high δ15N values owing to nursing in the first layer (first year) of dentin.

Bones are well-preserved samples, but they can only be obtained from cadavers. Because of their slow turnover rate, it is suitable for the study of prolonged lactation in animals. Jansen et al. [14] and Vales et al. [57] investigated the ontogenetic dietary changes in harbor porpoises (*Phocoena phocoena*) and South American fur seals (*Arctocephalus australis*) using bone samples. However, the time lag between the weaning and the δ15N signature in the bone should be considered when analyzing the results.

Most blood and milk samples used for the studies on lactation and feeding ecology of marine mammals were obtained from pinnipeds because they are easy to handle and can be sampled blood from pup and mother pairs, and milk and pup tissue pairs [9, 13, 43], in addition to lanugo hair and whisker samples, which is not possible in large whales.

### **2.4 Isotopic relationships among lactating female, milk, and offspring and between pregnant female and fetus: strategies of income and capital breeders**

**Figure 2** shows the δ15N and δ13C values in the muscle samples of KWs that were mass-stranded in Hokkaido. Nine corpses, including three lactating mothers and three calves at few months in age, (including two pairs of lactating mothers and calves), were recovered [58, 59]. Stranded lactating mothers could starve for several days before death, whereas milk was found in the stomach of the calf corpses.

#### **Figure 2.**

*Stable isotope ratios of carbon and nitrogen in muscle of killer whales from 3 calves and 3 lactating females massstranded in Hokkaido. Genetic analysis revealed two pairs of mother-calf relationship. See Table 1.*

The δ15N and δ13C values in the muscle samples were similar between the three calves and the three lactating mothers. The Δ15Noffspring-mother values of two pairs of whales were 1.7‰ and 1.6‰, and the Δ13Coffspring-mother values of two pairs were slightly positive 0.3‰ and 0.3‰ (**Figure 2**). To the best of our knowledge, the Δ15Noffspring-mother values in muscle samples (1.7‰ and 1.6‰) have not yet been reported in cetaceans within nature; however, these values were consistent with those calculated from blood and milk samples of pinniped pairs [9, 39, 60] and captive mammalian pairs [61]. Δ15Noffspring-mother in KWs (1.7‰ and 1.6‰) were smaller than those reported for <sup>Δ</sup>15Npredator-prey (<sup>3</sup>‰), as the lactation of KW calves was at an early stage of nursing before the δ15N-enriched peak (**Figures 3**–**5**).

In contrast, the Δ15Noffspring-mother values in the liver samples were slightly negative (0.5‰ and 0.5‰) (**Table 1**). These phenomena may be explained by the faster


*a total mercury concentration (μg/wet g). <sup>b</sup>*

*sum of 12 PCB isomers.*

*c (mg/lipid wet g).*

*From Endo et al. [58], and Haraguchi et al. [62].*

**Table 1.**

*Analytical results of stable isotope ratios of carbon and nitrogen and mercury concentrations in muscle and liver and PCB and* p,p*'-DDE concentrations in blubber of killer whales mass-stranded in Hokkadio (AKW pod).*

#### *Typical Changes in Carbon and Nitrogen Stable Isotope Ratios and Mercury Concentration… DOI: http://dx.doi.org/10.5772/intechopen.103067*

turnover rate of nitrogen in the liver than in the muscle [10, 47], and the increase in maternal δ15N values by starvation [63]. The Hg concentrations in the muscle and liver samples of lactating mothers were markedly higher than those of the calves, and this trend was particularly pronounced in the liver samples. In contrast to Hg, polychlorinated biphenyl (PCB), and 1,1-dichloro-2,2-bis(p-chlorophenyl) ethylene (*p,p*'-DEE) concentrations in blubber samples were significantly higher in calves than in lactating mothers because of the suckling of contaminated milk containing lipophilic compounds of PCB and *p,p*'-DEE [62].

Two types of reproductive strategies have been proposed for mammals [1–4, 9]. In capital breeders, lactating mothers catabolize their tissues to grow fetuses and produce milk. The trophic levels (δ15N values) of fetuses and milk are higher than those of pregnant females and lactating mothers, respectively (**Figure 1**); that is, the fetus and nursing offspring consume their mother's tissues [32, 33]. In contrast, in income breeders, fetuses and milk should not be enriched in 15N during parturition and lactation, as resources ingested by the pregnant female and lactating mother will not have been incorporated into her own tissues, and these are directly routed into fetal tissue and milk production.

Considering nitrogen in more detail, capital breeders show positive Δ15Nmilk-mother values: 2‰ in the northern elephant seal (*Mirounga angustirostris*) [13], 1‰ in polar bears (*Ursus maritimus*) [42], 0.5‰ in the southern right whale (*Eubalaena australis*) [38], and 0.3‰ in FW [3] (**Figure 1**). Contrary to the positive values in capital breeders, the <sup>Δ</sup>15Nmilk-mother values of income breeders are negative: 1.0 0.5‰in nine incomebreeder species [61], 0.8‰ in Steller sea lion (*Eumetopias jubatus*) [43] 0.3‰ in deer mice (*Peromyscus maniculatus*) [2], and 1.61‰ and 1.48‰ in two species of fur seals (*Arctocephalus gazella* and *Arctocephalus tropicalis*) [9]. However, Chilvers [64] recently reported a positive Δ15Nmilk-mother value of 0.4‰ in the New Zealand sea lion (*Phocarctos hookeri*), an income breeder, and suggested that the production of milk is not only from the diet but also through catabolizing of tissue. In contrast to milk, sampling of fetuses from marine mammals, especially cetaceans, is very rare, with only a single Δ15Nfetus-mother value of 1.6‰ reported in FW, a capital breeder [3].

#### **2.5 Mercury and cadmium**

As apex predators, odontocetes are exposed to high levels of pollutants, such as Hg and organochlorines, through feeding on contaminated fish [65–67]. In contrast, mysticetes, which feed lower on the food chain (zooplankton and small fish), typically have less significant exposure to pollutants. Cd is accumulates in molluscs, particularly cephalopods, and accumulates in predators via not only marine food webs but also species-specific physiologic mechanisms [18, 19, 68]. Cd is likely to be distributed preferentially in the kidney rather than in the liver of marine mammals, whereas Hg tends to be distributed preferentially in the liver [18, 19, 21, 58]. A sharp increase in the Hg burden in the liver is thought to be a weaning proxy for fish eaters in marine mammals, whereas a sharp increase in the Cd burden in the kidney could be a weaning proxy for cephalopod eaters [22]. The degree of Hg burden in mysticetes is generally low, but the Hg burden in opportunistic feeders of mysticetes could reflect the amount of fish consumed [69].

Methyl Hg, produced within marine ecosystems, is a neurotoxicant that is transported across the blood-brain barrier and placental barrier [26]. There is great concern regarding the neurotoxicological effects of Hg on infants of the human

population and piscivores [26]. Because of the difficulty of sampling from fetal tissue, Hg exposure in the fetus is usually estimated using the Hg concentrations in the scalp hair of infants [26], lanugo hair [27–29], dentin [56], and red blood cells of neonates [16]. Rea et al. [28] reported a high level of Hg (> 40 μg/g) as well as δ15N and δ13C values in the fur of sea lion pups, suggesting that their mothers ate fish contaminated with Hg at the high trophic position during the late gestation. Furthermore, the Hg concentration in red blood cells of suckling offspring could reflect the Hg burden in their fetus stage [26], as the majority of red blood cells are likely to be produced during late gestation [13, 61]. As hair is an excretory route for Hg, the Hg in hair is used for biomonitoring of Hg burden in human populations as well as pinnipeds, and in humans, scalp hair Hg is on average 250 times greater than that in whale blood [11]. Endo et al. [70] reported high Hg concentrations in the scalp hair of residents living in a whaling town and a positive correlation between the Hg concentrations and the δ15N values in their hair. Brookens et al. [71] and Hobson et al. [52] reported high Hg concentrations in the hair of Pacific harbor seals (*Phoca vitulina richardii*) and in the baleen plate of MWs and the correlations between Hg concentrations and δ15N values in these samples.

To the best of our knowledge, the only study on Hg analyzed in the fetus of marine mammals was conducted by Itano et al. [23] using hunted striped dolphins (*Stenella coeruleoalba*), which is heavily contaminated with Hg among odontocetes inhabiting waters around Japan [24]. Itano et al. [23] reported an increase in the Hg burden in striped dolphin fetuses with the increase in pregnancy term. Comparing muscle Hg concentrations in fetuses, suckling offspring, and mature females, they reported values of 0.900 0.310, 0.520 0.060, and 11.100 0.300 μg/wet g, respectively. In addition, they reported trace levels of Hg in milk (26 ng/g) and high levels of Hg in the blood of adult females (1.200 0.200 μg/g), which was 46 times higher than that in the milk. Thus, the Hg concentration in calves, which are contaminated with a high Hg burden through the placenta, is likely to be diluted by the growth dilution effect, even though a small amount of Hg could be taken from milk. In contrast, the Hg concentration in breast milk and blood in women at 6 weeks after delivery were 0.6 0.4 and 2.2 1.9 ng/mL, respectively, and that in blood was only 3 times higher [25]. No further studies on Hg transfer through the placenta and milk have been conducted in cetaceans.
