- Email: [email protected]
Effects of chemical species of selenium on maternal transfer during pregnancy and lactation
Life Sciences 84 (2009) 888–893
Contents lists available at ScienceDirect
Life Sciences j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / l i f e s c i e
Effects of chemical species of selenium on maternal transfer during pregnancy and lactation Yasumi Anan, Yasumitsu Ogra ⁎,1, Layla Somekawa, Kazuo T. Suzuki Graduate School of Pharmaceutical Sciences, Chiba University, Chiba 260-8675, Japan
a r t i c l e
i n f o
Article history: Received 10 October 2008 Accepted 31 March 2009 Keywords: Selenium Selenomethionine Selenite Selenate Selenoprotein P Milk Pregnancy Lactation
a b s t r a c t Aims: This study compares the transfer from mother to fetuses and pups of selenium (Se) in the form of selenite, selenate, and selenomethionine (SeMet) labeled with different homo-elemental isotopes. Main methods: To completely substitute endogenous Se with natural abundance with Se enriched with a single stable isotope (82Se), female Wistar rats delivered by mother fed 82Se-selenite were fed Se-deficient diet and drinking water containing 82Se-selenite immediately after weaning, and then mated with male Wistar rat at the age of 15–17 weeks. The pregnant rats were divided into two groups. One group was fed Sedeficient diet and drinking water containing 76Se-selenite, 78Se-selenate, and 77Se-SeMet from gestation days 11 to 20. The other group was fed the same diet and drinking water containing the three Se species after delivery for 10 days of lactation. Non-pregnant rats were also fed Se mixture and Se-deficient diet for 10 days. Key finding: Tissue and plasma Se concentrations showed significant changes among non-pregnant, pregnant, and lactating rats. The peak corresponding to selenoprotein P (Sel P) in serum of pregnant rats was reduced. The concentration of 77Se originating from SeMet was higher than those of 76Se from selenite and 78Se from selenate in the stomach content of pups. Significance: Inorganic Se species are more preferably transformed into Sel P than SeMet, and Sel P is effectively incorporated into placenta during pregnancy. On the other hand, SeMet is a more efficient Se source than inorganic Se species during lactation. © 2009 Elsevier Inc. All rights reserved.
Introduction Selenium (Se) is an essential trace element for animals including humans, and exists in the active center of selenoenzymes as a form of selenocysteine (SeCys). Naturally occurring Se species, i.e., inorganic Se (selenite and selenate) and selenoamino acids (SeCys and selenomethionine (SeMet)), are utilized as a nutritional source. Even though all of them are transformed into the assumed common intermediate (selenide) that is utilized for the synthesis of selenoproteins (selenoenzymes), or transformed into methylated metabolites such as selenosugar (Se-sugar) and trimethylselenonium (TMSe) for excretion (Birringer et al. 2002; Ip 1998; Suzuki 2005), it is surmised that their availability and metabolic pathway differ (Suzuki et al. 2006a). Selenite and selenate are directly reduced to the assumed intermediate selenide (Suzuki 2005), whereas selenoamino acids are transformed into selenide through several routes. For
⁎ Corresponding author. Graduate School of Pharmaceutical Sciences, Chiba University, 1-8-1 Chuo, Chiba 260-8675, Japan. Tel./fax: +81 43 226 2866. E-mail addresses: [email protected], [email protected] (Y. Ogra). 1 Present address: Laboratory of Chemical Toxicology and Environmental Health, Showa Pharmaceutical University, 3-3165 Higashi-Tamagawagakuen, Machida, Tokyo 194-8543, Japan. Tel./fax: +81 42 721 1563. 0024-3205/$ – see front matter © 2009 Elsevier Inc. All rights reserved. doi:10.1016/j.lfs.2009.03.023
instance, SeMet is transformed into selenide both by the β-lyase reaction after the successive trans-selenation reaction to SeCys and directly by the γ-lyase reaction (Okuno et al. 2001, 2005; Birringer et al. 2002; Suzuki 2005). In addition to the Se species, physiological condition may be also an important factor affecting Se metabolism. Indeed, it has been reported that the dose-dependent urinary excretion of Se differed between adult and young rats, probably because sources of the sugar moiety of Se-sugar are more abundant in adult rats than in young rats (Suzuki et al. 2005). Thus, it is expected that not only the ingested Se species but also the biological and physiological conditions of animals affect the availability and distribution of Se. It has been reported that Se requirement increases during pregnancy and lactation due to the increase in requirement for maternal transfer of Se to the fetus and the newborn (Smith and Picciano 1986). The effects of the chemical species of Se on Se concentration and selenoenzyme activities in mother and pup have been reported (Lane et al. 1991; Taylor et al. 2005). However, the precise mechanisms underlying the maternal transfer of Se depending on its chemical species remain unclear. We recently developed a new tracer method involving multiple enriched stable isotopes (Suzuki et al. 2006b). Briefly, endogenous natural abundance isotopes were initially substituted with a single enriched stable isotope in host
Y. Anan et al. / Life Sciences 84 (2009) 888–893
889
from each litter. The liver of fetuses and pups and the stomach content of pups were also collected, and stomach contents from rats in the same litter were pooled and treated as one sample. Determination of Se concentration in tissues and body fluids
Fig. 1. Schedule of breeding and dosing for non-pregnant, pregnant and lactating groups.
animal, and then multiple precursors labeled with different enriched stable isotopes were administered simultaneously to the animal. In the present study, we used rats whose endogenous Se with natural abundance was substituted with Se enriched with a stable isotope (82Se) by feeding Se-deficient diet and water containing 82Se-enriched selenite. Then, 76Se-selenate, 78Se-selenite, and 77Se-SeMet were administered simultaneously to rats at different reproductive stages, i.e., non-pregnancy, pregnancy, and lactation. 76Se, 78Se, and 77Se in tissues and body fluids were determined to be of selenate, selenite, and SeMet origin, respectively. Thus, this method makes it possible to clarify the effects of Se species on maternal transfer during pregnancy and lactation under exactly identical animal and analytical conditions.
78
Se- and
Speciation of Se in serum The HPLC system consisted of an online degasser, an HPLC pump (PU713; GL Science Co., Ltd., Tokyo), a Rheodyne six-port injector with a sample loop, and a column. A multi-mode gel filtration column, Shodex Asahipak GS-520 7G (7.5 i.d. × 500 mm, with a guard column, 7.5 i.d. × 75 mm; Showa Denko, Tokyo), was used for the separation of serum. The column was injected with a 200 µl aliquot of serum, and then eluted with 50 mM Tris–HCl, pH 7.4, at a flow rate of 1.0 ml/min. The eluate was introduced directly into a Babington nebulizer of the ICP-MS. Statistical analysis
Materials and methods Preparation of
A 200 mg or 200 µl portion of liver, kidney, muscle, plasma, pancreas, placenta, or stomach content was digested with 1 ml of mixed acid of concentrated nitric acid (HNO3) and 30% H2O2 (v/v = 1/1) in a test tube at room temperature for more than 1 day, then heated at 160–180 °C for several days until complete digestion was achieved. Se concentrations in samples were determined with an inductively coupled plasma-mass spectrometry (ICP-MS; Agilent 7500ce, Agilent Technologies, Hachiouji, Japan).
82
Se-selenite,
76
Se-selenate, and
77
Se-SeMet
Elemental forms of 76Se (97.0% enriched), 77Se (99.9%), 78Se (99.0%), and 82Se (98.9%) were purchased from Isoflex USA (San Francisco, CA, USA). 78Se- and 82Se-selenite were prepared by dissolving each elemental form in nitric acid, followed by adjustment to neutral pH with NaOH. 76Se-selenate was prepared by dissolving elemental 76Se in nitric acid, followed by oxidation with hydrogen peroxide (Kobayashi et al. 2001). 77Se-SeMet was synthesized from elemental 77Se and (S)-(−)-α-amino-γ-butyrolactone in three steps (Plenevaux et al. 1987). Animal experiments Twelve-week-old Wistar rats at gestation day 2 were purchased from Clea Japan Inc. (Tokyo, Japan), maintained in plastic cages at 22± 2 °C with a 12/12 h light/dark cycle, and fed Se-deficient diet (Oriental Yeast Co., Ltd.; Se concentration, b0.02 µg/g diet) and drinking water containing 82Se-selenite at a concentration of 0.2 µg Se/ml ad libitum until pups were weaned. To increase substitution efficiency, feeding the same diet and water was continued up to 15–17 weeks of age. Female rats were assigned to one of three groups, i.e., non-pregnant (n = 3), pregnant (n = 5), and lactating (n = 5) groups. The experimental schedule of breeding and feeding for each group was shown in Fig. 1. Rats of the non-pregnant group were fed Se-deficient diet and drinking water containing 76Se-selenate, 78Se-selenite, and 77Se-SeMet each at 0.1 µg Se/ml ad libitum for 10 days. Rats of pregnant and lactating groups were mated with male Wistar rats purchased from Clea Japan Inc. Rats of the pregnant group were fed the same diet and water as rats of the nonpregnant group ad libitum from gestation days 11 to 20. Rats of the lactating group were fed water containing 82Se-selenite and Se-deficient diet during pregnancy, and fed the same diet and water as rats of the non-pregnant group for 10 days after delivery. At the end of feeding, the rats including pups of the lactating group were sacrificed by exsanguination under ether anesthesia. Heparinized and non-heparinized blood, liver, kidney, muscle, pancreas, and placenta (only from the pregnant group) were collected. Three fetuses and pups were randomly selected
All statistical analyses were performed with StatView 4.51. Differences among three groups were tested using Scheffe's method along with one-factor ANOVA. A p value less than 0.05 was considered to be statistically significant. Results Changes in Se concentration with reproductive stage Total Se concentration was significantly changed in tissues and plasma of pregnant and lactating rats, except kidney (Fig. 2). Compared to the non-pregnant group, Se concentration in the liver was decreased in the pregnant group and further decreased in the lactating group. Plasma Se concentration was significantly lower in the pregnant group than in the lactating and non-pregnant groups. On the other hand, in the muscle and pancreas, Se concentration was significantly increased in the lactating group compared to the other
Fig. 2. Total Se concentrations in the liver, kidney, plasma, muscle, and pancreas of nonpregnant, pregnant, and lactating female rats fed Se-deficient diet and drinking water containing 76Se-selenate, 78Se-selenite, and 77Se-SeMet each at 0.1 µg Se/ml for 10 days ad libitum. Values with different superscript letters (a, b or c) are statistically significantly different among non-pregnant, pregnant, and lactating groups (p b 0.05).
890
Y. Anan et al. / Life Sciences 84 (2009) 888–893
Table 1 Percentage of Se originating from different Se sources with respect to total Se concentration in tissues and plasma of non-pregnant, pregnant, and lactating rats.
Liver
Kidney
Plasma
Muscle
Pancreas
Group
Selenate origin
Selenite origin
SeMet origin
Endogenous Se
NP P L NP P L NP P L NP P L NP P L
20.6 ± 1.7a 20.1 ± 0.9ab 20.4 ± 1.7a 18.5 ± 1.6a 18.2 ± 0.5a 18.2 ± 0.9a 24.9 ± 1.4a 23.4 ± 0.8a 25.9 ± 0.2a 6.4 ± 0.9a 4.6 ± 0.7a 5.6 ± 0.8a 12.7 ± 0.4a 10.3 ± 1.1a 7.7 ± 1.0a
22.4 ± 1.2a 21.7 ± 1.2a 22.7 ± 2.2a 20.9 ± 2.1a 19.7 ± 0.5a 20.0 ± 1.4a 27.7 ± 1.5a 25.9 ± 0.2a 29.7 ± 0.2b 5.2 ± 0.8a 3.2 ± 1.1a 4.3 ± 1.4a 12.5 ± 0.8a 9.4 ± 0.9a 7.4 ± 1.1a
15.0 ± 1.7b 17.7 ± 2.3b 18.8 ± 1.9a 13.3 ± 2.2a 15.5 ± 0.7b 15.0 ± 0.9b 16.6 ± 2.8b 19.2 ± 1.4b 19.6 ± 0.3c 11.7 ± 1.2b 12.3 ± 1.1b 10.1 ± 1.3b 33.0 ± 5.8b 44.6 ± 5.1b 58.2 ± 3.7b
42.0 ± 4.1 40.5 ± 1.2 38.0 ± 3.8 47.3 ± 3.3 46.7 ± 0.6 46.9 ± 3.0 30.9 ± 3.1 31.6 ± 1.3 24.9 ± 0.8 76.7 ± 2.6 79.9 ± 1.7 80.0 ± 3.4 41.8 ± 5.8 35.8 ± 3.2 26.7 ± 1.6
NP: non-pregnant rats, P: pregnant rats, and L: lactating rats. Values with different superscript letters (a, b or c) are statistically significantly different among selenate origin, selenite origin, and SeMet origin in each group (p b 0.05).
groups. Se concentration in the kidney was higher than those in other tissues and plasma, and there were no significant differences among the three groups. Comparison of availability of three Se species among non-pregnant, pregnant, and lactating rats The availability, i.e., incorporation of selenate, selenite, and SeMet into tissues and plasma, was not significantly different among nonpregnant, pregnant, and lactating groups (Table 1), suggesting that pregnancy and lactation hardly affect the availability of these
selenocompounds in the mother's body. Although no significant differences in availabilities of selenate and selenite were observed in all tissues of the three groups, the availability of selenite tended to be slightly higher than that of selenate in the liver, kidney, and plasma. Incorporation of SeMet was significantly lower than those of selenate and selenite in the liver, kidney, and plasma, while SeMet was taken up more efficiently than selenite and selenate in the pancreas and muscle of all groups. Fig. 3 shows changes in the concentration of Se originating from the three Se compounds (selenate, selenite, and SeMet) in three different reproductive stages. Changes in the concentration of Se originating from selenite and selenate well reflected the changes in total Se concentration (Fig. 2) in the liver and plasma. The concentration of Se originating from SeMet in the pancreas was markedly increased by pregnancy and lactation, whereas the concentrations of Se from selenate and selenite were not affected. Speciation of Se in serum using HPLC-ICP-MS Two peaks corresponding extracellular glutathione peroxidase (eGPx) and selenoprotein P (Sel P) were detected at retention times of 10.8 min and 13.0 min at m/z 76, 77, and 78 corresponding to selenate, SeMet, and selenite origins, respectively, in serum of non-pregnant, pregnant, and lactating rats (Fig. 4). Although the elution profiles of 76 Se, 78Se, and 77Se showed similar patterns, the peak heights of eGPx and Sel P at m/z 77 were smaller than those at m/z 76 and 78, suggesting that SeMet was less effectively transformed into these selenoproteins than selenite and selenate. Although the amount of eGPx was comparable among the three different reproductive groups, a marked reduction in the peak corresponding to Sel P was observed in pregnant rats. This result is consistent with the decrease in the plasma total Se concentration of pregnant rats (Fig. 2).
Fig. 3. Changes in Se concentration originating from three Se species, selenate, selenite and SeMet, in the liver (a), kidney (b), plasma (c), muscle (d), and pancreas (e) of nonpregnant, pregnant and lactating female rats fed Se-deficient diet and drinking water containing 76Se-selenate, 78Se-selenite, and 77Se-SeMet each at 0.1 µg Se/ml for 10 days ad libitum. Values with different superscript letters (a, b or c) are statistically significantly different among non-pregnant, pregnant, and lactating groups (p b 0.05).
Y. Anan et al. / Life Sciences 84 (2009) 888–893
891
Fig. 4. Elution profiles of Se isotopes in serum of non-pregnant, pregnant, and lactating female rats fed Se-deficient diet and drinking water containing 76Se-selenate, 78Se-selenite, and 77Se-SeMet each at 0.1 µg Se/ml for 10 days ad libitum.
Transfer of Se through placenta and milk to fetus and pups Se concentration in the fetal liver was comparable to that in the placenta (Table 2). There was no significant difference among the percentage of Se originating from the three Se species in the placenta and fetal liver (Table 2), suggesting that selenite, selenate, and SeMet were homogenously distributed to fetal liver through placenta. Se concentration in mother's milk was estimated by determining Se concentration in stomach content of pups, as stomach content consisted of only coagulated mother's milk. Se concentration in stomach content was 0.13 ± 0.03 µg/mg. The concentration and percentage of Se originating from inorganic species in the liver and stomach content of pups were significantly lower than those of Se originating from SeMet (Table 2). The concentration of Se originating from selenite in stomach content was below the detection limit. These results indicate that SeMet is a better source of Se than selenite and selenate in mother's milk. The percentages of endogenous Se in the stomach content and liver of pups were larger than those of exogenous Se, i.e., selenite, selenate and SeMet, implying that the major part of mother's milk was mobilized from the Se contained in maternal tissues rather than from the dietary Se.
pregnant rats and lactating dams were Se-deficient because Se was transferred to their fetuses and pups, respectively. The reduction in the concentration of liver Se originating from inorganic Se during pregnancy was more severe than that originating from SeMet (Fig. 3a). Inorganic Se, i.e., selenite and selenate, can be directly transformed into selenide, a critical metabolic intermediate of Se in the body, by simple reduction, and selenide is utilized for the biosynthesis of selenoproteins (Suzuki 2005). On the other hand, SeMet is used for the biosynthesis of selenoproteins after degradation to selenide via several metabolic pathways, and is also incorporated into general proteins in place of Met to produce the so-called Se-containing proteins (Suzuki 2005). Therefore, the amounts of selenite and selenate utilized for the biosynthesis of selenoproteins may be larger than that of SeMet, and thus, Se originating from selenite and selenate contained in selenoproteins may be more efficiently transferred to fetus accompanied by the transfer of selenoproteins, resulting in the reduction of those compounds in the dam's body. On the other hand, the reduction of liver Se was not different among selenate, selenite and SeMet origins in the lactating group (Fig. 3a), suggesting that Se-containing protein is also transferred to pups as well as selenoproteins during lactation. Se concentrations in the kidney and muscle were not sensitive to pregnancy, lactation, and ingested chemical species of Se (Fig. 3b and d). eGPx, a protein that acts to protect against
Discussion In the present study, the availability and efficiency of maternal transfer of three Se species were compared using a new tracer method. The same experimental conditions, including animals, sample preparation, and analytical conditions, were achieved by simultaneous administration of multiple selenocompounds labeled with compound-specific stable isotopes. Thus, this new method enabled more accurate comparison and quantitative evaluation of maternal transfer depending on chemical species than experiments using a single tracer in different animals, and was essential to achieving the aim of this study. Se concentration in the liver of pregnant and lactating rats was significantly reduced compared with that of non-pregnant rats (Fig. 2). In particular, Se concentration in the liver of lactating dams decreased by approximately 50% compared to that of non-pregnant rats. It is known that rat requires 1.5 times greater amount of Se during gestation (Smith and Picciano 1986). In this study, Se concentration in drinking water was adjusted to 0.3 µg/ml, which corresponded to the adequate level for non-pregnant rats. The results indicated that
Table 2 Concentrations of Se originating from different sources in the placenta and fetal liver of pregnant rats and liver and stomach contents of pups. Selenate origin Selenite origin SeMet origin Endogenous Se Total Pregnant rats Placenta 56 ± 11a (21.5 ± 1.3) Fetal liver 47 ± 4a (20.1 ± 1.2)
58 ± 11a (22.3 ± 1.0) 45 ± 5a (19.1 ± 2.0)
50 ± 6a (19.5 ± 2.0) 51 ± 5a (21.6 ± 2.0)
96 ± 15 (36.8 ± 1.1) 92 ± 11 (39.3 ± 4.1)
260±40 (100) 234±12 (100)
Pups of lactating group Liver 13 ± 8a (5.0 ± 2.6) Stomach 4 ± 1a content (3.2 ± 1.2)
7 ± 7a (2.8 ± 2.5) Not detectable (0.0)
43 ± 12b (17.8 ± 1.6) 23 ± 12b (17.7 ± 9.4)
177 ± 26 (74.3 ± 5.6) 100 ± 26 (79.1 ± 10.1)
241 ± 47 (100) 127 ± 31 (100)
Se concentration is expressed as ng Se/g tissue. Numbers in parentheses indicate percentage. Values with different superscript letters (a or b) are statistically significantly different among selenate origin, selenite origin, and SeMet origin in tissues (p b 0.05).
892
Y. Anan et al. / Life Sciences 84 (2009) 888–893
Fig. 5. Proposed maternal transfer pathways for selenate, selenite, and SeMet during pregnancy (a) and lactation (b). SeMet: selenomethionine, End Se: endogenous Se, Exo Se: exogenous Se.
oxidative stress in extracellular fluid and bloodstream, is biosynthesized in the kidney (Avissar et al. 1994), and the major part of Se in the kidney is recovered in the insoluble fraction (Hawkes et al. 1985). Therefore, Se might be pooled in the kidney to maintain the concentration of an important selenoprotein, eGPx, during pregnancy and lactation. Indeed, although the concentration of Sel P, a plasma selenoprotein biosynthesized in the liver, was affected by pregnancy, that of eGPx was stable (Fig. 4). The ratio of endogenous Se to total Se in the muscle was the highest among the tissues and plasma examined (Table 1), suggesting that the substitution rate of exogenous Se with endogenous Se was the lowest. Thus, the muscle does not seem to contribute much as an Se source to placenta and milk. Recently, a novel function of ApoER2, a specific receptor of Sel P expressed on cell surface, was reported (Burk et al. 2007; Olson et al. 2007, 2008). ApoER2 was predominantly expressed in the brain and testis. In addition, ApoER2 mRNA expression was observed in the placenta (Kim et al. 1996). These suggest that Sel P acts as an Se transporter to fetus via placenta, resulting in a reduction in the concentration of Sel P in pregnant rats. This coincides with the more severe reduction of liver Se originating in inorganic Se than in SeMet in pregnant rats because inorganic Se, selenite and selenate, is more efficiently assimilated into Sel P than SeMet (Fig. 4). Although the percentages of Se originating from selenate and selenite were higher than that from SeMet in the dam's plasma (Table 1), the percentages of Se originating from the three Se species were not significantly different in the placenta and fetal liver (Table 2). This suggests that SeMet itself is also transferred to fetus without incorporation into Sel P, via an SeMet-specific pathway. It is known that SeMet is preferably incorporated into the pancreas whereas inorganic Se species are hardly incorporated (Suzuki et al. 2006b). Indeed, the preferable incorporation of SeMet into the pancreas was observed (Table 1). Moreover, the concentrations of Se originating from SeMet in the pancreas of pregnant and lactating rats were higher than that of non-pregnant rats (Fig. 3e). Although an SeMet-assimilating protein is suggested to be present in the pancreas, it has not been identified yet (Suzuki et al. 2006b). Insulin and other peptide hormones may be candidates because pancreatic function, such as insulin secretion, is expected to be augmented during late pregnancy and lactation (Bell and Bauman 1997). Further studies are needed to identify the SeMet-assimilating protein in the pancreas.
The chemical species of Se in milk are not fully known yet. However, recently, SeCys, in addition to SeMet, was detected after digestion of milk proteins and derivatization of selenol group (–SeH) (Bierla et al. 2008). This suggests that there are some selenoproteins in milk, and all Se species are possible sources of them. On the other hand, in stomach content, the concentration of Se originating from SeMet was higher than that of Se originating from inorganic Se species (Table 2). This is in agreement with the previous observation that feeding SeMet to livestock, such as cows and pigs, more efficiently increases Se concentration in milk than feeding inorganic Se species (Yoon and McMillan 2006; Muñiz-Naveiro et al. 2006). Muñiz-Naveiro et al. (2007) suggested that SeMet is incorporated into general milk proteins in place of Met. Indeed, the concentration of Se originating from SeMet in the liver of pups was higher than that of Se from inorganic Se species (Table 2). Taken together, SeMet was more effectively transferred to pups via milk than inorganic Se species due to its incorporation into general milk proteins. There is no information of low molecular weight Se species in milk other than selenoproteins and Se-containing proteins. Such low molecular weight Se species as Se-sugars and methylated Se compounds may actually exist in extracellular fluid. Further studies, in particular, speciation studies, are needed. In conclusion, total Se concentrations in tissues and plasma except kidney were affected by gestation and lactation, and as proposed in Fig. 5, inorganic Se and SeMet were transferred to fetuses and pups via different routes. Inorganic Se species, such as selenite and selenate, are more preferably transformed into Sel P than SeMet, and Sel P is effectively incorporated into placenta via its specific receptor, ApoER2, during pregnancy. On the other hand, SeMet is a more efficient Se source than inorganic Se species during lactation because SeMet is incorporated into not only selenoproteins as well as inorganic Se species but also general milk proteins in place of Met. This indicates that SeMet is a better nutritional source of Se for pups than inorganic Se. Acknowledgments The authors wish to acknowledge a Grant-in-Aid for JSPS Fellows from Japan Society for the Promotion of Science (JSPS) (No. 06J01979), and a Grant-in-Aid from the Ministry of Education, Culture, Sports, Science and Technology, Japan (No. 19390033).
Y. Anan et al. / Life Sciences 84 (2009) 888–893
References Avissar N, Ornt DB, Yagil Y, Horowitz S, Watkins RH, Kerl EA, Takahashi K, Palmer IS, Cohen HJ. Human kidney proximal tubules are the main source of plasma glutathione peroxidase. American Journal of Physiology. Cell Physiology 266 (2 35–2), C367–C375, 1994. Bell AW, Bauman DE. Adaptations of glucose metabolism during pregnancy and lactation. Journal of Mammary Gland Biology and Neoplasia 2 (3), 265–278, 1997. Bierla K, Szpunar J, Lobinski R. Specific determination of selenoaminoacids in whole milk by 2D size-exclusion-ion-paring reversed phase high-performance liquid chromatography-inductively coupled plasma mass spectrometry (HPLC-ICP MS). Analytica Chimica Acta 624 (2), 195–202, 2008. Birringer M, Pilawa S, Flohe L. Trends in selenium biochemistry. Natural Product Reports 19 (6), 693–718, 2002. Burk RF, Hill KE, Olson GE, Weeber EJ, Motley AK, Winfrey VP, Austin LM. Deletion of apolipoprotein E receptor-2 in mice lowers brain selenium and causes severe neurological dysfunction and death when a low-selenium diet is fed. Journal of Neuroscience 27 (23), 6207–6211, 2007. Hawkes WC, Wilhelmsen EC, Tappel AL. Subcellular distribution of selenium-containing proteins in the rat. Journal of Inorganic Biochemistry 25 (2), 77–93, 1985. Ip C. Lessons from basic research in selenium and cancer prevention. Journal of Nutrition 128 (11), 1845–1854, 1998. Kim DH, Iijima H, Goto K, Sakai J, Ishii H, Kim HJ, Suzuki H, Kondo H, Saeki S, Yamamoto T. Human apolipoprotein E receptor 2. A novel lipoprotein receptor of the low density lipoprotein receptor family predominantly expressed in brain. Journal of Biological Chemistry 271 (14), 8373–8380, 1996. Kobayashi Y, Ogra Y, Suzuki KT. Speciation and metabolism of selenium injected with 82Se-enriched selenite and selenate in rats. Journal of Chromatography. B, Biomedical Sciences and Applications 760 (1), 73–81, 2001. Lane HW, Strength R, Johnson J, White M. Effect of chemical form of selenium on tissue glutathione peroxidase activity in developing rats. Journal of Nutrition 121 (1), 80–86, 1991. Muñiz-Naveiro Ó, Domínguez-González R, Bermejo-Barrera A, Bermejo-Barrera P, Cocho JA, Fraga JM. Study of the bioavailability of selenium in cows' milk after a supplementation of cow feed with different forms of selenium. Analytical and Bioanalytical Chemistry 385 (1), 189–196, 2006. Muñiz-Naveiro Ó, Domínguez-González R, Bermejo-Barrera A, Bermejo-Barrera P, Cocho JA, Fraga JM. Selenium speciation in cow milk obtained after supplementa-
893
tion with different selenium forms to the cow feed using liquid chromatography coupled with hydride generation-atomic fluorescence spectrometry. Talanta 71 (4), 1587–1593, 2007. Okuno T, Kubota T, Kuroda T, Ueno H, Nakamuro K. Contribution of enzymic alpha, gamma-elimination reaction in detoxification pathway of selenomethionine in mouse liver. Toxicology and Applied Pharmacology 176 (1), 18–23, 2001. Okuno T, Motobayashi S, Ueno H, Nakamuro K. Purification and characterization of mouse hepatic enzyme that converts selenomethionine to methylselenol by its alpha, gamma-elimination. Biological Trace Element Research 106 (1), 77–94, 2005. Olson GE, Winfrey VP, Nagdas SK, Hill KE, Burk RF. Apolipoprotein E receptor-2 (ApoER2) mediates selenium uptake from selenoprotein P by the mouse testis. Journal of Biological Chemistry 282 (16), 12290–12297, 2007. Olson GE, Winfrey VP, Hill KE, Burk RF. Megalin mediates selenoprotein P uptake by kidney proximal tubule epithelial cells. Journal of Biological Chemistry 283 (11), 6854–6860, 2008. Plenevaux A, Cantineau R, Guillaume M. Fast chemical synthesis of [75Se]L-selenomethionine. Applied Radiation and Isotopes 38 (1), 59–61, 1987. Smith AM, Picciano MF. Evidence for increased selenium requirement for the rat during pregnancy and lactation. Journal of Nutrition 116 (6), 1068–1079, 1986. Suzuki KT. Metabolomics of selenium: Se metabolites based on speciation studies. Journal of Health Science 51 (2), 107–114, 2005. Suzuki KT, Kurasaki K, Okazaki N, Ogra Y. Selenosugar and trimethylselenonium among urinary Se metabolites: dose- and age-related changes. Toxicology and Applied Pharmacology 206 (1), 1–8, 2005. Suzuki KT, Doi C, Suzuki N. Metabolism of 76Se-methylselenocysteine compared with that of 77Se-selenomethionine and 82Se-selenite. Toxicology and Applied Pharmacology 217 (2), 185–195, 2006. Suzuki KT, Somekawa L, Kurasaki K, Suzuki N. Simultaneous tracing of 76Se-selenite and 77 Se-selenomethionine by absolute labeling and speciation. Toxicology and Applied Pharmacology 217 (1), 43–50, 2006. Taylor JB, Finley JW, Caton JS. Effect of the chemical form of supranutritional selenium on selenium load and selenoprotein activities in virgin, pregnant, and lactating rats. Journal of Animal Science 83 (2), 422–429, 2005. Yoon I, McMillan E. Comparative effects of organic and inorganic selenium on selenium transfer from sows to nursing pigs. Journal of Animal Science 84 (7), 1729–1733, 2006.
Contents lists available at ScienceDirect
Life Sciences j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / l i f e s c i e
Effects of chemical species of selenium on maternal transfer during pregnancy and lactation Yasumi Anan, Yasumitsu Ogra ⁎,1, Layla Somekawa, Kazuo T. Suzuki Graduate School of Pharmaceutical Sciences, Chiba University, Chiba 260-8675, Japan
a r t i c l e
i n f o
Article history: Received 10 October 2008 Accepted 31 March 2009 Keywords: Selenium Selenomethionine Selenite Selenate Selenoprotein P Milk Pregnancy Lactation
a b s t r a c t Aims: This study compares the transfer from mother to fetuses and pups of selenium (Se) in the form of selenite, selenate, and selenomethionine (SeMet) labeled with different homo-elemental isotopes. Main methods: To completely substitute endogenous Se with natural abundance with Se enriched with a single stable isotope (82Se), female Wistar rats delivered by mother fed 82Se-selenite were fed Se-deficient diet and drinking water containing 82Se-selenite immediately after weaning, and then mated with male Wistar rat at the age of 15–17 weeks. The pregnant rats were divided into two groups. One group was fed Sedeficient diet and drinking water containing 76Se-selenite, 78Se-selenate, and 77Se-SeMet from gestation days 11 to 20. The other group was fed the same diet and drinking water containing the three Se species after delivery for 10 days of lactation. Non-pregnant rats were also fed Se mixture and Se-deficient diet for 10 days. Key finding: Tissue and plasma Se concentrations showed significant changes among non-pregnant, pregnant, and lactating rats. The peak corresponding to selenoprotein P (Sel P) in serum of pregnant rats was reduced. The concentration of 77Se originating from SeMet was higher than those of 76Se from selenite and 78Se from selenate in the stomach content of pups. Significance: Inorganic Se species are more preferably transformed into Sel P than SeMet, and Sel P is effectively incorporated into placenta during pregnancy. On the other hand, SeMet is a more efficient Se source than inorganic Se species during lactation. © 2009 Elsevier Inc. All rights reserved.
Introduction Selenium (Se) is an essential trace element for animals including humans, and exists in the active center of selenoenzymes as a form of selenocysteine (SeCys). Naturally occurring Se species, i.e., inorganic Se (selenite and selenate) and selenoamino acids (SeCys and selenomethionine (SeMet)), are utilized as a nutritional source. Even though all of them are transformed into the assumed common intermediate (selenide) that is utilized for the synthesis of selenoproteins (selenoenzymes), or transformed into methylated metabolites such as selenosugar (Se-sugar) and trimethylselenonium (TMSe) for excretion (Birringer et al. 2002; Ip 1998; Suzuki 2005), it is surmised that their availability and metabolic pathway differ (Suzuki et al. 2006a). Selenite and selenate are directly reduced to the assumed intermediate selenide (Suzuki 2005), whereas selenoamino acids are transformed into selenide through several routes. For
⁎ Corresponding author. Graduate School of Pharmaceutical Sciences, Chiba University, 1-8-1 Chuo, Chiba 260-8675, Japan. Tel./fax: +81 43 226 2866. E-mail addresses: [email protected], [email protected] (Y. Ogra). 1 Present address: Laboratory of Chemical Toxicology and Environmental Health, Showa Pharmaceutical University, 3-3165 Higashi-Tamagawagakuen, Machida, Tokyo 194-8543, Japan. Tel./fax: +81 42 721 1563. 0024-3205/$ – see front matter © 2009 Elsevier Inc. All rights reserved. doi:10.1016/j.lfs.2009.03.023
instance, SeMet is transformed into selenide both by the β-lyase reaction after the successive trans-selenation reaction to SeCys and directly by the γ-lyase reaction (Okuno et al. 2001, 2005; Birringer et al. 2002; Suzuki 2005). In addition to the Se species, physiological condition may be also an important factor affecting Se metabolism. Indeed, it has been reported that the dose-dependent urinary excretion of Se differed between adult and young rats, probably because sources of the sugar moiety of Se-sugar are more abundant in adult rats than in young rats (Suzuki et al. 2005). Thus, it is expected that not only the ingested Se species but also the biological and physiological conditions of animals affect the availability and distribution of Se. It has been reported that Se requirement increases during pregnancy and lactation due to the increase in requirement for maternal transfer of Se to the fetus and the newborn (Smith and Picciano 1986). The effects of the chemical species of Se on Se concentration and selenoenzyme activities in mother and pup have been reported (Lane et al. 1991; Taylor et al. 2005). However, the precise mechanisms underlying the maternal transfer of Se depending on its chemical species remain unclear. We recently developed a new tracer method involving multiple enriched stable isotopes (Suzuki et al. 2006b). Briefly, endogenous natural abundance isotopes were initially substituted with a single enriched stable isotope in host
Y. Anan et al. / Life Sciences 84 (2009) 888–893
889
from each litter. The liver of fetuses and pups and the stomach content of pups were also collected, and stomach contents from rats in the same litter were pooled and treated as one sample. Determination of Se concentration in tissues and body fluids
Fig. 1. Schedule of breeding and dosing for non-pregnant, pregnant and lactating groups.
animal, and then multiple precursors labeled with different enriched stable isotopes were administered simultaneously to the animal. In the present study, we used rats whose endogenous Se with natural abundance was substituted with Se enriched with a stable isotope (82Se) by feeding Se-deficient diet and water containing 82Se-enriched selenite. Then, 76Se-selenate, 78Se-selenite, and 77Se-SeMet were administered simultaneously to rats at different reproductive stages, i.e., non-pregnancy, pregnancy, and lactation. 76Se, 78Se, and 77Se in tissues and body fluids were determined to be of selenate, selenite, and SeMet origin, respectively. Thus, this method makes it possible to clarify the effects of Se species on maternal transfer during pregnancy and lactation under exactly identical animal and analytical conditions.
78
Se- and
Speciation of Se in serum The HPLC system consisted of an online degasser, an HPLC pump (PU713; GL Science Co., Ltd., Tokyo), a Rheodyne six-port injector with a sample loop, and a column. A multi-mode gel filtration column, Shodex Asahipak GS-520 7G (7.5 i.d. × 500 mm, with a guard column, 7.5 i.d. × 75 mm; Showa Denko, Tokyo), was used for the separation of serum. The column was injected with a 200 µl aliquot of serum, and then eluted with 50 mM Tris–HCl, pH 7.4, at a flow rate of 1.0 ml/min. The eluate was introduced directly into a Babington nebulizer of the ICP-MS. Statistical analysis
Materials and methods Preparation of
A 200 mg or 200 µl portion of liver, kidney, muscle, plasma, pancreas, placenta, or stomach content was digested with 1 ml of mixed acid of concentrated nitric acid (HNO3) and 30% H2O2 (v/v = 1/1) in a test tube at room temperature for more than 1 day, then heated at 160–180 °C for several days until complete digestion was achieved. Se concentrations in samples were determined with an inductively coupled plasma-mass spectrometry (ICP-MS; Agilent 7500ce, Agilent Technologies, Hachiouji, Japan).
82
Se-selenite,
76
Se-selenate, and
77
Se-SeMet
Elemental forms of 76Se (97.0% enriched), 77Se (99.9%), 78Se (99.0%), and 82Se (98.9%) were purchased from Isoflex USA (San Francisco, CA, USA). 78Se- and 82Se-selenite were prepared by dissolving each elemental form in nitric acid, followed by adjustment to neutral pH with NaOH. 76Se-selenate was prepared by dissolving elemental 76Se in nitric acid, followed by oxidation with hydrogen peroxide (Kobayashi et al. 2001). 77Se-SeMet was synthesized from elemental 77Se and (S)-(−)-α-amino-γ-butyrolactone in three steps (Plenevaux et al. 1987). Animal experiments Twelve-week-old Wistar rats at gestation day 2 were purchased from Clea Japan Inc. (Tokyo, Japan), maintained in plastic cages at 22± 2 °C with a 12/12 h light/dark cycle, and fed Se-deficient diet (Oriental Yeast Co., Ltd.; Se concentration, b0.02 µg/g diet) and drinking water containing 82Se-selenite at a concentration of 0.2 µg Se/ml ad libitum until pups were weaned. To increase substitution efficiency, feeding the same diet and water was continued up to 15–17 weeks of age. Female rats were assigned to one of three groups, i.e., non-pregnant (n = 3), pregnant (n = 5), and lactating (n = 5) groups. The experimental schedule of breeding and feeding for each group was shown in Fig. 1. Rats of the non-pregnant group were fed Se-deficient diet and drinking water containing 76Se-selenate, 78Se-selenite, and 77Se-SeMet each at 0.1 µg Se/ml ad libitum for 10 days. Rats of pregnant and lactating groups were mated with male Wistar rats purchased from Clea Japan Inc. Rats of the pregnant group were fed the same diet and water as rats of the nonpregnant group ad libitum from gestation days 11 to 20. Rats of the lactating group were fed water containing 82Se-selenite and Se-deficient diet during pregnancy, and fed the same diet and water as rats of the non-pregnant group for 10 days after delivery. At the end of feeding, the rats including pups of the lactating group were sacrificed by exsanguination under ether anesthesia. Heparinized and non-heparinized blood, liver, kidney, muscle, pancreas, and placenta (only from the pregnant group) were collected. Three fetuses and pups were randomly selected
All statistical analyses were performed with StatView 4.51. Differences among three groups were tested using Scheffe's method along with one-factor ANOVA. A p value less than 0.05 was considered to be statistically significant. Results Changes in Se concentration with reproductive stage Total Se concentration was significantly changed in tissues and plasma of pregnant and lactating rats, except kidney (Fig. 2). Compared to the non-pregnant group, Se concentration in the liver was decreased in the pregnant group and further decreased in the lactating group. Plasma Se concentration was significantly lower in the pregnant group than in the lactating and non-pregnant groups. On the other hand, in the muscle and pancreas, Se concentration was significantly increased in the lactating group compared to the other
Fig. 2. Total Se concentrations in the liver, kidney, plasma, muscle, and pancreas of nonpregnant, pregnant, and lactating female rats fed Se-deficient diet and drinking water containing 76Se-selenate, 78Se-selenite, and 77Se-SeMet each at 0.1 µg Se/ml for 10 days ad libitum. Values with different superscript letters (a, b or c) are statistically significantly different among non-pregnant, pregnant, and lactating groups (p b 0.05).
890
Y. Anan et al. / Life Sciences 84 (2009) 888–893
Table 1 Percentage of Se originating from different Se sources with respect to total Se concentration in tissues and plasma of non-pregnant, pregnant, and lactating rats.
Liver
Kidney
Plasma
Muscle
Pancreas
Group
Selenate origin
Selenite origin
SeMet origin
Endogenous Se
NP P L NP P L NP P L NP P L NP P L
20.6 ± 1.7a 20.1 ± 0.9ab 20.4 ± 1.7a 18.5 ± 1.6a 18.2 ± 0.5a 18.2 ± 0.9a 24.9 ± 1.4a 23.4 ± 0.8a 25.9 ± 0.2a 6.4 ± 0.9a 4.6 ± 0.7a 5.6 ± 0.8a 12.7 ± 0.4a 10.3 ± 1.1a 7.7 ± 1.0a
22.4 ± 1.2a 21.7 ± 1.2a 22.7 ± 2.2a 20.9 ± 2.1a 19.7 ± 0.5a 20.0 ± 1.4a 27.7 ± 1.5a 25.9 ± 0.2a 29.7 ± 0.2b 5.2 ± 0.8a 3.2 ± 1.1a 4.3 ± 1.4a 12.5 ± 0.8a 9.4 ± 0.9a 7.4 ± 1.1a
15.0 ± 1.7b 17.7 ± 2.3b 18.8 ± 1.9a 13.3 ± 2.2a 15.5 ± 0.7b 15.0 ± 0.9b 16.6 ± 2.8b 19.2 ± 1.4b 19.6 ± 0.3c 11.7 ± 1.2b 12.3 ± 1.1b 10.1 ± 1.3b 33.0 ± 5.8b 44.6 ± 5.1b 58.2 ± 3.7b
42.0 ± 4.1 40.5 ± 1.2 38.0 ± 3.8 47.3 ± 3.3 46.7 ± 0.6 46.9 ± 3.0 30.9 ± 3.1 31.6 ± 1.3 24.9 ± 0.8 76.7 ± 2.6 79.9 ± 1.7 80.0 ± 3.4 41.8 ± 5.8 35.8 ± 3.2 26.7 ± 1.6
NP: non-pregnant rats, P: pregnant rats, and L: lactating rats. Values with different superscript letters (a, b or c) are statistically significantly different among selenate origin, selenite origin, and SeMet origin in each group (p b 0.05).
groups. Se concentration in the kidney was higher than those in other tissues and plasma, and there were no significant differences among the three groups. Comparison of availability of three Se species among non-pregnant, pregnant, and lactating rats The availability, i.e., incorporation of selenate, selenite, and SeMet into tissues and plasma, was not significantly different among nonpregnant, pregnant, and lactating groups (Table 1), suggesting that pregnancy and lactation hardly affect the availability of these
selenocompounds in the mother's body. Although no significant differences in availabilities of selenate and selenite were observed in all tissues of the three groups, the availability of selenite tended to be slightly higher than that of selenate in the liver, kidney, and plasma. Incorporation of SeMet was significantly lower than those of selenate and selenite in the liver, kidney, and plasma, while SeMet was taken up more efficiently than selenite and selenate in the pancreas and muscle of all groups. Fig. 3 shows changes in the concentration of Se originating from the three Se compounds (selenate, selenite, and SeMet) in three different reproductive stages. Changes in the concentration of Se originating from selenite and selenate well reflected the changes in total Se concentration (Fig. 2) in the liver and plasma. The concentration of Se originating from SeMet in the pancreas was markedly increased by pregnancy and lactation, whereas the concentrations of Se from selenate and selenite were not affected. Speciation of Se in serum using HPLC-ICP-MS Two peaks corresponding extracellular glutathione peroxidase (eGPx) and selenoprotein P (Sel P) were detected at retention times of 10.8 min and 13.0 min at m/z 76, 77, and 78 corresponding to selenate, SeMet, and selenite origins, respectively, in serum of non-pregnant, pregnant, and lactating rats (Fig. 4). Although the elution profiles of 76 Se, 78Se, and 77Se showed similar patterns, the peak heights of eGPx and Sel P at m/z 77 were smaller than those at m/z 76 and 78, suggesting that SeMet was less effectively transformed into these selenoproteins than selenite and selenate. Although the amount of eGPx was comparable among the three different reproductive groups, a marked reduction in the peak corresponding to Sel P was observed in pregnant rats. This result is consistent with the decrease in the plasma total Se concentration of pregnant rats (Fig. 2).
Fig. 3. Changes in Se concentration originating from three Se species, selenate, selenite and SeMet, in the liver (a), kidney (b), plasma (c), muscle (d), and pancreas (e) of nonpregnant, pregnant and lactating female rats fed Se-deficient diet and drinking water containing 76Se-selenate, 78Se-selenite, and 77Se-SeMet each at 0.1 µg Se/ml for 10 days ad libitum. Values with different superscript letters (a, b or c) are statistically significantly different among non-pregnant, pregnant, and lactating groups (p b 0.05).
Y. Anan et al. / Life Sciences 84 (2009) 888–893
891
Fig. 4. Elution profiles of Se isotopes in serum of non-pregnant, pregnant, and lactating female rats fed Se-deficient diet and drinking water containing 76Se-selenate, 78Se-selenite, and 77Se-SeMet each at 0.1 µg Se/ml for 10 days ad libitum.
Transfer of Se through placenta and milk to fetus and pups Se concentration in the fetal liver was comparable to that in the placenta (Table 2). There was no significant difference among the percentage of Se originating from the three Se species in the placenta and fetal liver (Table 2), suggesting that selenite, selenate, and SeMet were homogenously distributed to fetal liver through placenta. Se concentration in mother's milk was estimated by determining Se concentration in stomach content of pups, as stomach content consisted of only coagulated mother's milk. Se concentration in stomach content was 0.13 ± 0.03 µg/mg. The concentration and percentage of Se originating from inorganic species in the liver and stomach content of pups were significantly lower than those of Se originating from SeMet (Table 2). The concentration of Se originating from selenite in stomach content was below the detection limit. These results indicate that SeMet is a better source of Se than selenite and selenate in mother's milk. The percentages of endogenous Se in the stomach content and liver of pups were larger than those of exogenous Se, i.e., selenite, selenate and SeMet, implying that the major part of mother's milk was mobilized from the Se contained in maternal tissues rather than from the dietary Se.
pregnant rats and lactating dams were Se-deficient because Se was transferred to their fetuses and pups, respectively. The reduction in the concentration of liver Se originating from inorganic Se during pregnancy was more severe than that originating from SeMet (Fig. 3a). Inorganic Se, i.e., selenite and selenate, can be directly transformed into selenide, a critical metabolic intermediate of Se in the body, by simple reduction, and selenide is utilized for the biosynthesis of selenoproteins (Suzuki 2005). On the other hand, SeMet is used for the biosynthesis of selenoproteins after degradation to selenide via several metabolic pathways, and is also incorporated into general proteins in place of Met to produce the so-called Se-containing proteins (Suzuki 2005). Therefore, the amounts of selenite and selenate utilized for the biosynthesis of selenoproteins may be larger than that of SeMet, and thus, Se originating from selenite and selenate contained in selenoproteins may be more efficiently transferred to fetus accompanied by the transfer of selenoproteins, resulting in the reduction of those compounds in the dam's body. On the other hand, the reduction of liver Se was not different among selenate, selenite and SeMet origins in the lactating group (Fig. 3a), suggesting that Se-containing protein is also transferred to pups as well as selenoproteins during lactation. Se concentrations in the kidney and muscle were not sensitive to pregnancy, lactation, and ingested chemical species of Se (Fig. 3b and d). eGPx, a protein that acts to protect against
Discussion In the present study, the availability and efficiency of maternal transfer of three Se species were compared using a new tracer method. The same experimental conditions, including animals, sample preparation, and analytical conditions, were achieved by simultaneous administration of multiple selenocompounds labeled with compound-specific stable isotopes. Thus, this new method enabled more accurate comparison and quantitative evaluation of maternal transfer depending on chemical species than experiments using a single tracer in different animals, and was essential to achieving the aim of this study. Se concentration in the liver of pregnant and lactating rats was significantly reduced compared with that of non-pregnant rats (Fig. 2). In particular, Se concentration in the liver of lactating dams decreased by approximately 50% compared to that of non-pregnant rats. It is known that rat requires 1.5 times greater amount of Se during gestation (Smith and Picciano 1986). In this study, Se concentration in drinking water was adjusted to 0.3 µg/ml, which corresponded to the adequate level for non-pregnant rats. The results indicated that
Table 2 Concentrations of Se originating from different sources in the placenta and fetal liver of pregnant rats and liver and stomach contents of pups. Selenate origin Selenite origin SeMet origin Endogenous Se Total Pregnant rats Placenta 56 ± 11a (21.5 ± 1.3) Fetal liver 47 ± 4a (20.1 ± 1.2)
58 ± 11a (22.3 ± 1.0) 45 ± 5a (19.1 ± 2.0)
50 ± 6a (19.5 ± 2.0) 51 ± 5a (21.6 ± 2.0)
96 ± 15 (36.8 ± 1.1) 92 ± 11 (39.3 ± 4.1)
260±40 (100) 234±12 (100)
Pups of lactating group Liver 13 ± 8a (5.0 ± 2.6) Stomach 4 ± 1a content (3.2 ± 1.2)
7 ± 7a (2.8 ± 2.5) Not detectable (0.0)
43 ± 12b (17.8 ± 1.6) 23 ± 12b (17.7 ± 9.4)
177 ± 26 (74.3 ± 5.6) 100 ± 26 (79.1 ± 10.1)
241 ± 47 (100) 127 ± 31 (100)
Se concentration is expressed as ng Se/g tissue. Numbers in parentheses indicate percentage. Values with different superscript letters (a or b) are statistically significantly different among selenate origin, selenite origin, and SeMet origin in tissues (p b 0.05).
892
Y. Anan et al. / Life Sciences 84 (2009) 888–893
Fig. 5. Proposed maternal transfer pathways for selenate, selenite, and SeMet during pregnancy (a) and lactation (b). SeMet: selenomethionine, End Se: endogenous Se, Exo Se: exogenous Se.
oxidative stress in extracellular fluid and bloodstream, is biosynthesized in the kidney (Avissar et al. 1994), and the major part of Se in the kidney is recovered in the insoluble fraction (Hawkes et al. 1985). Therefore, Se might be pooled in the kidney to maintain the concentration of an important selenoprotein, eGPx, during pregnancy and lactation. Indeed, although the concentration of Sel P, a plasma selenoprotein biosynthesized in the liver, was affected by pregnancy, that of eGPx was stable (Fig. 4). The ratio of endogenous Se to total Se in the muscle was the highest among the tissues and plasma examined (Table 1), suggesting that the substitution rate of exogenous Se with endogenous Se was the lowest. Thus, the muscle does not seem to contribute much as an Se source to placenta and milk. Recently, a novel function of ApoER2, a specific receptor of Sel P expressed on cell surface, was reported (Burk et al. 2007; Olson et al. 2007, 2008). ApoER2 was predominantly expressed in the brain and testis. In addition, ApoER2 mRNA expression was observed in the placenta (Kim et al. 1996). These suggest that Sel P acts as an Se transporter to fetus via placenta, resulting in a reduction in the concentration of Sel P in pregnant rats. This coincides with the more severe reduction of liver Se originating in inorganic Se than in SeMet in pregnant rats because inorganic Se, selenite and selenate, is more efficiently assimilated into Sel P than SeMet (Fig. 4). Although the percentages of Se originating from selenate and selenite were higher than that from SeMet in the dam's plasma (Table 1), the percentages of Se originating from the three Se species were not significantly different in the placenta and fetal liver (Table 2). This suggests that SeMet itself is also transferred to fetus without incorporation into Sel P, via an SeMet-specific pathway. It is known that SeMet is preferably incorporated into the pancreas whereas inorganic Se species are hardly incorporated (Suzuki et al. 2006b). Indeed, the preferable incorporation of SeMet into the pancreas was observed (Table 1). Moreover, the concentrations of Se originating from SeMet in the pancreas of pregnant and lactating rats were higher than that of non-pregnant rats (Fig. 3e). Although an SeMet-assimilating protein is suggested to be present in the pancreas, it has not been identified yet (Suzuki et al. 2006b). Insulin and other peptide hormones may be candidates because pancreatic function, such as insulin secretion, is expected to be augmented during late pregnancy and lactation (Bell and Bauman 1997). Further studies are needed to identify the SeMet-assimilating protein in the pancreas.
The chemical species of Se in milk are not fully known yet. However, recently, SeCys, in addition to SeMet, was detected after digestion of milk proteins and derivatization of selenol group (–SeH) (Bierla et al. 2008). This suggests that there are some selenoproteins in milk, and all Se species are possible sources of them. On the other hand, in stomach content, the concentration of Se originating from SeMet was higher than that of Se originating from inorganic Se species (Table 2). This is in agreement with the previous observation that feeding SeMet to livestock, such as cows and pigs, more efficiently increases Se concentration in milk than feeding inorganic Se species (Yoon and McMillan 2006; Muñiz-Naveiro et al. 2006). Muñiz-Naveiro et al. (2007) suggested that SeMet is incorporated into general milk proteins in place of Met. Indeed, the concentration of Se originating from SeMet in the liver of pups was higher than that of Se from inorganic Se species (Table 2). Taken together, SeMet was more effectively transferred to pups via milk than inorganic Se species due to its incorporation into general milk proteins. There is no information of low molecular weight Se species in milk other than selenoproteins and Se-containing proteins. Such low molecular weight Se species as Se-sugars and methylated Se compounds may actually exist in extracellular fluid. Further studies, in particular, speciation studies, are needed. In conclusion, total Se concentrations in tissues and plasma except kidney were affected by gestation and lactation, and as proposed in Fig. 5, inorganic Se and SeMet were transferred to fetuses and pups via different routes. Inorganic Se species, such as selenite and selenate, are more preferably transformed into Sel P than SeMet, and Sel P is effectively incorporated into placenta via its specific receptor, ApoER2, during pregnancy. On the other hand, SeMet is a more efficient Se source than inorganic Se species during lactation because SeMet is incorporated into not only selenoproteins as well as inorganic Se species but also general milk proteins in place of Met. This indicates that SeMet is a better nutritional source of Se for pups than inorganic Se. Acknowledgments The authors wish to acknowledge a Grant-in-Aid for JSPS Fellows from Japan Society for the Promotion of Science (JSPS) (No. 06J01979), and a Grant-in-Aid from the Ministry of Education, Culture, Sports, Science and Technology, Japan (No. 19390033).
Y. Anan et al. / Life Sciences 84 (2009) 888–893
References Avissar N, Ornt DB, Yagil Y, Horowitz S, Watkins RH, Kerl EA, Takahashi K, Palmer IS, Cohen HJ. Human kidney proximal tubules are the main source of plasma glutathione peroxidase. American Journal of Physiology. Cell Physiology 266 (2 35–2), C367–C375, 1994. Bell AW, Bauman DE. Adaptations of glucose metabolism during pregnancy and lactation. Journal of Mammary Gland Biology and Neoplasia 2 (3), 265–278, 1997. Bierla K, Szpunar J, Lobinski R. Specific determination of selenoaminoacids in whole milk by 2D size-exclusion-ion-paring reversed phase high-performance liquid chromatography-inductively coupled plasma mass spectrometry (HPLC-ICP MS). Analytica Chimica Acta 624 (2), 195–202, 2008. Birringer M, Pilawa S, Flohe L. Trends in selenium biochemistry. Natural Product Reports 19 (6), 693–718, 2002. Burk RF, Hill KE, Olson GE, Weeber EJ, Motley AK, Winfrey VP, Austin LM. Deletion of apolipoprotein E receptor-2 in mice lowers brain selenium and causes severe neurological dysfunction and death when a low-selenium diet is fed. Journal of Neuroscience 27 (23), 6207–6211, 2007. Hawkes WC, Wilhelmsen EC, Tappel AL. Subcellular distribution of selenium-containing proteins in the rat. Journal of Inorganic Biochemistry 25 (2), 77–93, 1985. Ip C. Lessons from basic research in selenium and cancer prevention. Journal of Nutrition 128 (11), 1845–1854, 1998. Kim DH, Iijima H, Goto K, Sakai J, Ishii H, Kim HJ, Suzuki H, Kondo H, Saeki S, Yamamoto T. Human apolipoprotein E receptor 2. A novel lipoprotein receptor of the low density lipoprotein receptor family predominantly expressed in brain. Journal of Biological Chemistry 271 (14), 8373–8380, 1996. Kobayashi Y, Ogra Y, Suzuki KT. Speciation and metabolism of selenium injected with 82Se-enriched selenite and selenate in rats. Journal of Chromatography. B, Biomedical Sciences and Applications 760 (1), 73–81, 2001. Lane HW, Strength R, Johnson J, White M. Effect of chemical form of selenium on tissue glutathione peroxidase activity in developing rats. Journal of Nutrition 121 (1), 80–86, 1991. Muñiz-Naveiro Ó, Domínguez-González R, Bermejo-Barrera A, Bermejo-Barrera P, Cocho JA, Fraga JM. Study of the bioavailability of selenium in cows' milk after a supplementation of cow feed with different forms of selenium. Analytical and Bioanalytical Chemistry 385 (1), 189–196, 2006. Muñiz-Naveiro Ó, Domínguez-González R, Bermejo-Barrera A, Bermejo-Barrera P, Cocho JA, Fraga JM. Selenium speciation in cow milk obtained after supplementa-
893
tion with different selenium forms to the cow feed using liquid chromatography coupled with hydride generation-atomic fluorescence spectrometry. Talanta 71 (4), 1587–1593, 2007. Okuno T, Kubota T, Kuroda T, Ueno H, Nakamuro K. Contribution of enzymic alpha, gamma-elimination reaction in detoxification pathway of selenomethionine in mouse liver. Toxicology and Applied Pharmacology 176 (1), 18–23, 2001. Okuno T, Motobayashi S, Ueno H, Nakamuro K. Purification and characterization of mouse hepatic enzyme that converts selenomethionine to methylselenol by its alpha, gamma-elimination. Biological Trace Element Research 106 (1), 77–94, 2005. Olson GE, Winfrey VP, Nagdas SK, Hill KE, Burk RF. Apolipoprotein E receptor-2 (ApoER2) mediates selenium uptake from selenoprotein P by the mouse testis. Journal of Biological Chemistry 282 (16), 12290–12297, 2007. Olson GE, Winfrey VP, Hill KE, Burk RF. Megalin mediates selenoprotein P uptake by kidney proximal tubule epithelial cells. Journal of Biological Chemistry 283 (11), 6854–6860, 2008. Plenevaux A, Cantineau R, Guillaume M. Fast chemical synthesis of [75Se]L-selenomethionine. Applied Radiation and Isotopes 38 (1), 59–61, 1987. Smith AM, Picciano MF. Evidence for increased selenium requirement for the rat during pregnancy and lactation. Journal of Nutrition 116 (6), 1068–1079, 1986. Suzuki KT. Metabolomics of selenium: Se metabolites based on speciation studies. Journal of Health Science 51 (2), 107–114, 2005. Suzuki KT, Kurasaki K, Okazaki N, Ogra Y. Selenosugar and trimethylselenonium among urinary Se metabolites: dose- and age-related changes. Toxicology and Applied Pharmacology 206 (1), 1–8, 2005. Suzuki KT, Doi C, Suzuki N. Metabolism of 76Se-methylselenocysteine compared with that of 77Se-selenomethionine and 82Se-selenite. Toxicology and Applied Pharmacology 217 (2), 185–195, 2006. Suzuki KT, Somekawa L, Kurasaki K, Suzuki N. Simultaneous tracing of 76Se-selenite and 77 Se-selenomethionine by absolute labeling and speciation. Toxicology and Applied Pharmacology 217 (1), 43–50, 2006. Taylor JB, Finley JW, Caton JS. Effect of the chemical form of supranutritional selenium on selenium load and selenoprotein activities in virgin, pregnant, and lactating rats. Journal of Animal Science 83 (2), 422–429, 2005. Yoon I, McMillan E. Comparative effects of organic and inorganic selenium on selenium transfer from sows to nursing pigs. Journal of Animal Science 84 (7), 1729–1733, 2006.