Assessment of selenium bioavailability from naturally produced high-selenium soy foods in selenium-deficient rats

Journal of Trace Elements in Medicine and Biology 24 (2010) 223–229

Contents lists available at ScienceDirect

Journal of Trace Elements in Medicine and Biology journal homepage: www.elsevier.de/jtemb

Bioavailability

Assessment of selenium bioavailability from naturally produced high-selenium soy foods in selenium-deficient rats夽,夽夽 Lin Yan ∗ , Philip G. Reeves 1 , LuAnn K. Johnson USDA, ARS, Grand Forks Human Nutrition Research Center, 2420 2nd Avenue North, Grand Forks, ND 58202, United States

a r t i c l e

i n f o

Article history: Received 19 October 2009 Accepted 13 April 2010 Keywords: Selenium Bioavailability Soybean Glutathione peroxidase Thioredoxin reductase Rats

a b s t r a c t We assessed the bioavailability of selenium (Se) from a protein isolate and tofu (bean curd) prepared from naturally produced high-Se soybeans. The Se concentrations of the soybeans, the protein isolate and tofu were 5.2 ± 0.2, 11.4 ± 0.1 and 7.4 ± 0.1 mg/kg, respectively. Male weanling Sprague–Dawley rats were depleted of Se by feeding them a 30% Torula yeast-based diet (4.1 ␮g Se/kg) for 56 days, and then they were replenished with Se for an additional 50 days by feeding them the same diet containing 14, 24 or 30 ␮g Se/kg from the protein isolate or 13, 23 or 31 ␮g Se/kg from tofu, respectively. l-Selenomethionine (SeMet) was used as a reference. Selenium bioavailability was determined on the basis of the restoration of Se-dependent enzyme activities and tissue Se concentrations in Se-depleted rats, comparing those responses for the protein isolate and tofu to those for SeMet by using a slope-ratio method. Dietary supplementation with the protein isolate or tofu resulted in linear or log-linear, dose-dependent increases in glutathione peroxidase activities in blood and liver and in thioredoxin reductase activity in liver. Furthermore, supplementation with the protein isolate or tofu resulted in linear or log-linear, dosedependent increases in the Se concentrations of plasma, liver, muscle and kidneys. These results indicated an overall bioavailability of approximately 101% for Se from the protein isolate and 94% from tofu, relative to SeMet. We conclude that Se from naturally produced high-Se soybeans is highly bioavailable in this model and that high-Se soybeans may be a good dietary source of Se. Published by Elsevier GmbH.

Introduction The nutritional essentiality of selenium (Se) was first reported in 1957 when it was shown to prevent diet-induced liver necrosis in laboratory animals [1]. Later, Se was found to be an integral part of the catalytic site of several enzymes, including glutathione peroxidase (GPX) [2] and thioredoxin reductase (TRR) [3]. An adequate intake of Se prevents Se-deficient diseases in humans [4] and in livestock [5], and dietary supplementation with Se is associated with many potential health benefits for humans, including an enhancement in immune responses [6], an improvement of thyroid health [7] and a reduction in cancer risk [8].

夽 Supported by the U.S. Department of Agriculture, ARS, research project 545051000-035-00D. 夽夽 The U.S. Department of Agriculture, Agricultural Research Service, Northern Plains Area is an equal opportunity/affirmative action employer and all agency services are available without discrimination. Mention of trade names or commercial products in this article is solely for providing specific information and does not imply recommendation or endorsement by the U.S. Department of Agriculture. ∗ Corresponding author. Tel.: +1 701 795 8499; fax: +1 701 795 8220. E-mail addresses: [email protected] (L. Yan), [email protected] (L.K. Johnson). 1 Deceased. 0946-672X/$ – see front matter. Published by Elsevier GmbH. doi:10.1016/j.jtemb.2010.04.002

Selenium from dietary sources can be metabolized through different pathways, which determine its functions and destination in the body. There are two major Se compartments, a selenocysteine (SeCys) compartment comprised a relatively small number of proteins containing Se incorporated as SeCys by a highly specific co-translational process [9–11], and a non-specific SeMet compartment comprised general proteins in which SeMet is incorporated as a mimic of its sulfur-analog methionine. The amount of Se in the SeCys compartment is regulated by the availability of Se at low levels of intake. The SeMet compartment is relatively large, potentially including all sites of methionine in body proteins. The non-specific incorporation of SeMet is not specifically regulated. Selenomethionine is a dominant form of Se in plant foods [12–14], which upon its absorption can be non-specifically incorporated into the SeMet compartment or metabolized via enzymatic or nonenzymatic steps through selenide, an intermediate metabolite in Se metabolism, to enter the SeCys compartment or be methylated and excreted in urine. Soy has been a major source of dietary protein for humans for centuries. Consumption of foods of soy origin has been associated with improved cardiovascular health [15] and reduced cancer risk [16]. The United States is one of the largest soy producers and exporters in the world, and the Northern Plains states where Se is rich in soils [17] are among the major soy producers of the country.

224

L. Yan et al. / Journal of Trace Elements in Medicine and Biology 24 (2010) 223–229 Table 1 Composition of the basal dieta .

The Se contents of soybeans reflect the Se contents of the producing soils, which vary widely among the major soy producing states. Soybeans produced in Northern Plains have greater Se contents than those produced elsewhere. The soybeans we obtained from South Dakota, a Northern Plains state, contained 5.2 mg Se/kg, much higher than the average of 0.2 mg Se/kg in soybeans produced overall in the United States [18]. However, the nutritional value and potential health benefit of high-Se soybeans have not been assessed. The purpose of the present study was to determine the bioavailability of Se from two food products prepared from naturally produced high-Se soybeans, soy protein isolate and tofu (bean curd). Soy protein isolate has wide applications in food industry, and tofu is one of the most commonly consumed traditional soy foods.

The study was approved by the Animal Care and Use Committee of the USDA-ARS Grand Forks Human Nutrition Research Center. The procedures followed the guidelines of the National Institute of Health for the experimental use of laboratory animals [19]. High-selenium soybeans Soybeans (Dekalb 2551) were obtained from Karlen Ranch, Lyman, South Dakota. The raw soybeans were taken from the bulk combine harvest in 2007; the beans contained 5.2 ± 0.2 mg Se/kg (n = 3). Preparation of soy protein isolate and tofu

Diet preparation The Se-deficient basal diet was prepared based on the AIN-93G formulation [20] with the protein source being 30% Torula yeast (Table 1). Torula yeast is relatively rich in Fe, P, K, Zn, Mn and Mg [21], so a diet containing 30% yeast has adequate amounts of these minerals for rodents. Thus, a mineral mix (Table 1) was prepared containing only those minerals needed to meet National Research Council (NRC) recommendations [22] and those included in the AIN-93G formulation [20]. A Se-adequate diet, according to the AIN-93G formulation [20], was prepared as a positive control by adding SeMet to the basal diet to bring the total Se concentration to 150 ␮g Se/kg. As we proposed that Se would be the active

g/kg

Corn starch Torula yeast Sucrose Soybean oil Cellulose Mineral mixb AIN-93G vitamin mix l-Methionine l-Cystine l-Tryptophan Choline bitartrate

427.4 300.0 100.0 70.0 50.0 35.0 10.0 2.9 3.5 0.16 1.0

a At 30% of the diet, Torula yeast (21) provided adequate amounts of all essential amino acids for rodents except cystine (1.8 g/kg), methionine (2.4 g/kg) and tryptophan (1.84 g/kg) based on NRC recommendations for cystine (4.9 g/kg), methionine (4.9 g/kg) and tryptophan (2 g/kg) (22). Thus, these amino acids were added to the diet to meet the recommendations. At 30% of the diet, the yeast provided 0.9 g/kg choline, we added additional 0.4 g choline/kg (1.0 g choline bitartrate/kg) to the diet to meet that provided by the AIN-93G formulation (20). b Mineral mix contained calcium carbonate, anhydrous, 40.04% Ca, 555.26 g/kg; sodium chloride, 39.34% Na, 52.17 g/kg; sodium meta-silicate, 9 hydrate, 9.88% Si, 1.45 g/kg; chromium potassium sulfate, 12 hydrate, 10.42% Cr, 0.275 g/kg; copper carbonate, 57.47% Cu, 0.143 g/kg; boric acid, 17.5% B, 0.082 g/kg; sodium fluoride, 45.24% F, 0.064 g/kg; nickel carbonate, 45% Ni, 0.032 g/kg; lithium chloride, 16.38% Li, 0.017 g/kg; potassium iodate, 59.3% I, 0.010 g/kg; ammonium paramolybdate, 4 hydrate, 54.34% Mo, 0.008 g/kg; ammonium vanadate, 43.55% V, 0.007 g/kg; powdered sucrose 390.482 g/kg.

Materials and methods

Soybeans were finely ground to flour and defatted by using hexane at room temperature. After removal of hexane by evaporation, the defatted flour was mixed with water (1:10, w/w) at 32 ◦ C for 15 min and centrifuged at 2800 × g for 20 min to collect a soluble extract. The insoluble portion was again mixed with water (1:6, w/w) at 32 ◦ C for 15 min and centrifuged to obtain a second soluble extract. The soluble extracts were combined, adjusted with HCl to pH 4.5 and centrifuged to collect a precipitate. The precipitate was mixed with water (1:6, w/w) at 32 ◦ C for 15 min and centrifuged to obtain a curd. To obtain protein isolate, the curd was resolubilized with water, adjusted with NaOH to pH 7.4, pasteurized at 100 ◦ C for 15 min to denaturize trypsin inhibitors, and then freeze-dried. To make tofu, soybeans were soaked with deionized water overnight, and then mixed with warm water (1:5, w/w) and ground thoroughly into a smooth solution. The solution was cooked at 100 ◦ C for 15 min before filtering through cheesecloth to separate soy pulp from the milk. The milk was treated with CaSO4 to obtain a coagulate that was filtered to remove water, pressed to tofu, and then freeze-dried. The Se concentrations of defatted soy flour, protein isolate and tofu were 7.5 ± 0.1, 11.4 ± 0.1 and 7.4 ± 0.1 mg/kg (n = 5), respectively.

Ingredient

component of high-Se soybeans that would affect the endpoints measured, the protein isolate was added to the basal diet at 1.8, 2.6 or 3.5 g/kg and tofu at 2.7, 4.1 or 5.4 g/kg to provide 20, 30 or 40 ␮g Se/kg diet, respectively. Each mixed diet was analyzed for Se before it was provided to animals, and the results of the analysis are in Table 2. Experimental design The experiment was initiated with 108 male weanling Sprague–Dawley rats (strain: SAS:VAF; Charles River, Wilmington, MA); 96 were fed the basal diet (4.1 ␮g Se/kg) to deplete Se,

Table 2 Selenium content of the experimental diets. Targeted Se concentration (␮g/kg) 30% Torula yeast basal diet Selenium-supplemented diets Selenomethionine

5 10 20 50 70 100 150

Analyzed Se concentration (␮g/kg)a 4.1 ± 1.8 10.9 20.5 57.7 79.1 96.1 140.8

± ± ± ± ± ±

1.3 1.3 2.0 1.2 0.4 13.2

Protein isolate

20 30 40

13.8 ± 1.3 23.8 ± 1.4 30.4 ± 1.0

Tofu

20 30 40

12.9 ± 0.3 22.5 ± 0.2 31.4 ± 0.4

a Values are means ± SD, n = 21 for the basal diet (5 ␮g Se/kg), n = 6 for the Seadequate diet (150 ␮g Se/kg) and n = 3 for each of the other diets.

L. Yan et al. / Journal of Trace Elements in Medicine and Biology 24 (2010) 223–229

and 12 were fed the Se-adequate diet (150 ␮g Se/kg) as positive controls. Rats fed the basal diet for 56 days were considered Sedepleted based on reduced GPX activity in whole blood, compared to that from rats fed the Se-adequate diet. After Se-depletion, 12 rats remained on the basal diet, and the other 84 were assigned to 12 groups and fed the basal diet supplemented with 10, 20, 50, 70, 100 or 150 ␮g Se/kg from SeMet (n = 8), 14, 24 or 30 ␮g Se/kg from the protein isolate (n = 6) or 13, 23 or 31 ␮g Se/kg from tofu (n = 6), respectively. The rats were fed the supplemented diets for 50 days, which resulted in Se repletion based on GPX activity in whole blood of rats supplemented with SeMet. The rats were individually housed in stainless steel cages with wire-mesh bottoms in a room maintained at 50% relative humidity, 22 ◦ C, and a 12-h light/dark cycle. All rats had free access to diet and deionized water, and they were weighed weekly. Food intake was measured on days 74–78 during the Se-repletion period (n = 4 for each group). At the end of the experiment, rats were anesthetized with a mixture of ketamine and xylazine. Blood, liver, gastronemius muscle and kidneys were collected and stored at −80 ◦ C for enzyme and Se analyses. Liver, muscle and kidneys were freeze-dried prior to Se analysis. Enzyme assays Glutathione peroxidase activity was determined in whole blood and liver by the method of Paglia and Valentine [23] as modified by Lawrence and Burk [24] using H2 O2 as the substrate in the presence of azide. The activity in whole blood was expressed as units/mg hemoglobin (Hb) and in liver as units/mg protein; one unit of activity was defined as the amount of enzyme required to oxidize 1.0 ␮mol NADPH/min. Thioredoxin reductase activity was determined in liver by the method of Hill et al. [25] as modified by Hintze et al. [26]. A unit of activity was defined as 1.0 ␮mol thionitrobenzoate formed/(min mg protein). Protein was determined by the Bradford method (BioRad, Hercules, CA). Selenium analysis Samples of diets, plasma and organs were digested by a mixture of nitric acid (16 mol/L), hydrochloric acid (12 mol/L) and magnesium nitrate (hexahydrate solution, 40% in deionized water); the volume of the digestion mixture was adjusted accordingly to the sample weight. The initial digestion was performed by refluxing on a 120 ◦ C hot plate for 24 h. The samples were then dried and ashed in a 490 ◦ C muffle furnace for 12 h. The ash was dissolved in hydrochloric acid (12 mol/L) and brought up to the volume with deionized water. Digested samples were analyzed by hydridegeneration, inductively coupled argon plasma mass spectrometry (PerkinElmer DRCII instrument, PerkinElmer Corp., Wellsley, MA), equipped with automated hydride-generation and flow injection system. Each sample was digested in triplicate. Three isotopes, Se77 , Se78 and Se82 , were determined in each replicate, and three values were obtained for each isotope. The average of these 27 measurements was taken as the final result for each sample analyzed. The method detection limits of Se for the quantification of Se77 , Se78 and Se82 were 1.7, 1.35 and 1.4 ng/ml, respectively. The certified reference standard Durum Flour (Standard Reference Material #8436, National Institute of Standards and Technology, Gaithersburg, MD) was used to ensure the accuracy of the measurements. Our analysis of Durum Flour found 1281 ± 31 ng Se/g, well within the published range for this reference material (1140–1320 ng Se/g). Results of the analysis were expressed as ␮g/kg for diet, ␮mol/L for plasma and ␮mol/kg (dry weight) for liver, muscle and kidneys.

225

Statistical analyses Student’s t-test was used to compare differences between groups fed the basal and the Se-adequate diets. One-way analysis of variance followed by Tukey contrasts was used to test for differences between the Se-deficient group and the groups fed the protein isolate or tofu with different amounts of Se. A slope-ratio model [28] was used to determine the relative bioavailability of Se from the protein isolate and tofu compared to Se from SeMet. Linearity of the respective regression lines was ascertained for each source of Se after which a single multiple regression model was derived to determine the slope and intercept of the responses for the three Se sources: the protein isolate, tofu and SeMet [29]. Confidence limits for relative bioavailability were obtained by Fieller’s method [28]. All data are presented as means ± SD. Differences with P ≤ 0.05 are considered significant. All statistical analyses were performed using SAS version 9.2 (SAS Institute, Inc., Cary, NC).

Results Our targeted levels of Se supplementation were 20, 30 and 40 ␮g Se/kg, but the actual analyzed concentrations of Se were 14, 24 and 30 ␮g/kg with the protein isolate-supplemented diets and 13, 23 and 31 ␮g/kg with the tofu-supplemented diets, respectively (Table 2). This difference between the targeted and analyzed Se concentrations was caused by a variation in Se from the available lots of Torula yeast, which varied from 2.4 ± 0.3 to 48.3 ± 0.6 ␮g Se/kg. Because the analyzed Se concentrations of these diets met the requirements of the slope-ratio model [28] used in this study, (a) graded levels of dietary Se and (b) intakes of Se must not exceed the amount required to fully replenish the response measure, these diets were used for the experiment without further adjustment for Se concentrations. The analyzed Se content of the basal diet was very low (4.1 ± 1.8 ␮g/kg) compared with that of the Se-adequate diet (140.8 ± 13.2 ␮g/kg) (Table 2). Feeding rats this basal diet resulted in growth retardation. After 21 days on the diet, the average weight gain of rats fed the basal diet was 4.9 ± 0.2 g/d compared with rats on the Se-adequate diet (6.1 ± 0.1 g/d; P = 0.001), and this difference remained statistically significant throughout the experiment (Fig. 1). Dietary supplementation with Se from the protein isolate or tofu tended to increase body weight; however, neither increase was statistically significant (Fig. 1). The 5-day average food intake of rats fed the basal diet and the Se-adequate diet during the Se-repletion period was 16.8 ± 0.8 g/d and 17.8 ± 0.6 g/d (P = 0.05), respectively, and dietary supplementation with the protein isolate or tofu did not result in significant changes in food intake (data not shown). Feeding rats the basal diet reduced Se-dependent enzyme activities. At the end of the Se-depletion period, blood GPX activity of the rats fed the basal diet was 6% of the rats fed the Se-adequate diet (34.9 ± 2.8 U/mg Hb vs. 577.1 ± 27.5 U/mg Hb; n = 5), and at the end of the experiment it was only 2.1% of the rats maintained on the Se-adequate diet throughout the experiment (15.6 ± 3.7 U/mg Hb vs. 743.0 ± 100.4 U/mg Hb; n = 8). Dietary supplementation with SeMet resulted in a dose-dependent increase in blood GPX activity (Fig. 2A). At the end of the Se-repletion period, the GPX activity was 236.3 ± 80.6, 421.2 ± 77.7 and 502.8 ± 61.0 U/mg Hb in groups supplemented with 50, 100 and 150 ␮g Se/kg from SeMet, which represented a 31.8, 56.7 and 67.7% restoration of GPX activity compared with the rats fed the Se-adequate diet throughout the experiment. Adding the protein isolate or tofu to the basal diet resulted in a log-linear, dose-dependent increase in GPX activity (Fig. 2A). The slope-ratio model [28] showed that Se bioavailability was 119.4% for the protein isolate and 86.8% for tofu, relative to SeMet (Table 3).

226

L. Yan et al. / Journal of Trace Elements in Medicine and Biology 24 (2010) 223–229

Fig. 1. Body weight changes of rats during the Se-depletion and the Se-repletion periods. Student’s t-test was used to compare the differences between rats fed the basal and the Se-adequate diets throughout the experiment. One-way analysis of variance followed by Tukey contrasts was used to compare differences among groups during the Se-repletion period. Dietary supplementation with the protein isolate or tofu (data not shown) during the Se-repletion period tended to increase body weight, however, neither increase was significant. Mean ± SD, n = 12 for rats fed the basal and the Se-adequate diets and n = 6 for rats fed the protein isolate- or tofu-supplemented diets.

The reduction in liver GPX activity by the Se-depletion was similar to that of blood GPX activity. At the end of the experiment, hepatic GPX activity of the rats fed the basal diet was no greater than 1% of those maintained on the Se-adequate diet (14.7 ± 6.3 U/mg protein vs. 3947.7 ± 1135.4 U/mg protein; n = 8). The enzyme activity responded to dietary SeMet supplementation (Fig. 2B). At the end of the Se-repletion period, the liver GPX activity was 1051.7 ± 269.5, 3273.1 ± 715.0 and 3476.1 ± 619.7 U/mg protein in groups supplemented with 50, 100 and 150 ␮g Se/kg as SeMet, which represented a 26.6, 82.9 and 88.1% restoration of GPX activity compared with the group fed the Se-adequate diet throughout the experiment. Dietary supplementation with Se from the protein isolate or tofu resulted in a log-linear, dose-dependent increase in liver GPX activity (Fig. 2B), indicating that Se bioavailability was 100.8% for the protein isolate and 90.6% for tofu, relative to SeMet (Table 3). Hepatic TRR activity responded to changes in dietary Se (Fig. 2C). At the end of the experiment, the TRR activity of the rats fed the Table 3 The relative biological value (RBV; bioavailability) of selenium from soy protein isolate and tofu compared with selenomethionine in restoring Se-dependent enzyme activity and tissue selenium retention in ratsa , b . % RBV (95% confidence interval) Protein isolate

Tofu

Blood GPX Liver GPX Liver TRR Plasma Se Liver Se Muscle Se Kidney Se

119.4 (108.1, 130.6) 100.8 (88.6, 113.2) 74.6 (50.3, 98.9) 98.6 (89.7, 107.6) 117.2 (106.7, 127.7) 101.6 (88.6, 114.5) 99.5 (92.3, 106.7)

86.8 (71.1, 102.6) 90.6 (79.0, 102.2) 74.2 (47.4, 101.0) 99.3 (91.4, 107.3) 115.1 (104.6, 125.6) 78.4 (64.0, 92.8) 114.9 (106.9, 122.9)

Mean ± SDc

101.7 ± 14.7

a

94.2 ± 16.4

Standard response curve for Se-dependent enzyme activities and tissue Se was made by feeding rats diets containing various amounts of Se as SeMet. The enzyme activities and tissue Se of rats fed the protein isolate- or tofu-supplemented diet were compared with the values on the standard response curve. b The %RBV was estimated by using the slope-ratio method [28] for enzyme activities or the parallel line assay for tissue Se. c The overall mean of each % RBV column.

Fig. 2. Curves showing the response of blood GPX (A), liver GPX (B) and liver TRR activities (C) to dietary supplementation with SeMet (solid circle), the protein isolate (open circle) and tofu (solid diamond). For the slope-ratio analyses, blood and liver GPX activities were log-transformed to achieve linearity. The Se concentrations were 14, 24 and 30 ␮g/kg with the isolate-supplemented diets and 13, 23 and 31 ␮g/kg with the tofu-supplemented diets. Mean ± SD, n = 8 for SeMet supplementation and n = 6 for the protein isolate or tofu supplementation.

basal diet was 34% of that of the rats maintained on the Se-adequate diet (2.4 ± 0.3 U/mg protein vs. 7.9 ± 0.3 U/mg protein; n = 8). At the end of the Se-repletion period, the enzyme activity was 4.7 ± 0.3, 6.3 ± 0.6 and 7.4 ± 0.5 U/mg protein in groups supplemented with 50, 100 and 150 ␮g Se/kg as SeMet, which represented a 59.5, 79.7 and 93.7% restoration of TRR activity, respectively. Dietary supplementation with Se from the protein isolate or tofu resulted in a linear, dose-dependent increase in hepatic TRR activity (Fig. 2C), indicating Se bioavailabilities of 74.6% and 74.2% for the protein isolate and tofu, respectively, relative to SeMet (Table 3). Feeding rats the basal diet resulted in tissue Se-depletion. At the end of the experiment, Se concentrations of plasma, liver, muscle and kidneys of the rats fed the basal diet were 2, 1, 1, and 6%, respectively, compared to those fed the Seadequate diet (0.1 ± 0.1 ␮mol/L vs. 6.5 ± 0.6 ␮mol/L for plasma, 0.5 ± 0.2 ␮mol/kg vs. 40.0 ± 4.8 ␮mol/kg for liver, 0.1 ± 0.1 ␮mol/kg vs. 7.5 ± 0.6 ␮mol/kg for muscle, and 4.2 ± 0.7 ␮mol/kg vs. 72.2 ± 5.7 ␮mol/kg for kidneys). Dietary supplementation with SeMet resulted in a dose-dependent increase in tissue Se (Fig. 3).

L. Yan et al. / Journal of Trace Elements in Medicine and Biology 24 (2010) 223–229

227

Fig. 3. Curves showing the response of plasma (A), liver (B), muscle (C) and kidneys (D) to dietary supplementation with SeMet (solid circle), the protein isolate (open circle) and tofu (solid diamond). For the slope-ratio analyses, liver Se concentrations were log-transformed to achieve linearity. The Se concentrations were 14, 24 and 30 ␮g/kg with the isolate-supplemented diets and 13, 23 and 31 ␮g/kg with the tofu-supplemented diets. Mean ± SD, n = 8 for SeMet supplementation and n = 6 for the protein isolate or tofu supplementation.

At the end of the repletion period, the maximum restoration of Se was 93.8, 89.0, 77.3 and 90.0%, respectively, in plasma, liver, muscle and kidneys of the rats replenished with the 150 ␮g Se/kg diet (6.1 ± 0.9 ␮mol/L in plasma, 35.6 ± 3.7, 5.8 ± 0.9 and 65.0 ± 7.0 ␮mol/kg in liver, muscle and kidneys) compared to the rats fed the same diet throughout the experiment. Adding the protein isolate or tofu to the basal diet during the Se-repletion period resulted in linear or log-linear, dose-dependent increases in Se in those tissues (Fig. 3). The relative bioavailabilities of Se from the protein isolate were 98.6% (plasma), 117.2% (liver), 101.6% (muscle) and 99.5% (kidneys) and those of Se from tofu were 99.3% (plasma), 115.1% (liver), 78.4% (muscle) and 114.9% (kidneys), respectively, compared to that from SeMet (Table 3). Discussion In the present study, we assessed the bioavailabilities of Se from soy protein isolate and tofu, prepared from naturally produced high-Se soybeans, on the basis of restoration of Se-dependent enzyme activities and tissue Se retention in Se-depleted rats. A lowSe status was induced in rats by feeding them a Torula yeast-based diet, containing only 4.1 ␮g Se/kg, for a sufficient time, and then replenishing Se by supplementing the diet with high-Se soy protein isolate or tofu. Previous findings from our laboratory [27,30] and others [31] demonstrated the validity of this method in assessing the bioavailability of Se from foods. We demonstrated that dietary supplementation with Se from the protein isolate or tofu in a range from 10 to 30 ␮g Se/kg generated linear or log-linear, nonplateauing responses in Se-dependent enzyme activities and tissue Se contents, which could be used to extrapolate bioavailabilities relative to that of SeMet. The Se from the protein isolate and tofu was comparably effective in restoring seleno-enzyme activities and tissue Se concentrations. Soy protein isolate is a highly refined protein product

with most non-protein components removed, whereas tofu coagulated from soy milk is a crude protein product retaining lipids and carbohydrates. A major difference between these two products is their protein contents. In the present study, Se concentration was increased approximately twofold in the protein isolate and 40% in tofu compared with raw soybeans. These findings are in agreement with a previous report that the Se content of a processed soy product is dependent upon the protein content of the product [32]. Thus, different methods in processing soy foods may affect their protein contents as well as Se contents, but not the bioavailability of Se. The bioavailability of Se is defined as the fraction of ingested dietary Se that is utilized for normal physiological functions. The main biological function of GPX is to protect organisms from oxidative damage [33], and TRR plays important roles in antioxidant defense and in cell cycling control [34]. Glutathione peroxidase and TRR activities were used to assess the amount of Se from the protein isolate and tofu incorporated into the SeCys compartment for specific selenoprotein synthesis. Dietary supplementation with either the protein isolate or tofu produced dose-dependent increases in GPX and TRR activities that were similar to those of SeMet, which indicated that Se from these two soy products are comparably efficient in digestibility, absorbability and metabolic conversion to functional SeCys-enzymes as that in SeMet. While both seleno-enzyme activities and tissue Se responded to the amount of bioavailable Se consumed, restoration of SeCys-enzyme activities in Se-depleted animals has particular physiological relevance because of their biological functions. The retention of Se in tissues is an indirect measure of Se bioavailability. The Se in blood, liver, muscle and kidneys constitutes 60% of total body Se in humans [35]. In the present study, Se from either the protein isolate or tofu was comparable to Se from SeMet in replenishing Se concentrations in plasma, liver, muscle and kidneys. Increases in Se concentrations in these tissues are most likely the result of non-specific incorporation of SeMet into tissue

228

L. Yan et al. / Journal of Trace Elements in Medicine and Biology 24 (2010) 223–229

proteins. The observed increases in both tissue Se concentrations and seleno-enzyme activities in rats fed the protein isolate- and tofu-supplemented diets indicate that Se from these soy products is not only catabolized to selenide to enter the SeCys compartment but also significantly retained in the non-specific SeMet compartment. In the present study, a slope-ratio model [28] was used to determine the relative bioavailability of Se from the protein isolate and tofu compared to Se from SeMet. In this model, linear or log-linear regression lines are generated for both test and standard compounds, and the relative bioavailability is expressed as the ratio of the slope of the test compound to that of the standard. Furthermore, it requires: (a) graded levels of the test compounds, (b) intakes of the test compound that do not exceed the amount required to fully replenish the response measure, and (c) regression lines for test and reference treatments have a common intercept. Using these guidelines, we designed to use the protein isolate and tofu in amounts that provided 20, 30, and 40 ␮g Se/kg diet. Although the analyzed Se concentrations of the protein isolate- and tofusupplemented diets did not achieve targeted values, they were within the design concentration range and gave dose-dependent relationships. For all regressions, we used the standard response levels between 5 and 70 ␮g Se/kg, because at levels greater than 70 ␮g Se/kg the responses for some enzymes and tissue Se became non-linear or reached a plateau. Linearity of the respective regression lines was ascertained for each source of Se after which a single multiple regression model was derived to determine the slope and intercept of the responses for the three Se sources, the protein isolate, tofu and SeMet [29]. As a result, this approach generated linear or log-linear, dose-dependent increases in Se enzyme activities and tissue Se and accurate estimates of bioavailabilities relative to those from SeMet. Soy contains isoflavones, a group of bioactive components with potential health benefits. In vitro studies showed that genistein [36,37] and daidzein [37] protect cultured cells against oxidative stress; the protection is accompanied by increased GPX activity [37,38]. Animal studies revealed that feeding rats a 20% isoflavonecontaining soy protein diet increases hepatic GPX activity [39,40]. The present study assessed GPX and TRR activities as markers of Se bioavailability from soy. Considering that no greater than 1% protein isolate or tofu was added to the Se-deficient diet, the impact of isoflavones from these additions on enzyme activities quantified would be very minimal in this experiment. In the present study, growth was significantly retarded in rats fed the basal diet. Torula yeast was used as the protein source for the diet because it was very low in Se. Except for Se, the diet contained all required nutrients, including sulfur-containing amino acids, minerals and vitamins in amounts that met or exceeded NRC recommendations [22] and the AIN-93G formulation [20]. The only differences between the basal diet and experimental diets were their amounts and sources of Se. We noted a non-significant increase in body weight in rats fed the protein isolate- or tofusupplemented diet during the Se-repletion period. This suggests that the amount of Se in the diet was responsible for the differences in body weight observed in this study. The bioavailability of Se varies from different dietary sources. Generally, organic forms of Se (e.g. SeMet, Se-yeast and wheat Se) are more bioavailable than inorganic Se (e.g. selenite and selenate). This difference can be explained by the incorporation of organic sources of Se instead of methionine into tissue proteins where it is a storage site for Se that is readily available to the body upon tissue protein turnover. There are regions in the world (e.g. Europe and parts of China) where Se is deficient from natural resources, and intake of Se in many countries is below recommended levels [41]. While Se supplements are a good source of Se, they may not be readily available for many individuals. It has been recommended

that people should consume nutrients through whole food, rather from specific supplements [42]. Soy is a staple crop that people have already accepted and consumed as a source of dietary protein. It has been documented that SeMet is a dominant form of Se in soy [13,14]. Thus, naturally produced high-Se soybeans may provide needed Se for humans, particularly for those in Se-deficient regions. In conclusion, the present study demonstrated that Se from soy protein isolate and tofu was highly bioavailable. Compared with SeMet, these natural sources of Se were capable of restoring Sedependent enzyme activities and tissue Se concentrations in Sedeficient rats. Thus, naturally produced high-Se soybeans may be a good dietary source of Se. Acknowledgements The authors gratefully acknowledge Mr. Brad Karlen, Karlen Ranch, for providing us high-Se soybeans to this project, Dr. Barbara Bryan, the Solae Company, for advising us in preparation of the protein isolate and tofu and the assistance of the following staff of the Grand Forks Human Nutrition Research Center: James Lindlauf for preparing the experimental diets, Lana DeMars for conducting enzyme analyses, Craig Lacher and William Martin for conducting Se analysis, Kay Keehr for preparing the protein isolate and tofu, Denice Schafer and her staff for providing high quality animal care, and Dr. Forrest Nielsen for his input in revising this manuscript. References [1] Schwarz K, Foltz CM. Selenium as an integral part of factor 3 against dietary necrotic liver degeneration. J Am Chem Soc 1957;79:3292–3. [2] Rotruck JT, Pope AL, Ganther HE, Swanson AB, Hafeman DG, Hoekstra WG. Selenium: biochemical role as a component of glutathione peroxidase. Science 1973;179:588–90. [3] Gladyshev VN, Jeang KT, Stadtman TC. Selenocysteine, identified as the penultimate C-terminal residue in human T-cell thioredoxin reductase, corresponds to TGA in the human placental gene. Proc Natl Acad Sci U S A 1996;93:6146–51. [4] Ge K, Xue A, Bai J, Wang S. Keshan disease – an endemic cardiomyopathy in China. Virchows Arch A Pathol Anat Histopathol 1983;401:1–15. [5] Ammerman CB, Miller SM. Selenium in ruminant nutrition: a review. J Dairy Sci 1975;58:1561–77. [6] Broome CS, McArdle F, Kyle JA, Andrews F, Lowe NM, Hart CA, et al. An increase in selenium intake improves immune function and poliovirus handling in adults with marginal selenium status. Am J Clin Nutr 2004;80:154–62. [7] Negro R, Greco G, Mangieri T, Pezzarossa A, Dazzi D, Hassan H. The influence of selenium supplementation on postpartum thyroid status in pregnant women with thyroid peroxidase autoantibodies. J Clin Endocrinol Metab 2007;92:1263–8. [8] Clark LC, Combs GFJ, Turnbull BW, Slate EH, Chalker DK, Chow J, et al. Effects of selenium supplementation for cancer prevention in patients with carcinoma of the skin. A randomized controlled trial. Nutritional Prevention of Cancer Study Group. JAMA 1996;276:1957–63. [9] Low SC, Harney JW, Berry MJ. Cloning and functional characterization of human selenophosphate synthetase, an essential component of selenoprotein synthesis. J Biol Chem 1995;270:21659–64. [10] Sunde RA, Evenson JK. Serine incorporation into the selenocysteine moiety of glutathione peroxidase. J Biol Chem 1987;262:933–7. [11] Tormay P, Wilting R, Lottspeich F, Mehta PK, Christen P, Bock A. Bacterial selenocysteine synthase – structural and functional properties. Eur J Biochem/FEBS 1998;254:655–61. [12] Olson OE, Novacek EJ, Whitehead EI, Palmer IS. Investigations on selenium in wheat. Phytochemistry 1970;9:1181–8. [13] Yasumoto K, Suzuki T, Yoshida M. Identification of selenomethionine in soybean protein. J Agric Food Chem 1988;36:463–7. [14] Sathe SK, Mason AC, Weaver CM. Some properties of a selenium-incorporating sulfur-rich protein in soybeans (Glycine max L.). J Agric Food Chem 1992;40:2077–83. [15] Anderson JW, Johnstone BM, Cook-Newell ME. Meta-analysis of the effects of soy protein intake on serum lipids. N Engl J Med 1995;333:276–82. [16] Yan L, Spitznagel EL. Soy consumption and prostate cancer risk in men: a revisit of a meta-analysis. Am J Clin Nutr 2009;89:1155–63. [17] Kubota J, Allaway WH, Carter DL, Cary EE, Lazar VA. Selenium in crops in the United States in relation to selenium-responsive diseases of animals. J Agric Food Chem 1967;15:448–53. [18] Wolnik KA, Fricke FL, Capar SG, Braude GL, Meyer MW, Satzger RD, et al. Elements in major raw agricultural crops in the United States. 2. Other elements in lettuce, peanuts, potatoes, soybeans, sweet corn, and wheat. J Agric Food Chem 1983;31:1244–9.

L. Yan et al. / Journal of Trace Elements in Medicine and Biology 24 (2010) 223–229 [19] National Research Council. In: National Research Council Institute of Laboratory Animal Resources, Commission on Life Sciences. Guide for the care and use of laboratory animals, Washington, DC: National Academy Press; 1996. [20] Reeves PG, Nielsen FH, Fahey GCJ. AIN-93 purified diets for laboratory rodents: final report of the American Institute of Nutrition ad hoc writing committee on the reformulation of the AIN-76A rodent diet. J Nutr 1993;123:1939–51. [21] Jurgens MH. Feed stuffs used in livestock diets. In: Animal feeding and nutrition. 8th ed. Dubuque: Kendall/Hunt; 1997. p. 81–234. [22] National Research Council. Nutrient requirements of laboratory animals. Washington, DC: National Academy Press; 1995. p. 3–79. [23] Paglia DE, Valentine WN. Studies on the quantitative and qualitative characterization of erythrocyte glutathione peroxidase. J Lab Clin Med 1967;70:158–69. [24] Lawrence RA, Burk RF. Glutathione peroxidase activity in selenium-deficient rat liver. Biochem Biophys Res Commun 1976;71:952–8. [25] Hill KE, McCollum GW, Burk RF. Determination of thioredoxin reductase activity in rat liver supernatant. Anal Biochem 1997;253:123–5. [26] Hintze KJ, Wald KA, Zeng H, Jeffery EH, Finley JW. Thioredoxin reductase in human hepatoma cells is transcriptionally regulated by sulforaphane and other electrophiles via an antioxidant response element. J Nutr 2003;133:2721–7. [27] Reeves PG, Leary PD, Gregoire BR, Finley JW, Lindlauf JE, Johnson LK. Selenium bioavailability from buckwheat bran in rats fed a modified AIN-93G torula yeast-based diet. J Nutr 2005;135:2627–33. [28] Finney DJ. Statistical method in biological assay. London, UK: Charles Griffin & Company; 1978. [29] Littell RC, Henry PR, Lewis AJ, Ammerman CB. Estimation of relative bioavailability of nutrients using SAS procedures. J Anim Sci 1997;75:2672–83. [30] Reeves PG, Gregoire BR, Garvin DF, Hareland GA, Lindlauf JE, Johnson LK, et al. Determination of selenium bioavailability from wheat mill fractions in rats by using the slope-ratio assay and a modified torula yeast-based diet. J Agric Food Chem 2007;55:516–22. [31] Ornsrud R, Lorentzen M. Bioavailability of selenium from raw or cured selenomethionine-enriched fillets of Atlantic salmon (Salmo salar) assessed in selenium-deficient rats. Br J Nutr 2002;87:13–20.

229

[32] Weaver CM, Davis J, Marks HS, Sensmeier RK. Selenium content of processed soybeans. J Food Sci 1988;53:300–1. [33] Mills GC. Hemoglobin catabolism. I. Glutathione peroxidase, an erythrocyte enzyme which protects hemoglobin from oxidative breakdown. J Biol Chem 1957;229:189–97. [34] Sun QA, Wu Y, Zappacosta F, Jeang KT, Lee BJ, Hatfield DL, et al. Redox regulation of cell signaling by selenocysteine in mammalian thioredoxin reductases. J Biol Chem 1999;274:24522–30. [35] Sunde RA. Selenium. In: Stipanuk MH, editor. Biochemical, physiological, molecular aspects of human nutrition. New York: W.B. Saunders; 2006. p. 1091–126. [36] Xu SZ, Zhong W, Ghavideldarestani M, Saurabh R, Lindow SW, Atkin SL. Multiple mechanisms of soy isoflavones against oxidative stress-induced endothelium injury. Free Radic Biol Med 2009;47:167–75. [37] Hernandez-Montes E, Pollard SE, Vauzour D, Jofre-Montseny L, Rota C, Rimbach G, et al. Activation of glutathione peroxidase via Nrf1 mediates genistein’s protection against oxidative endothelial cell injury. Biochem Biophys Res Commun 2006;346:851–9. [38] Suzuki K, Koike H, Matsui H, Ono Y, Hasumi M, Nakazato H, et al. Genistein, a soy isoflavone, induces glutathione peroxidase in the human prostate cancer cell lines LNCaP and PC-3. Int J Cancer 2002;99:846–52. [39] Appelt LC, Reicks MM. Soy induces phase II enzymes but does not inhibit dimethylbenz[a]anthracene-induced carcinogenesis in female rats. J Nutr 1999;129:1820–36. [40] Lee JS. Effects of soy protein and genistein on blood glucose, antioxidant enzyme activities, and lipid profile in streptozotocin-induced diabetic rats. Life Sci 2006;79:1578–84. [41] Rayman MP. The use of high-selenium yeast to raise selenium status: how does it measure up? Br J Nutr 2004;92:557–73. [42] Hunt JR. Tailoring advice on dietary supplements: an opportunity for dietetics professionals. J Am Diet Assoc 2002;102:1754–5.