Acute and Subchronic Toxicity Studies on Sel-Plex ®,a Standardized,Registered High-selenium Yeast

Abstract

Selenium has been recognized as an essential nutrient for human health; however, its bioavailability is primarily dependent upon the type of selenium, elemental versus organic. In geographic areas low in selenium, there is the potential for animals (including humans) to become selenium deficient and this potential deficiency can be remedied by consumption of exogenous selenium, including selenium-enriched yeast (Saccharomyces cerevisiae) that contains high levels of organic selenium (e.g., selenized yeast). The present studies were conducted to investigate potential oral toxicity of a unique selenized yeast preparation (Sel-Plex^®) when administered to (1) adult female CHS Swiss mice ICo:OFI (IOPS Caw); (2) adult female CHS Sprague-Dawley rats; and (3) adult male and female Sprague-Dawley CD rats. For the 28- and 90-day toxicity studies, (1) adult male and female Sprague-Dawley CRL:CD^® (SD) IGS BR strain rats and (2) adult male and female 6- to 7-month-old Beagle dogs were used. The LD₅₀ for mice was ^® ≥2000 mg Sel-Plex® /kg (≥4.06 mg Se/kg) and for rats, was greater ^® than ≥2000 mg Sel-Plex® /kg (≥4.06 mg Se/kg). In the two 28-day studies, for rats, the no observed adverse effects level (NOAEL) was 50 mg Sel-Plex^®/kg/day (0.1 mg Se/kg/day), and for the dogs, the NOAEL was 22.5 mg Sel-Plex^®/kg/day (0.045 mg Se/kg/day). For the two 90-day studies, for rats the NOAEL for Sel-Plex^® was 114 mg/kg/day (0.23 mg Se/kg/day), and for dogs, the NOAEL was 30 mg Sel-Plex^®/kg/day (0.06 mg Se/kg/day): the latter being the NOAEL in the most sensitive species.

Selenium (Se) has been recognized as an essential nutrient for human health for over 40 years, and is a component of proteins and enzymes required for various biological functions, such as antioxidant defense, reduction of inflammation, thyroid hormone production, fertility, and DNA synthesis (Brown and Arthur 2001; Whanger 2002). Selenium has been utilized as a feed supplement for livestock in geographical areas that are naturally low in selenium. It has been demonstrated that selenium deficiencies can also occur in human diets, and may be clinically manifested as muscular weakness and myalgia, fatigue, loss of appetite, cardiomegaly, cardiac arrhythmia, and palpitations (Raisbeck 2000; Kanekura et al. 2005). Selenium toxicity, termed selenosis, is caused by excessive selenium intake and has been documented in humans in seleniferous regions of China where selenium-contaminated foods were consumed as a sole food source. It is also reported in individuals who consumed excessive amounts of selenium supplements. Clinical symptoms of selenosis include hair and nail loss, skin lesions, diarrhea, and a garlic-like odor of the breath and in body secretions. The exact consumption level at which selenosis occurs is a matter of debate; residents in seleniferous regions of South Dakota have been reported to consume approximately 0.7 mg Se/day with no toxic symptoms, but findings in China indicate that about 3 to 5 mg Se/day will cause selenosis (Yang et al. 1983; Fan and Kizer 1990; Longnecker et al. 1991).

Animal studies on selenosis have described liver cirrhosis, splenomegaly, and significant growth depression in rats fed levels of 6.4 mg/kg or more selenium from sodium selenite or from seleniferous wheat. Selenium administration at doses of 8.0 mg/kg or more caused additional effects such as pancreatic enlargement, anemia, and elevated serum bilirubin levels (Halverson, Palmer, and Guss 1966). Conversely, selenium has been shown to protect against mercury and cadmium toxicity, as well as inhibit genotoxicity (WHO 1987).

The amount of selenium in the human body is not wholly predicated by the amount of selenium in the diet, as its bioavailability is primarily dependent upon the type of selenium, i.e., elemental versus organic. Elemental selenium is poorly available, whereas organically bound selenium has a much greater bioavailability (Brown and Arthur 2001; Schrauzer 2001). Selenium functions through an association with proteins, termed selenoproteins. As selenium is structurally similar to sulfur, selenium will replace the sulfur atom in stoichiometric amounts in the normal biosynthetic pathways of plants, and is mainly found incorporated into methionine, to form selenomethionine (Sors, Ellis, and Salt 2005). Selenomethionine, synthesized in plants, is randomly incorporated in place of methionine in a variety of proteins in both plants and mammals. Most selenium in animal tissues is present as selenomethionine or selenocysteine (Allan, Lacourciere, and Stadtman 1999; Schrauzer 2003).

Certain yeasts are able to absorb inorganic selenium and produce selenomethionine, an organic selenium that is more easily absorbed by mammals (Encinar et al. 2003). There are many brands of selenized yeast on the market; however, not all have been fully characterized or standardized to the concentration or forms of elemental/organic selenium content. Currently there are a number of studies underway that are designed to substantiate the contention that selenium supplementation to the diet would provide benefits ranging from the prevention of the progression of cancers, including nonmelanoma skin cancers and prostate cancers as well as prevention of cardiovascular events (Greenwald et al. 2002; Larsen et al. 2004; Duffield-Lillico et al. 2003; Reid et al. 2004; Stranges et al. 2006). To date there has not been persuasive statistical significance that selenium supplementation can address these life-threatening effects, but that does not diminish the need for adequate selenium intake and the demonstrable antioxidant effects related to this element. One reason for this inconsistency may relate to the wide variability even amongst selenized yeasts. Larsen et al. (2004) speculated that differences in intake regimens, selenium content, speciation of selenium moieties, and their attendant bioavailability may explain the inconsistencies observed in clinical trials and safety studies to date. For example, selenomethionine, the predominant organic seleno compound isolated from selenized yeasts, has varied from a low of 27% to values routinely around 54% to 60% in several different selenized yeast preparations, and even between lots of the same preparation (Larsen et al. 2004). Sel-Plex^® is standardized to contain a total selenium content of 2000 mg/kg (ppm) with organoselenium equal to or greater than 98% of the total selenium content.

The present studies were conducted to investigate potential toxicity of a unique, standardized, and registered high-selenium yeast preparation when administered orally to rats or dogs.

METHODS AND MATERIALS

Test Articles

Sodium selenite was provided by Alltech (Meath, Ireland). A unique, standardized, registered, high-selenium food-grade baker’s yeast (Saccharomyces cerevisiae, S. cerevisiae CNCM 1-3060 extract; Sel-Plex^® ), also provided by Alltech,1 was selected after screening in excess of 150 distinct strains of S. cerevisiae for a unique and proprietary strain with the ability to accumulate appreciable levels of selenium to form the final commercial product, Sel-Plex^®, which contains S. cerevisiae CNCM 1-3060 with reproducible levels of selenium-containing proteins. The unique S. cerevisiae CNCM 1-3060 strain’s genomic DNA karyotype has been identified via pulsed field electrophoresis and polymerase chain reaction. It is known that yeast strains incorporate selenium into cellular proteins in a strain-specific manner (Encinar et al. 2003), such that the resulting yeast preparation is expected to demonstrate unique bioavailability and thus, unique toxicity or safety. S. cerevisiae CNCM 1-3060 has incorporated selenium from selenium-containing media into cellular proteins. For this reason, the yeast preparation tested in the described studies will be termed “Sel-Plex^®.” Sel-Plex^® is a spray-dried selenium yeast preparation with a total selenium content of 2000 mg/kg (ppm). The organic selenium content, predominately selenomethionine, is equal to or greater than 98% of the selenium content.

Animals

For the rat studies, animals were acclimatized for at least 5 to 7 days before initiation of dosing. At dosing, animals were 8 to 12 weeks old. All animals were clinically monitored at the time of delivery and during acclimation period, and those found unsuitable were excluded. Animals were housed in standard cages with sawdust bedding or polypropylene grid cages, with a controlled 12-h light/dark cycle. Temperature was maintained at 19°C to 25°C with a relative humidity of 30% to 70%. Potable water and standardized diet were provided ad libitum.

For the first and second acute studies, adult female CHS Sprague-Dawley rats and female CHS Swiss ICo:OFI (IOPS Caw) mice were obtained from Depre Breeding Centre (Saint Doulchard, France). For the third acute oral toxicity study, male and female Sprague-Dawley CD rats were obtained from Charles River Limited (Margate, Kent, UK). For the 28-day and 90-day subchronic toxicity studies, adult male and female Sprague-Dawley CRL:CD^® (SD) IGS BR strain rats were obtained from Charles River Limited (Margate, Kent, UK). Animals were acclimatized for at least 7 days before initiation of dosing. At dosing, animals were 8 to 12 weeks old for the 28-day study, and 6 to 8 weeks old for the 90-day study.

For the 28-day range-finding and 90-day subchronic toxicity studies in dogs, adult male and female 6- to 7-month-old Beagle dogs were obtained from Marshall Farms (North Rose, NY, USA) and acclimatized for at least 7 days. They were individually housed in kennels containing wood shavings bedding (SICSA, Alfortville, France), except when urine collection was required, and a controlled 12-h light/dark cycle. Temperature was maintained at 15°C to 25°C with a relative humidity of 30% to 70%. Potable water was provided ad libitum and 300 g/dog/day of standardized UAR 125 C3 pelleted diet (SAFE, Villemoisson, Epinay-sur-Orge, France) was provided.

All animal studies complied with EC Council Directive No. 86/609/EEC, 24NOV1986 (EC 1986) and the appropriate parts of the Animal Welfare Act Regulations, 9CFR Parts 1, 2, and 3 (Federal Register 1989). These studies also met or exceeded the requirements of OECD Guidelines (1997) and FDA 1993 Redbook II Guidelines (FDA 1993), including OECD (1998), and FDA Principles of Good Laboratory Practices (FDA 2006).

Experimental Design—Acute Oral Toxicity

For both the Swiss mice and the Sprague-Dawley rats, a dose of 2000 mg/kg of Sel-Plex^® corresponding to 4.06 mg Se/kg) or 50 mg/kg sodium selenite (corresponding to 22.5 mg Se/kg) was administered. For each test substance, three females were dosed, followed a short time later by a second group of three females. The two test substances were administered as single 10 ml/kg homogeneous suspensions in corn oil via gavage to animals deprived of food for 24 h. A third acute study was also conducted in six fasted Sprague-Dawley rats (three male, three female) given Sel-Plex^® at doses of 25 mg/kg (0.051 mg Se/kg), 200 mg/kg (0.406 mg Se/kg), or 2000 mg/kg (4.06 mg Se/kg). A fourth group of six rats was given sodium selenite at 25 mg/kg (11.25 mg Se/kg). All test substances in this study were prepared in distilled water and administered via gavage.

Experimental Design—28-Day Subchronic Oral Toxicity Studies in Rats and Dogs

For the 28-day range-finding study in rats, Sel-Plex^® was administered for up to 28 days via dietary admixture to six rats (three males and three females) per group (Table 1). For the 28-day range-finding study in dogs, Sel-Plex^® was orally administered (via gavage) for 28 days to two male and two female Beagle dogs per group at the concentrations indicated in Table 1. Dose volumes for the dog study were 5 ml/kg for all concentrations. Each animal in both the rat and dog study was observed at least twice a day for mortality or signs of morbidity during the treatment period.

Experimental Design—90-Day Subchronic Oral Toxicity Studies in Rats and Dogs

The animals in the subchronic oral toxicity studies received a mean achieved dose according to the concentrations outlined in Table 2. For the 90-day study in rats, Sel-Plex^® was administered by dietary admixture to three groups (10 male, 10 female per group). A fourth group of 20 rats (10 male, 10 female) was given sodium selenite at a mean achieved dose of 0.8 mg/kg/day. In the 90-day study in dogs, the test article was administered for 13 weeks by gavage to three groups of eight dogs (four male, four female per group). Sodium selenite was administered to eight dogs (four male, four male) at 1.3 mg/kg/day (Table 2).

In-life Observations—Acute Oral Toxicity Studies in Rats and Mice

Both mice and rats received full clinical examinations for signs of toxicity, morbidity, and mortality outside the housing cage at 30 and 90 min after dose administration, on Day 7, and at the completion of the study, respectively. The full clinical examination included behavioral, neurological and autonomic profiles. The animals were reviewed cage-side for signs of toxicity, morbidity and mortality at the following time points: rats 3 to 4 h post dose; mice 2 to 3 h post dose; both species once daily for 14 days. In the third oral toxicity study, rats were observed 0.5, 1, 2, and 4 h post doing, and then once daily for 14 days for signs of toxicity, morbidity, and mortality. For all three studies, animals were weighed on day 0 (dosing) and on Days 7, 14, and at death.

In-life Observations—Subchronic Oral Toxicity in Rats and Dogs

For the day range-finding study in rats, all animals were examined for overt signs of toxicity, morbidity, or behavioral change daily. Animals were weighed weekly and at death. Food consumption was recorded by cage group, twice weekly. In the subchronic, 90-day repeat-dose study in rats, all animals were examined once daily for overt signs of toxicity, morbidity, or mortality. Before treatment and weekly thereafter, all animals were observed for functional or behavioral change. Animals were weighed at Day 0, weekly thereafter, and at death. Food consumption was recorded weekly for each cage group. Ophthalmoscopic examination was performed on high-dose and control groups.

For the 28-day range-finding study in Beagle dogs, all animals were observed twice daily for mortality and signs of morbidity. Food consumption was recorded daily and animals were weighed on Days 4, 8, 11, and 14. In the 90-day study in Beagle dogs, the dogs were observed at least once per day, and were weighed prior to distribution to a group, on Day 1 of treatment, and once weekly thereafter. Daily food intake per animal was estimated 5 days before initiation and during the course of the study. Detailed clinical observations were made at initiation and then once per week. Ophthalmologic examinations were performed on all animals prior to initiation, and all high-dose and positive-control animals were reexamined at study conclusion.

Clinical Chemistry

In rats, blood was sampled on day 90 for hematological and blood chemistry analysis. The hematological analyses included hemoglobin, hematocrit, erythrocyte count, erythrocytic indices, leukocyte count, leukocyte indices, platelet count, reticulocyte count, prothrombin time, and activated partial thromboplastin time. Plasma analyses included alanine amino-transferase, albumin, albumin/globulin ratio, alkaline phosphatase, aspartate amino-transferase, calcium, chloride, creatinine, glucose, inorganic phosphorus, sodium, potassium, total bilirubin, total cholesterol, total protein, and urea. In dogs, fasting blood and urine samples were collected on Day 0, Week 6, and at the end of the exposure period. Hematological analyses included erythrocyte count, erythrocytic indices, leukocyte count, leukocyte indices, platelet count, reticulocyte count, fibrinogen, prothrombin time, and activated partial thromboplastin time. Plasma analyses included urea, glucose, total protein, albumin, albumin/globulin ratio, sodium, potassium, chloride, calcium, inorganic phosphorus, aspartate amino-transferase, alanine amino-transferase, lactate dehydrogenase, creatine kinase, γ -glutamyltransferase, alkaline phosphatase, creatinine, total cholesterol, triglycerides, and total bilirubin. Urinary parameters included appearance, color, pH, specific gravity, proteins, glucose, ketones, bilirubin, nitrites, blood, urobilinogen, and sediment cytology. Bone marrow smears were prepared from the sternum of animals at the end of the study and bone marrow differential cell count was determined.

Pathology

In the acute mouse and rat studies, the animals were euthanized on Day 15 by exsanguination after cervical dislocation. All animals were subjected to necropsy, and the gonads/reproductive tract, heart, intestines, kidney, liver, lungs and stomach were examined both for gross and microscopic pathology.

In the rat studies (28-day and 90-day), survivors at the end of the study were euthanized by intravenous overdose of sodium pentobarbitone and examined for gross pathological change. In the 28-day studies, kidneys, liver, spleen, and testes were removed and weighed. In the 90-day studies, the following organs were removed and weighed: adrenals, brain, epididymides, heart, kidneys, liver, ovaries, spleen, testes, thymus, and uterus. Samples were preserved in 10% formalin for microscopic examination. Tissues that were collected (but not weighed) and also subjected to gross and microscopic examination included aorta (thoracic), bone and bone marrow, cecum, colon, eye, duodenum, ileum, jejunum, lungs, lymph nodes, mammary gland, skeletal muscle, esophagus, pancreas, pituitary, prostate, rectum, salivary glands, sciatic nerve, seminal vesicles, skin, spinal cord, stomach, thyroid/parathyroid, tongue, trachea, and urinary bladder.

In the dog studies (28-day and 90-day), survivors were euthanized by exsanguination after intravenous injection of thiopental sodium subjected to necropsy and examined for gross pathologies at the end of the study. In the 28-day study, adrenals, brain, kidney, liver, ovaries, spleen, testes, thymus, thyroids, and parathyroids were removed and weighed. No histological examinations were performed in this study. In the 90-day studies, the following organs were removed and weighed: adrenals, brain, heart, kidney, liver, ovaries, spleen, testes, thymus, thyroids, and parathyroids. Samples were preserved in 10% formalin for microscopic examination. Tissues that were collected (but not weighted) and also subjected to gross and microscopic examination included: aorta, eye, gall bladder, prostate, salivary glands, sciatic nerve, skin, stomach, tongue, trachea, urinary bladder, and vagina.

Statistical Analyses

In the rat studies, data were processed to give group mean values and standard deviations. Hematological, blood chemistry, organ weight (absolute and relative), weekly weight gain, quantitative functional performance, and sensory reactivity data were assessed by linear regression analysis for dose-response relationships between control and treatment groups, followed by one-way analysis of variance (ANOVA) incorporating Levene’s test for homogeneity of variance (Levene 1960). Control and positive control data were assessed by a two-tailed t test incorporating Levine’s test. Where variances were shown to be homogenous, pairwise comparisons were conducted using Dunnett’s test (Dunnett 1955, 1964). Where Levene’s test showed unequal variances, the data were analyzed using nonparametric methods, including Kruskal-Wallis ANOVA (Kruskal and Wallis 1952 Kruskal and Wallis 1953) and Mann-Whitney U test and Wilcoxon rank sum test (Wilcoxon 1945). Histopathological data were analyzed to determine significant differences between control and treatment groups for the individual sexes using either the chi-squared analysis for differences in the incidence of lesions occurring with an overall frequency of ≥1, or the Kruskal-Wallis one-way non-parametric analysis of variance for the comparison of severity grades for the more frequently observed graded conditions.

The low number of animals used in the 28-day dog study precluded the use of statistical analysis. In the 90-day dog study, normally distributed data were assessed for homogeneity using Bartlett or Fisher tests. Nonhomogenous data sets were analyzed using Dunn’s, Mann-Whitney, or Wilcoxon tests; homogeneous data sets were compared using the Student’s t test or Dunnett’s test. Data sets that were not normally distributed were logarithmically transformed and compared using the Dunn test (Dunn 1964).

RESULTS

Acute Oral Toxicity Studies in Rats and Mice

No mortalities occurred in Sprague-Dawley rats or Swiss mice in the two acute oral toxicity studies of Sel-Plex^® at the maximum dose used, 2000 mg/kg body weight (bw) (4.06 mg Se/kg bw). With sodium selenite, all of the mice (3/3) and all of the rats (3/3) died at the 50 mg/kg bw dose (22.5 mg Se/kg bw) (Table 3). At the middle dose, 200 mg Sel-Plex^®/kg bw (0.406 mg Se/kg bw), 0/6 mice and 1/6 rats died. On Day 1, 3 hours after treatment with 2000 mg Sel-Plex^®/kg bw, a marked decrease in locomotor activity was observed. From Day 5, the right eye of one animal was opaque. No other clinical abnormalities were observed. Mean weight in treated animals was normal, and no organ or tissue exhibited gross pathological change (Table 3). In mice, at a dose of 2000 mg Sel-Plex^®/kg bw (4.06 mg Se/kg bw), a marked decrease in locomotor activity was observed in one animal. No other clinical signs were observed.

In the third acute rat study, there were no significant clinical signs or systemic toxicity at any of the doses tested. Hunched shoulders were noted in males up to one day after dosing at the highest dose. No abnormalities were noted at necropsy.

To summarize, the LD₅₀ for mice was ≥2000 mg Sel-Plex^®/kg bw (≥4.06 mg Se/kg bw); for rats, the LD₅₀was greater than ≥2000 mg Sel-Plex^®/kg bw (≥4.06 mg Se/kg bw). The LD₅₀ for sodium selenite for both rats and mice was between 30 and 50 mg/kg bw (13.5 to 22.5 mg Se/kg bw).

Subchronic Oral Toxicity Studies in Rats and Dogs

No mortality occurred in any of the rats or dogs administered Sel-Plex^® for 28 or 90 days. One dog in the 90-day study died on day 29 after treatment with 1.3 mg sodium selenite/kg/day (0.6 mg Se/kg day).

28-Day Subchronic Oral Toxicity Observations

In the 28-day rat study, hunched posture was noted in the 2500 mg/kg Sel-Plex^®/day (5.1 mg Se/kg/day) dose group from day 4 onward, with emaciation evident by Day 8. Therefore, the 2500 mg/kg/day dose group was terminated without further treatment. Similar observations were noted in the 1000 mg Sel-Plex^®/kg/day (2.0 mg Se/kg/day) dose group, with hunched posture and generalized fur loss apparent by Day 3; this group was terminated on Day 9 of the study. The 250 mg Sel-Plex^®/kg/day (0.51 mg Se/kg/day) dose group developed a hunched posture at Day 18, which continued to study termination. All animals treated with 1000 or 2500 mg/kg/day Sel-Plex^® showed substantial bodyweight losses on Days 4 and 8. Nonsignificant reductions in bodyweight gain were detected in the 250 mg/kg/day dose group throughout the study period. No adverse effects were noted in the 50 mg Sel-Plex^®/kg/day (0.1 mg Se/kg/day) dose group during the study. No significant hematology or clinical chemistry changes were noted in the 250 mg/kg/day dose group. As the 1000 and 2500 mg/kg/day Sel-Plex^® dose groups were terminated prior to the end of the study, hematological and blood chemistry analyses were not performed.

In the 28-day subchronic study in dogs, excessive salivation was noted from Day 1 in one dog administered 112.5 mg Sel-Plex^®/kg/day (0.225 mg Se/kg/day), and intermittently in the female 112.5 mg/kg/day dose group. Excessive salivation was noted in males and females administered 562.5 mg Sel-Plex^®/kg/day (1.125 mg/kg/day), before or shortly after treatment. However, as this is commonly observed in dogs treated via gavage, excessive salivation was not considered an adverse effect. One male in the 562.5 mg Sel-Plex^®/kg/day dose group showed vomiting on Days 16 to 19, with hypoactivity on Day 17 and soft feces on Days 20 and 21. As this occurred in the high-dose group, a treatment relationship cannot be ruled out. Although no remarkable hematological or clinical chemistry changes were noted, a slightly decreased erythrocyte count, hemoglobin concentration, and packed cell volume were noted in the male 112.5 and 562.5 mg Sel-Plex^®/kg/day dose groups. No changes were noted in the female dose groups (data not shown for either sex). No toxicological changes were noted at pathological investigation in the 22.5 or 112.5 mg Sel-Plex^®/kg/day (0.045 mg Se and 0.225 mg Se/kg/day, respectively) dose groups.

90-Day Subchronic Oral Toxicity Observations

In the 90-day study in rats, there were lower body weight gains in all Sel-Plex^® and sodium selenite treatment groups compared to controls, especially in females during the first three weeks of the study. The effect was most persistent in the sodium selenite and high-dose Sel-Plex^® groups (Table 4). As a consequence, there was reduced dietary intake for all groups for the duration of the study (Figures 1 and 2). No clinically observable signs appeared in rats in response to the Sel-Plex ^® at 114 mg/kg/day (0.23 mg Se/kg/day). Male rats receiving 179 mg Sel-Plex^®/kg/day (0.36 mg Se/kg/day) developed a hunched posture in Week 2. Females from this treatment group also showed hunched postures and generalized fur loss. Females receiving a maximum dosage of 303 mg Sel-Plex^®/kg/day (0.61 mg Se/kg/day) showed persistent fur loss with sporadic incidents of ‘tiptoe gait.’ Piloerection and emaciation were also observed in one female for a short period in the second week of treatment. Reduced body weight gain was noted in rats administered the Sel-Plex^®at 179 and 303 mg/kg/day. Behavioral assessments determined a hunched posture in 179 and 303 mg/kg/day dose groups, with a tiptoe gait also noted at the highest dose. Sensory reactivity assessment noted an increase in startle reflex at the highest Sel-Plex^® dose and the positive-control group. Effects of sodium selenite at 0.8 mg/kg/day (0.35 mg Se/kg/day) were similar to those in rats administered Sel-Plex^® at 303 mg/kg/day (0.61 mg Se/kg/day).

Significant reductions in male rat heart, spleen, and thymus organ weights were noted in the 179 and 303 mg/kg/day Sel-Plex^® dose groups, and in the sodium selenite dose group (Table 4). Female organ weights (heart, kidneys, ovaries, and thymus) were also significantly decreased in 179 and 303 mg Sel-Plex^®/kg/day dose groups and the sodium selenite dose group (Table 4). Gross pathology evaluations resulted in no treatment-related macroscopic findings detected for male or female rats dosed at 114 mg/kg/day. The majority of the females in the 303 mg/kg/day group showed an accentuated lobular pattern in the liver; and this observation was observed in one female in the 179 mg/kg/day group. Similar liver morphology changes were observed in the sodium selenite treatment group. Histopathological evaluation of the rats found no differences in incidence or severity between control and 114 mg/kg/day Sel-Plex^® dose groups. Female rats at 179 and 303 mg/kg/day had centrilobular hepatocyte enlargement (p <.05) and hypertrophy of the zona glomerulosa cells of the adrenal glands (p <.001). Similar observations were noted with the sodium selenite-treated animals.

Clinical chemistry analysis of the female rats indicated decreased albumin and albumin/globulin ratios in the 303 mg/kg/day dose, compared to control values (Table 5). Increased alkaline phosphatase and bilirubin levels were noted in the 303 mg/kg/day Sel-Plex^® dose group, compared to controls. In addition, significant increases in bilirubin levels were noted in the 114 and 179 mg/kg/day Sel-Plex^® female dose groups (p <.05). Decreased glucose and albumin levels, and increased alkaline phosphatase and bilirubin (p <.001) levels were noted in the female sodium selenite dose group. In male rats, no significant blood chemistry values were noted in any of the Sel-Plex ^® treatment groups. A significant decrease in glucose was noted in the sodium selenite group (140 ± 8 versus 150 ± 11 mg/dl control), as well as significant increases in alanine amino-transferase (58 ± 10 IU/L) and cholesterol (83 ± 7 mg/dl) values, when compared to controls (49 ± 8 and 69 ± 8 mg/dl, respectively).

In the 90-day dog study, there were no Sel-Plex^® effects observed with food consumption or body weight gain; however, sodium selenite resulted in lower body weight gain in female dogs. Excessive salivation was noted in a dose-related manner in Sel-Plex^® –treated dogs, and in the sodium selenite–treated dogs. Excessive salivation, vomiting, and regurgitation are clinical signs commonly observed in dogs treated via gavage and therefore are not considered as adverse events, although in this study they were not observed in the controls and therefore may be treatment related. Emaciated appearance was noted in 2/4 male dogs fed Sel-Plex^® at 300 mg/kg/day (0.6 mg Se/kg/day). Further, a higher incidence of liquid/soft feces was observed in the Sel-Plex^® dose groups.

In the 300 mg Sel-Plex^®/kg/day dose group, erythrocyte count, hemoglobin concentration, packed erythrocyte volume (PCV), mean cell hemoglobin concentration (MCHC) were all lower in Weeks 6 and/or 13, when compared to control and prestudy values (Table 6). The lower mean cell hemoglobin concentration in males given 30 mg/kg/day Sel-Plex^® and in females given 100 mg/kg/day Sel-Plex^® (p <.05) was not associated with changes in other related parameters, and therefore a treatment relationship is doubtful. In Week 13, female dogs fed 100 or 300 mg/kg/day showed higher cholesterol levels, and increased triglyceride levels were noted in female dogs administered 300 mg/kg/day Sel-Plex^® (data not shown). The 100 and 300 mg/kg bw/day Sel-Plex^® male dose groups, the 300 mg/kg/day female dose group, and the male and female sodium selenite dose groups (1.3 mg/kg/day) had lower relative thymus weights, but these were not statistically significant reductions (Table 7). A notable, although not statistically significant, reduction in the size of the thymus was observed in the 300 mg/kg/day dose groups, which corresponded to a higher incidence and severity of thymic lymphoid depletion noted in the 300 mg Sel-Plex^®/kg bw/day and the sodium selenite treatment groups (Table 7). There were no adverse ophthalmological findings in either rats or dogs from these subchronic studies (data not shown).

DISCUSSION

These studies were conducted to investigate the potential acute and subchronic toxicity of Sel-Plex^® in various animal models. Sel-Plex^® was shown to have a significantly lower acute oral toxicity than sodium selenite in both the rat and mouse. Acute toxicity limit tests conducted in rats and mice showed the maximum test concentration of 2000 mg Sel-Plex^®/kg bw (4.06 mg Se/kg bw) were without mortality. A third acute study indicated that the LD₅₀ was ≥2500 mg Sel-Plex^®/kg bw (5.1 mg Se/kg bw) in rats. A reduced dietary intake was noted throughout the observation period for all treatment groups, which may be considered as a consequence of the unpalatability of the dietary mixture. This resulted in a decreased bodyweight gain in all treatment groups, particularly the females.

The subchronic rat study indicates that administration of 303 mg/kg/day Sel-Plex^® (0.61 mg Se/kg/day) preparation induced hepatotoxic effects in female rats, as indicated by morphological changes in the liver. Previous studies also indicate that selenium at toxic doses induces hepatoxicity (Halverson, Palmer, and Guss 1966). Subchronic, repeat-dose testing for 90 days in the Sprague-Dawley rat and the Beagle dog corroborate the findings of other investigators who described a more efficient selenium status from animals that were fed organic selenium compared to inorganic selenium salts (sodium selenite) (Pehrson et al. 1999; Rayman 2004; Reeves et al. 2005). The studies described herein demonstrate that selenium administered via a yeast strain, i.e., organic Se, is less toxic than comparable levels of selenium administered via sodium selenite. The treatment-related effects in rats fed 0.8 mg sodium selenite/kg/day (0.35 mg Se/kg/day) were similar in nature and severity to those fed 303 mg Sel-Plex^®/kg/day (0.61 mg Se/kg/day). The no observed adverse effect level (NOAEL) for Sel-Plex^® in rats was 114 mg/kg/day (0.23 mg Se/kg/day). In Beagle dogs, no deaths occurred in dogs dosed as high as 300 mg Sel-Plex^®/kg/day (0.6 mg Se/kg/day); whereas, one in four dogs died when dosed with an equivalent amount of selenium (0.6 mg Se/kg/day) in the form of sodium selenite (1.3 mg/kg/day). The NOAEL is posited at 30 mg Sel-Plex^®/kg/day (0.06 mg Se/kg/day) in Beagle dogs, based on clinical chemistry changes noted at 100 and 300 mg Sel-Plex^®/kg/day.

Agricultural livestock have been fed selenium supplements for decades, and studies have shown that selenized yeast supplementation increases selenium stores more efficiently, and positively affects growth, immune status, and liver oxidative status (Sobajic, Mihailovic, and Miric 1998; Mahmoud and Edens 2003; Mihalikova et al. 2005; Weiss and Hogan 2005). From the acute and subchronic studies presented here, and the data published in the literature regarding the relatively nontoxic activity of selenized yeast in various animal species, we conclude that Sel-Plex^® has a NOAEL of 30 mg/kg/day (0.06 mg Se/kg/day) in the most sensitive sex and species.

It is interesting to note that the NOAEL in the most sensitive species is relatively similar, based on selenium content, to the tolerable upper limit (UL) espoused by the Institute of Medicine (IoM 2000). The value for adults, 19 years and older is 400 μg/day (0.4 mg/day or 5.1 μmol/day) selenium without distinction of elemental versus organic presentation. For a 60 adult, this value is converted to 0.0067 mg, approximately 10-fold safety factor from the dog NOAEL. A definitive NOAEL most likely lies between the 30 and 100 mg Sel-Plex^®/kg/day (0.06 and 0.2 mg Se/kg/day, respectively); however, additional research will need to be conducted to define the slope of the dose-response curve.