Selenium tolerance in sheep and selenium supplementation methods for beef cattle

MISSING IMAGE

Material Information

Title:
Selenium tolerance in sheep and selenium supplementation methods for beef cattle
Physical Description:
ix, 146 leaves : ill. ; 29 cm.
Language:
English
Creator:
Davis, Paul Armand
Publication Date:

Subjects

Subjects / Keywords:
Animal Sciences thesis, Ph. D
Dissertations, Academic -- Animal Sciences -- UF
Genre:
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 2004.
Bibliography:
Includes bibliographical references.
Statement of Responsibility:
by Paul Armand Davis.
General Note:
Printout.
General Note:
Vita.

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 024494216
oclc - 847495563
System ID:
AA00022519:00001

Full Text











SELENIUM TOLERANCE IN SHEEP AND SELENIUM SUPPLEMENTATION
METHODS FOR BEEF CATTLE
















By

PAUL ARMAND DAVIS


A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA


2004












L J


-w
I' ~ ~2


N
~1 -f
































Copyright 2004

by

Paul Armand Davis































To our Heavenly Father and my loving family.














ACKNOWLEDGMENTS


The author wishes to express the utmost gratitude to Dr. Lee R. McDowell,

chairman of the supervisory committee, for his guidance, patience, encouragement and

direction throughout the Ph.D. program and during preparation of this dissertation.

Appreciation is given to Drs. Tim Marshall, Claus Buergelt, and Richard Weldon for

service and dedication on the graduate committee and for advice and understanding.

Special thanks are extended to Dr. McDowell for his unselfish attitude and including me

on several experiments, publications, and educational opportunities. Gratefulness beyond

words is in order to Dr. McDowell for showing me that one will get further by lifting

others up than by putting anyone down. Dr. Todd Thrift deserves thanks for his

willingness to listen and provide both professional and personal advice.

A huge debt of gratitude and appreciation is extended to Mr. Bert Faircloth, Mr.

Charles Stephens, Mr. Steve Chandler, Mr. Jesse Savell, and Mr. Brantley Ivey at the

Santa Fe Beef Unit for their help in conducting the experiments and collecting data. Mr.

Dean Glicco deserves particular thanks for his care of the experimental animals and for

becoming a trusted friend and confidant to the author. The author thanks Ms. Nancy

Wilkinson for sharing her invaluable knowledge of lab procedures and for her patience in

instruction of laboratory techniques. Mrs. Lorraine McDowell deserves thanks for her

work with electron microscopy on animal tissues.








Fellow graduate students Deke Alkire, Carlos Alosilla, Bradley Austin, Nathan

Krueger, Edgar Rodriguez, and Oswaldo Rosendo deserve thanks for their help with

sample collection. Also, Eric Matsuda-Fugisaki provided some needed assistance in the

laboratory and is much appreciated. Likewise, appreciation is extended to all graduate

students in the Animal Sciences Department for support and camaraderie during this

program of study. Pam Gross deserves recognition for her unyielding willingness to help

others and her wonderful attitude.

United States Sugar Corporation and its employees deserve thanks and

recognition for their donations of liquid feeds. A note of appreciation goes to Dr. Jon

Nelson and Southeastern Minerals, Bainbridge, Georgia, for donation of mineral

supplements, to Flint River Mills for transportation of mineral supplements, to Alltech,

Nicholasville, Kentucky, for donation of Sel-Plex, and to Mr. Dane Bernis for

preparation of experimental diets.

The author wishes to acknowledge, though not necessarily in a positive manner,

Hurricanes Charley, Frances, Ivan, and Jeanne. In the worst hurricane season in many

decades, the storms caused chaos across the state of Florida by damaging property and

generally disrupting daily life.

Last, but certainly not least, the author wishes to thank Ms. Rachel Van Alstyne

for her assistance with laboratory analyses and for her unselfishness and dedication as a

friend and colleague. She has truly been the kind of friend that a friend would like to

have.













TABLE OF CONTENTS

ACKNOW LEDGM ENTS ............................................................................................ iv

ABSTRACT ..................................................................................................................... viii

CHAPTER

1 INTRODUCTION ..................................................................................................... 1

2 REVIEW OF LITERATURE ................................................................................... 3

Benefits of Selenium Supplem entation to Livestock ............................................... 3
M ethods of Selenium Supplementation to Livestock ............................................... 8
Absorption, Transport, Storage, and Excretion of Selenium ................................. 12
Differences in Efficacy of Selenium due to Source ............................................... 14
Selenium Toxicosis ................................................................................................. 17

3 TOLERANCE OF INORGANIC SELENIUM IN RANGE-TYPE EWES DURING
GESTATION AND LACTATION ........................................................................ 22

Introduction ................................................................................................................. 22
M aterials and M ethods ............................................................................................ 23
Results and Discussion ............................................................................................ 26
Implications ................................................................................................................. 39
Summary ..................................................................................................................... 40

4 EFFECTS OF SELENIUM LEVELS IN EWE DIETS ON SELENIUM IN MILK
AND PLASMA AND TISSUE SELENIUM CONCENTRATIONS OF LAMBS ... 48

Introduction ................................................................................................................. 48
M aterials and M ethods ............................................................................................ 49
R esu lts ......................................................................................................................... 5 1
Discussion ................................................................................................................... 56
Implications ................................................................................................................. 61
Summary ..................................................................................................................... 61








5 COMPARATIVE EFFECTS AND TOLERANCE OF VARIOUS DIETARY
LEVELS OF SE AS SODIUM SELENITE OR SE YEAST ON BLOOD, WOOL,
AND TISSUE SE CONCENTRATIONS OF WETHER SHEEP ......................... 67

Introduction ................................................................................................................. 67
M aterials and M ethods ............................................................................................ 68
Results and D iscussion ............................................................................................ 71
Im plications ................................................................................................................. 81
Sum m ary ..................................................................................................................... 82

6 EFFECTS OF FORM OF PARENTERAL OR DIETARY SELENIUM
SUPPLEMENTATION ON BODY WEIGHT AND BLOOD, LIVER, AND MILK
CON CEN TRA TION S IN BEEF COW S ............................................................... 88

Introduction ................................................................................................................. 88
M aterials and M ethods ............................................................................................ 89
Results ......................................................................................................................... 92
Discussion ................................................................................................................... 95
Implications ................................................................................................................. 99
Sum m ary ..................................................................................................................... 99

7 TISSUE AND BLOOD SELENIUM CONCENTRATIONS AND PERFORMANCE
OF BEEF CALVES FROM DAMS RECEIVING DIFFERENT FORMS OF
SELEN IUM SUPPLEM EN TATION ....................................................................... 106

Introduction ............................................................................................................... 106
M aterials and M ethods .............................................................................................. 107
Results ....................................................................................................................... 110
D iscussion ................................................................................................................. 113
Im plications ............................................................................................................... 118
Sum m ary ................................................................................................................... 118

8 SUM M ARY AN D CON CLU SION S ....................................................................... 125

LITERATURE CITED ................................................................................................... 133

BIOGRA PHICAL SKETCH .......................................................................................... 145














Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy
SELENIUM TOLERANCE IN SHEEP AND SELENIUM SUPPLEMENTATION

METHODS FOR BEEF CATTLE

By

Paul Armand Davis

December 2004

Chair: Lee Russell McDowell
Major Department: Animal Sciences
A series of experiments to evaluate and compare methods, sources, and dietary

levels of selenium was carried out utilizing sheep and cattle. Experiments using sheep

were conducted to gather further data on 1) the tolerance of dietary inorganic Se by ewes

during lamb production, 2) the effects of high levels of dietary Se fed to ewes on their

lambs, and 3) the tolerance of organic or inorganic Se by mature wethers. A cow-calf

herd was used to evaluate and compare effects of using different forms of dietary or

parenteral Se on weight change and blood, milk, and liver Se concentrations of beef cows

and their calves. In ewes fed Se, as sodium selenite, above requirements, Se

concentrations in blood, wool, and soft tissues generally increased (P < 0.05) as dietary

Se increased. Ewes tolerated up to 20 mg/kg dietary Se without suffering from toxicosis.

Lambs born to ewes receiving high levels of dietary Se had increased plasma Se (P <

0.05) as Se in ewe diets increased. No signs of Se toxicosis were observed in lambs

regardless of Se concentration in the ewe diets. Wethers, fed up to 40 mg/kg Se as








sodium selenite or Se yeast for 60 wk, had Se concentrations in serum, whole blood,

wool, and soft tissues which increased as dietary Se increased (P < 0.05). In general, Se

yeast vs selenite was more effective at increasing Se in blood, wool, and soft tissues (P <

0.05). Enzyme activity and histopathological evaluation of soft tissues from ewes and

wethers indicated no evidence of Se toxicosis. From the two sheep experiments,

maximum tolerance for both forms of dietary Se is greater than 40 mg/kg. Cows

receiving Se supplementation as Se yeast maintained adequate concentrations of Se in

plasma, whole blood, and liver and generally had higher (P < 0.05) concentrations than

cows receiving inorganic Se. Calves from cows receiving Se via free-choice minerals

had higher (P < 0.05) weight gains than from cows receiving injectable selenate. Calves

whose dams received Se yeast generally had higher Se (P < 0.05) in blood and liver.













CHAPTER 1
INTRODUCTION

Selenium (Se) has had a long and storied history in animal nutrition. Since its

discovery, at the bottom of a vat of sulfuric acid, by Jbns Jacob Berzelius, a Swedish

chemist, in 1817, Se has played the role of toxic element, essential nutrient, carcinogen,

and contributor in cancer prevention. However, it seems that selenium's greatest legacy

is one of a toxic agent to livestock. As early as 1295, Se was documented as detrimental

as Marco Polo described a poisonous plant which, when eaten by horses, caused their

hooves to drop off (Komroff, 1926). Likewise, a U.S. Army surgeon, stationed at Fort

Randall in 1856, described much the same conditions afflicting horses in the Nebraska

Territory (Madison, 1860). Selenium was identified as the principal toxic agent in

conditions described as "blind staggers" and "alkali disease" throughout Wyoming and

the Dakotas. In 1957, Se was shown to prevent liver necrosis in rats and afterward was

deemed an essential nutrient (Schwarz and Foltz, 1957). Though much of the world is

afflicted with Se deficiency, supplementation of Se using dietary or parenteral forms will

generally resolve the problem. Selenium toxicities require more effort but can be

successfully combated, by not overdosing livestock with supplemental Se, monitoring Se

content of feedstuffs, and by using certain animal management techniques.

With its many implications as a toxic element, the use of Se, as a supplement to

livestock, garners much caution from feed manufacturers, animal scientists, and

nutritionists. The current estimate of the maximum tolerable level for dietary Se in








domestic animals is 2 mg/kg (National Research Council [NRC], 1980). This estimate

does not consider differences inmetabolism of Se by different species and makes no

differentiation in the maximum tolerable level for the different chemical forms of Se,

such as Se yeast or sodium selenite. Previous research has shown that the absorption of

an oral dose of inorganic Se differs between ruminant and monogastric species (Wright

and Bell, 1966). Likewise, studies in cattle and swine have shown a marked difference in

the efficacy of organic vs inorganic Se to increase blood, milk, and tissue Se

concentrations (Pehrson et al., 1999; Kim and Mahan, 2001; Gunter et al., 2003).

Furthermore, some evidence exists to suggest that the maximum tolerable level of Se for

livestock is grossly underestimated and to discredit the notion that the range between

optimal and toxic levels of Se is narrow (Glenn et al., 1964a; Kim and Mahan, 2001;

Cristaldi et al., in press).

To further the body of knowledge in this subject area, a series of experiments were

carried out with sheep and cattle. Experiments using sheep were conducted to gather

further data on 1) the amount of dietary inorganic Se that can be tolerated by ewes during

lamb production, 2) the effects of Se supplementation to ewes on their lambs, and 3) the

amount of organic or inorganic Se that can be tolerated by mature wethers. A cow-calf

herd was utilized to evaluate and compare effects of using different forms of dietary or

injectable Se on body weight change and blood, milk, and liver Se concentrations of beef

cows and their calves.













CHAPTER 2
REVIEW OF LITERATURE

Benefits of Selenium Supplementation to Livestock

Selenium's role in animal nutrition was drastically changed when it was identified

as the third factor involved in preventing liver necrosis in rats (Schwarz and Foltz, 1957).

After this first evidence for the essentiality of Se, benefits for many other species were

discovered. Patterson et al. (1957) demonstrated that Se would prevent exudative

diathesis in chicks and Eggert et al. (1957) showed that hepatosis dietetica (liver necrosis)

could be prevented in swine by feeding Se. In calves and lambs, Se was successful in

preventing white muscle disease (WMD), a condition also known as nutritional

myodegeneration (Hogue, 1958; Muth et al., 1958).

Corah and Ives (1991) reported that insufficient Se could be linked to a variety of

disorders in beef cattle. Among the reproductive disorders observed were retained

placenta, infertility, abortions, births of premature, weak, or dead calves, cystic ovaries,

metritis, delayed conception, erratic estrus periods, and poor fertilization. In addition to

problems in reproduction, a condition known as "ill-thrift" has also been reported in

cattle (Corah and Ives, 1991; Underwood and Suttle, 1999) and also affects sheep. "Ill-

thrift" is defined as a syndrome that includes subclinical growth deficit, clinical

unthriftiness with rapid loss in weight, as well as some mortality. Selenium deficiency

has also been linked to cases of mastitis in dairy cattle that occurred more frequently and

lasted longer than mastitis in cattle with adequate Se intake (Smith et al., 1985).








Perhaps Se is best known for its role as an essential constituent of glutathione

peroxidase (Rotruck et al., 1973) and four Se-dependent glutathione peroxidases have

been identified and designated as glutathione peroxidase 1, 2, 3, and 4 (Lei et al., 1998).

These four enzymes benefit animal health by protecting cellular and subcellular

membranes against oxidative damage. Also, it appears that Se-dependent glutathione

peroxidases provide a second line of defense against peroxidation of vital phospholipids

(McDowell, 2003). Vitamin E provides the first line of defense against the peroxidation

of phospholipids in membranes.

Adequate dietary or supplemental Se is an effective way to combat the

aforementioned problems in growth and reproduction of livestock. Likewise, in the

presence of adequate Se, the glutathione peroxidase system works in synergy to protect

animal cells against lipid peroxidation. In general, livestock species have minimum

requirements of dietary Se which range from 0.05 to 0.30 mg/kg (McDowell, 2003).

There are numerous examples throughout the scientific literature that cite benefits in

growth, reproduction, and prevention of WMD and other anomalies due to adequate

dietary or supplementary Se.

The dietary Se requirement for all classes of sheep ranges from 0.10 to 0.20 mg/kg

(NRC, 1985). However, the minimum dietary Se level necessary to prevent WMD varies

as reported in the literature. Oldfield et al. (1963) reported that 0.06 mg/kg was the

minimum dietary Se level required to prevent WMD in lambs. However, researchers in

New Zealand indicated that lambs had normal growth and remained free of clinical signs

of Se deficiency when grazing pastures containing 0.03 to 0.04 mg/kg (Hartley and

Grant, 1961). Oldfield et al. (1963) further reported that ewes fed a ration containing








only 0.02 mg/kg Se gave birth to lambs with WMD, but by supplementing 0.10 mg/kg Se

in the ewe diet, WMD was prevented consistently. It is suggested that at least some of

the variation in the Se requirements necessary to prevent WMD is due to sparing or

interfering nutrients, such as Vitamin E or sulfur, and that differences also reflect Se

losses due to drying, as well as analytic error (McDowell, 2003). Maas et al. (1984)

suggested that even in cases of Se deficiency, lambs returned to a normal Se status with

one or two i.m. injections containing 1 mg Se and 68 IU of vitamin E. Selenium

supplementation to ewes at a level of 2.25 mg/d reduced both the incidence and severity

of WMD in white-faced lambs (Gardner and Hogue, 1967).

Selenium supplementation has been reported to have some effects on growth and

rate of gain in sheep and cattle. Spears et al. (1986) reported increased summer gains in

calves that received Se and vitamin E supplementation vs those calves receiving no Se

supplementation. Likewise, Perry et al. (1976) reported a 10% increase in ADG of steers

when feedlot diets were supplemented with 0.1 mg/kg Se. Furthermore, an 8% increase

in ADG of finishing beef cattle was again reported when 0.1 mg/kg Se was added to the

diet (Burroughs et al., 1963). Increases in BW gains of 20% were attained when Friesian

heifer calves were supplemented with Se at a rate of 3 mg/d (Wichtel et al., 1996). In

lambs, data from Oldfield et al. (1963) indicated that lambs with the lowest blood Se had

the lowest BW at six wk of age. However, reports of a positive response in growth or

BW gain are inconsistent in sheep and cattle. Ammerman et al. (1980) reported no

differences in weaning weights of calves nursing Se supplemented mothers vs calves

whose dams had received no supplemental Se. Hereford x Angus calves showed no

difference in ADG from birth to weaning due to supplemental Se (Castellan et al., 1999).








Likewise, ADG, feed consumption, and gain:feed were not affected by supplementation

of 0.1 mg/kg dietary Se fed for 13 wk or 0.2 mg/kg dietary Se fed for six wk in separate

studies, two using sheep and one using cattle (Ullrey et al., 1977).

Newborn and suckling calves and lambs can receive Se via their dams from either

maternal transfer or increased Se in milk. Recent studies indicate that blood Se in

newborn calves can be increased through Se supplementation of their dams

(Abdelrahman and Kincaid, 1995; Gunter et al., 2003; Valle et al., 2003). Likewise,

positive correlations between Se concentration in dam's milk and Se concentration of calf

whole blood have been observed in calves up to 70 d of age (Pehrson et al., 1999).

Evidence also exists that milk Se can be increased by level and duration of Se

supplementation in lactating cows (Conrad and Moxon, 1979). Like blood and tissue,

milk Se is affected by dietary Se level (Conrad and Moxon, 1979; Givens et al., 2004)

and Se readily crosses the placenta to the fetus (Van Saun et al., 1989). A strong

relationship of dietary Se to Se in milk of dairy cows was reported with up to 18.08% of

dietary Se being recovered in milk (Maus et al., 1980). Koller et al. (1984) supplemented

first-calf Hereford heifers with dietary Se and concluded that Se readily crosses the

placenta in beef cattle. Furthermore, those authors added that low Se concentrations in

the blood of dams could cause the fetus to gather more Se and result in fetal blood Se that

is higher than that of the mother. In sheep, Cuesta et al. (1995) showed increased

colostrum Se from ewes receiving supplemental Se and that milk Se was higher after one

mo of lactation. Also, Jacobsson et al. (1965) concluded that Se administered to ewes

could be transmitted to lambs through the placenta and the milk after a study using radio-

labeled sodium selenite and selenomethionine. In a study utilizing swine, Wuryastuti et








al. (1993) documented the importance of Se and vitamin E for maintaining immune

function in livestock. Those authors measured immune responses of blood, colostrum

and milk leukocytes of sows and concluded that greater phagocytic and microbicidal

activity could be realized in milk and colostrum through supplementation with Se and

vitamin E.

Reproductive problems in beef cattle such as retained placenta, infertility,

abortions, births of premature, weak, or dead calves, cystic ovaries, metritis, delayed

conception, erratic estrus periods, and poor fertilization may be successfully overcome

with Se supplementation (Corah and Ives, 1991). Awadeh et al. (1998a) concluded that

Se intakes of pregnant cows could be an important factor in weak calf disorders and that

passive immunity and heat production by newborn calves using brown adipose tissue

could both be influenced by maternal Se intakes. A Se deficiency in the diet of dairy

cattle was reported to be a contributor to a high incidence of retained placentas (Trinder

et al., 1973). Data from studies using dairy cows have shown that supplemental Se and

vitamin E to animals receiving Se deficient diets are beneficial in decreasing the

incidence of retained placentas (Julien et al., 1976; Hemken et al., 1978). Smith et al.

(1988) studied the effects of Se on disease resistance in dairy cattle and concluded that

many dairy herds have inadequate dietary intakes of Se and vitamin E. Those authors

added that insufficient intakes of these nutrients could result in increased cases of

mastitis, metritis, and retained placenta, and recommended Se supplementation at a level

to maintain blood Se at a minimum of 200 gg/L. Weiss et al. (1990) studied the

relationships between Se and mammary gland health in commercial dairy herds and








concluded that high serum Se concentrations were associated with reduced rates of

clinical mastitis and low somatic cell counts in the milk tank.

Reproductive problems such as increased services per conception or increased

calving interval which could, at least in part, be attributed to male fertility may also be

improved by Se supplementation. Heimann et al. (1981) showed that the pituitary gland

and reproductive tissues exhibited higher Se concentrations than many other body tissues.

Julien and Murray (1977) reported that percent motility in bovine spermatozoa increased

significantly as concentration of Se in sperm increased. Thus, supplemental Se may have

a positive effect on sperm quality and ultimately on male fertility. However, Segerson

and Johnson (1981) observed no differences in sperm number, viability, or Se content

from Se supplemented bulls compared to sperm from unsupplemented controls.

Methods of Selenium Supplementation to Livestock

The benefits of Se supplementation to livestock are many and Se deficiencies are

easily combated with adequate Se supplementation. Several methods of Se

supplementation exist and successful uses of all methods have been reported. The

method of Se supplementation chosen by livestock producers may be dependent on

factors such as Se content of soils, local grains and forages, species produced, class of

livestock and stage of production, facilities for animal handling, as well as knowledge,

previous experience, and personal preference.

Many areas of the United States have Se deficient soils (McDowell, 2003) and thus

produce grains and forages which are low in Se. Likewise, many regions of the world

have been mapped as Se deficient and may benefit from the administration of Se to

livestock (Oldfield, 2002). In a survey of blood Se status in beef cattle encompassing

more than 250 herds in 18 states in several regions of the U.S., more than 18% of cattle








were classified as marginally deficient (51 to 80 gg/L) or severely deficient (< 50 gg/L)

in blood Se (Dargatz and Ross, 1996). Percentages of cattle classified as deficient varied

with region of the country. Herds in the Central U.S. had the least occurrence of Se

deficiency, while the Southeast, including Florida, had the greatest incidence of Se

deficiency at more than 40%. Stowe and Herdt (1992) also suggest that many cattle in

the U.S. are in a state of Se deficiency.

Selenium supplementation to livestock is accomplished using three or four primary

methods. Addition of Se to livestock feeds and/or minerals, use of injectable Se

preparations (usually in combination with vitamin E), use of sustained release

intrareticular Se supplements, and possibly the use of seleniferous grains or forages

grown on high Se soils (Ammerman and Miller, 1975) are the methods most often used to

supplement Se. One additional option to increase Se intake of livestock is the use of Se

containing fertilizers on forage and pasture (Valle, 2001). The addition of Se to

feedstuffs was not an option until 1974 when the Food and Drug Administration (FDA)

allowed for supplementation of up to 0.1 mg/kg Se as selenite or selenate for swine and

poultry (Schmidt, 1974). An amendment to this FDA order allowed the use of

supplemental Se for sheep in 1978 and a subsequent amendment in 1979 allowed for use

in dairy and beef cattle. Currently, use of 0.3 mg/kg dietary Se is approved for

supplementation in poultry, swine, sheep, and cattle (McDowell, 2003).

Regardless of method chosen for Se supplementation, Se deficiencies are more

easily combated than are toxicities, which generally require more animal and/or pasture

management. In sheep and beef cattle production systems, producers most often choose

to use injectable Se products or supplement Se through free-choice mineral mixtures.








Judson et al. (1991) evaluated long-acting Se treatments for ewes and lambs in a 200 wk

experiment. Those authors reported that a 100 mg injection of barium selenate was more

effective at increasing blood Se of ewes and their lambs than was an intraruminal Se

pellet or no Se supplementation. Data show a near five-fold increase in blood Se from

lambs from injectable selenate treated ewes vs lambs from unsupplemented dams.

Norton and McCarthy (1986) evaluated injectable Se products for prevention of WMD in

lambs and reported increased plasma and milk Se in ewes that received the injectable Se

vs unsupplemented controls. Likewise, those authors showed increases in lamb plasma

Se due to the frequency of use of injectable Se. In a series of University of Florida

studies, the use of injectable Se, as sodium selenite and barium selenate, in a cow-calf

herd was evaluated and compared to inclusion of organic Se in free-choice mineral

mixtures (Valle et al., 2002; 2003). Those authors reported that, in general, injectable Se

as selenate and selenite affected plasma, liver, colostrum, and milk in a similar manner.

Though the injectable products did increase Se levels in blood, milk, and tissue compared

with blood, milk, and tissue Se concentrations from unsupplemented animals, both

injectable forms of Se were generally less effective than the addition of organic Se to

free-choice minerals. The calves born to and suckling cows that received injectable Se

had plasma Se concentrations which were similar to plasma Se concentrations of calves

from unsupplemented dams. Selenium supplementation via free-choice minerals proved

more effective at raising and maintaining Se status of Florida beef cows and their calves.

Gunter et al. (2003) compared effects of Se supplementation as sodium selenite or

Se yeast added to free-choice minerals on performance and Se status of beef cows and

calves in Arkansas. Mineral mixtures were formulated to contain 26 mg/kg Se and were








offered free-choice. No differences in performance between unsupplemented controls or

cattle receiving either form of Se were observed. However, differences in blood Se of

supplemented vs unsupplemented cattle were reported. Likewise, Se yeast treated cows

and their calves had higher blood Se than cows and calves receiving selenite Se. Those

authors concluded that calves are at risk for Se deficiency if their dams are not

supplemented with Se and that even when selenite Se is provided, calves may still be at

risk. Sheep may also be supplemented with Se which is included in mineral mixtures and

salt licks. Norwegian researchers reported no incidences of WMD in lambs and

increased Se in blood and colostrum when Se fortified mineral mixtures and salt licks

were offered to ewes and lambs (Overnes et al., 1985).

In addition to the use of injectable Se or the inclusion of Se in free-choice mineral

mixtures, livestock producers may use an intraruminal or intrareticular bolus or pellet

which provides a sustained release of Se. Judson et al. (1991) reported that the use of an

intraruminal Se pellet and steel grinder increased blood Se of ewes and lambs compared

with controls. However, the Se pellet and grinder system was not as effective as an

injection of barium selenate at increasing blood Se. Campbell et al. (1990) used

crossbred beef cows to evaluate the safety and efficacy of Se boluses and Se pellets.

Both methods of Se supplementation were shown to be both safe and effective; however,

blood Se of cows receiving either method of Se supplementation increased until d 119 of

the study and was decreased by d 220. As in previous studies, both methods produced

blood Se higher than the blood Se from unsupplemented controls. Abdelrahman and

Kincaid (1995) evaluated the effects of administration of an intraruminal Se bolus on

colostrum, plasma, and whole blood Se concentration of dairy cows. Those authors








reported that the Se bolus was an effective method of increasing Se in colostrum, plasma,

and whole blood. Likewise, calves born to Se supplemented cows had higher Se

concentrations in plasma, whole blood, and liver than calves born to cows receiving no

supplemental Se. In this study, the administration of a sustained release Se bolus to cows

proved to be an effective method of Se supplementation to newborn calves.

The need for Se supplementation to livestock is great as evidenced by the many

benefits of supplemental Se on animal health and performance. This need is further

elucidated by surveys such as reported by Dargatz and Ross (1996), which reported a

relatively high percentage of beef cattle in the U.S. that were classified as Se deficient.

Selenium supplementation generally adds only a negligible amount to the cost of

livestock production and producers have several effective means of Se supplementation

to choose from.

Absorption, Transport, Storage, and Excretion of Selenium

Ruminant animals differ from monogastric animals in their ability to absorb and/or

retain Se. Wright and Bell (1966) reported retention of a dose of sodium selenite to be

29% for sheep and 77% for swine. In both sheep and swine, Se absorption occurred in

the small intestine and cecum with some additional absorption in the colon for swine. No

absorption of Se occurred in the rumen of sheep or the stomach of swine (Wright and

Bell, 1966). These authors also reported net absorption of Se to be 36% for sheep and

86% for swine. Less absorption of Se in ruminants seems to be due to the reduction of

inorganic Se to insoluble forms by rumen microorganisms (Butler and Peterson, 1961;

Peterson and Spedding, 1963; Hidiroglou et al., 1968). Inorganic Se is more readily

reduced within the rumen than organic forms of Se such as Se yeast. Diet also affected








Se absorption as sheep on a high concentrate diet had higher plasma Se than sheep

receiving a high forage diet. (Koenig et al., 1997).

In contrast to many other minerals consumed by livestock, which use homeostasis

as a primary status regulator, Se status of animals seems to have little effect on intestinal

absorption. In a study utilizing rats, urinary excretion was shown as the only relevant

means of Se homeostasis (Windisch and Kirchgessner, 2000) as urinary excretion is

directly related to dietary level while fecal Se excretion is quite static (Burk et al., 1972).

When Se absorption was regressed on Se intake of dairy cows, a strongly linear

relationship was observed (Harrison and Conrad, 1984). However, Se intakes reported in

that study were relatively low and ranged from 0.437 to 3.136 mg/d. Most dairy cows

consume closer to 6 mg Se/d, based on supplementation in the diet of 0.3 mg/kg Se.

Absorbed Se is associated with plasma protein and transported in the blood plasma

until it enters tissues (McDowell, 2003). Selenoprotein P is the plasma protein with

which most Se is associated in individuals with adequate or deficient dietary Se, while

most plasma Se is associated with albumin when Se intake is in excess (Xia et al., 2000).

In addition to plasma, Se is also found in muscle and glandular tissues.

Generally, when ranked on a Se concentration basis, tissues follow the general order of

kidney > liver > heart > skeletal muscle, regardless of species, when Se is fed at an

adequate or deficient level (Comb and Combs, 1986). The kidney may be the highest in

Se concentration as it is primary organ of excretion. However, when Se is fed at levels

above requirement, liver surpasses kidney in terms of Se concentration (Cristaldi et al., in

press).








Urine, feces, and exhalation are the primary excretion routes of Se. Amount and

distribution of excreted Se within these routes are affected by chemical form of Se, total

Se intake, and diet composition including antagonists (McDowell, 2003). Urine is the

major excretory pathway and Se excretion via urine increases with Se status of the

animal. Fecal excretion remains nearly constant and exhalation of Se becomes a major

route only when Se concentrations are at a toxic level (McDowell, 2003). The amount of

Se exhaled increases as dietary Se increases (McConnell and Roth, 1966) and one

characteristic of animals which excrete Se via respiration is breath with a garlicky odor.

Selenium excretion in ruminant animals is dependent on method of administration.

When Se is provided orally, ruminants excrete more Se in feces. However, when Se is

given parenterally, more Se is excreted in urine (Wright and Bell, 1966). This is

supported by the concept that rumen microorganisms reduce dietary Se to insoluble forms

(Butler and Peterson, 1961; Peterson and Spedding, 1963; Hidiroglou et al., 1968) and

thus increase fecal excretion of unabsorbed Se.

Differences in Efficacy of Selenium due to Source

The efficacy of Se to increase blood and tissue Se concentrations in animals varies

with source of Se. In general, Se is deposited in tissues and blood Se is more increased

when supplemental Se is of the organic form (McDowell, 2003). The primary sources of

inorganic Se are sodium selenite and sodium or barium selenate, while Se yeast and

seleniferous grains and plants are the primary sources of organic Se. Sodium selenite and

selenate are often added to free-choice mineral mixtures for livestock. Likewise, those

two chemical forms are used in injectable Se products. Selenomethionine is the major Se

compound found in grains used for livestock feeds and in Se yeast. Se-

methylselenocystine is the Se compound found most abundantly in seleniferous plants,








while some inorganic Se is found in grains and plants (Whanger, 2002). In animal tissues,

selenate is the major inorganic form and selenocystine is the predominant organic form.

Selenomethionine is found initially when this amino acid is fed; however seleno-

methionine is converted to selenocystine after some time (Whanger, 2002). With such

differentiation in the sources of Se within plant and animal tissues, it seems reasonable

that differences in efficacy due to form of Se administered would exist and there are

numerous examples in the scientific literature to support this concept.

Goehring et al. (1984a) evaluated the effects of high dietary levels of Se from

selenite or seleniferous grains on blood and tissue concentrations in swine. Those authors

documented that Se from seleniferous grains increased Se in blood and tissue compared

to selenite Se fed at the same level. Awadeh et al. (1998a) reported increased blood Se in

crossbred beef cows consuming free-choice minerals containing 60 mg/kg Se as Se yeast

compared to 60 mg/kg selenite Se. Furthermore, cows receiving free-choice minerals

containing 60 mg/kg Se as Se yeast had a lower percentage protein in albumin compared

to cows receiving minerals with 60 mg/kg selenite Se. Selenium from Se yeast has been

documented by several groups of researchers as more effective than Se selenite or

selenate at increasing blood and liver Se levels in beef cows (Pehrson et al., 1999; Valle

et al., 2002; Gunter et al., 2003) and in dairy cattle (Ortman and Pehrson, 1999).

As with blood and tissue Se, milk Se has been more effectively increased by using

organic Se vs inorganic Se in beef cattle, dairy cattle, and swine. Selenium yeast

produced milk Se more than 100% higher than selenite or selenate Se when 3 mg of Se

from each source were fed to Swedish dairy cows (Ortman and Pehrson, 1999). Hereford

cows supplemented with Se yeast produced milk with markedly higher Se concentrations








than did cows receiving supplemental selenite Se in early and late lactation (Pehrson et

al., 1999). In a two-yr study utilizing Florida beef cows, milk Se was consistently higher

from cows receiving free-choice minerals with Se yeast compared to cows receiving Se

as selenite or selenate injections (Valle et al., 2002). Also, calves suckling the cows

which received the organic Se had higher Se concentrations in plasma and liver (Valle et

al., 2003). Researchers at Ohio State University fed Se as Se yeast or sodium selenite at

dietary levels of 0.15 and 0.30 mg/kg to gestating and lactating sows. Colostrum and

subsequent milk Se concentrations were consistently at least two-fold higher from sows

receiving organic Se than from sows receiving selenite Se (Mahan, 2000). Data from

New Zealand indicated that the transfer of Se into cows' milk was markedly more

efficient, up to three-fold more, with selenized yeast than with sodium selenate (Knowles

et al., 1999).

The effectiveness of different sources of Se for supplementation continues to be

evaluated even though Se has been recognized as nutritionally essential since the late

1950s. Selenium provided by different supplementation methods and from different

sources leads to different physiological responses in the animals that serve mankind.

With evidence of an increasing ability to manipulate the Se content of milk and animal

tissues which are commonly consumed by humans, it seems to be possible to supplement

Se to humans through method and source of Se supplementation to livestock. Givens et

al. (2004) reported a decline in Se intake by the people of Great Britain. After

conducting an experiment which validated previous findings that the milk of dairy cows

could be increased by feeding an organic Se source, those authors further explored the

idea of increasing human consumption of Se by altering the Se content of foods. It seems








that a worthy challenge exists for animal scientists and food scientists to work

collaboratively to identify effective programs for administration of Se to dairy and food

animals so that the milk and meat subsequently produced can be more nutritious for

humankind.

Selenium Toxicosis

Selenium deficiencies are prevalent in many parts of the world (McDowell, 2003)

and the benefits of Se supplementation continue to be elucidated. However, it seems that

Se is still most often implicated as an element which is toxic to livestock. This belief

most likely stems from diary-style documentation, observations, and research findings

beginning as early as 1295 when Marco Polo described an agent in plants which when

eaten by horses caused their hooves to fall off (Komroff, 1926). Six hundred years later

similar afflictions began to be described in the Great Plains region of the United States.

In 1856, a U.S. Army surgeon reported the occurrence of a disease, fatal to U.S. Cavalry

horses, similar to the affliction described by Marco Polo (Madison, 1860). The horses in

the Nebraska Territory near Fort Randall lost hair and had debilitating conditions of the

hoof. By the 1890s, farmers and stockmen who settled in northern Nebraska and South

Dakota observed similar conditions in livestock (Moxon and Rhian, 1943). Selenium

toxicity in livestock and laboratory animals has been reported from the 1930s to the

present day. Some reports were observations of animals receiving seleniferous grains or

grazing seleniferous plants (Franke, 1934; Franke and Potter, 1935; Moxon, 1937). Other

researchers have intentionally induced or attempted to induce Se toxicities, while several

reports of Se toxicity are a result of accidental overdosing with injectable Se.

Rosenfeld and Beath (1964) suggested that Se poisoning in livestock occurs in

three distinct phases: acute toxicity and the two phases of chronic toxicity, alkali disease








and blind staggers. Acute Se toxicity can be caused by ingesting a large amount of

supplemental Se, overdosing with parenteral Se, or by ingesting a large amount of

seleniferous plants. Certain plants, mostly species of Astragalus, may contain up to

10,000 mg/kg Se and cereal crops, grasses, and other forages may contain up to 50 mg/kg

Se (Aitken, 2001). Fatalities of sheep, cattle, and hogs have been reported in regions

known to grow seleniferous plants (National Academy of Sciences [NAS], 1983), with

some deaths occurring within 24 h (Rosenfeld and Beath, 1964). Clinical signs of acute

Se toxicity may include elevated body temperature, labored breathing, diarrhea, and often

death. Alkali disease and blind staggers types of Se toxicosis occur with more time and

involve feedstuffs containing less Se. Clinical signs of chronic Se toxicity include

anorexia, apathy, diarrhea, weight loss, hair loss, and hoof malformations (Glenn et al.,

1964a). Animals in the blind staggers phase of Se toxicosis may wander, stumble, and

lack appetite initially and then become somewhat paralyzed and almost blind in the later

stage. The later stage appears suddenly and death usually occurs within hours (Rosenfeld

and Beath, 1964). Chronic forms of selenosis have been induced by feeding grains

containing 5 to 40 mg/kg Se (Schoening, 1936; Rosenfeld and Beath, 1964).

Glenn et al. (1964a) induced death in ewes after oral dosing of up to 50 mg/d of Se

as sodium selenate for 93 d and concluded that minimum toxic oral dose of Se as selenate

depended on susceptibility, level of Se administered, and duration of administration.

Those authors later reported liver Se concentrations of up to 29 mg/kg in experimentally

poisoned ewes (Glenn et al., 1964c). Evaluation of the tissues of the ewes in the previous

study showed that most tissue damage in Se toxicosis is confined to the heart (Glenn et

al., 1964b). No kidney damage and few instances of liver damage were reported.








Blodgett and Bevill (1987) induced death in sheep by feeding 0.7 to 1.0 mg Se/kg BW as

selenite in as little as 6.75 h. Liver Se concentrations of more than 17 mg/kg and whole

blood Se of 2.7 mg/L were reported. After receiving an oral 5 mg selenite Se/kg BW,

lambs died within 6 h and when the same dosage was given as an injection, lambs lived

up to 60 h (Smyth et al., 1990). After evaluating the major organs, those authors

concluded that the heart is most damaged in a case of Se toxicosis as the heart has great

affinity for Se especially in lethal doses. Caravaggi et al., (1970) injected Merino lambs

with 0.425 to 0.500 mg Se/kg BW, induced death, and determined the LD50 for lambs to

be 0.455 mg Se/kg BW. Twenty lambs received 10 mg of selenite Se orally in an attempt

to prevent WMD. Of those 20, seven died within 16 h, eight developed diarrhea but

recovered, and five lambs were apparently unaffected (Marrow, 1968). Cristaldi et al.

(2004) fed up to 10 mg/kg dietary Se as selenite to growing wether sheep and reported no

signs of Se toxicity. Those authors reported whole blood Se concentrations of up to 1.2

mg/L, wool Se of 2.5 mg/kg, and liver Se concentrations of nearly 15 mg/kg Se on a dry

basis and no evidence of Se toxicity from histopathological evaluation.

Selenium toxicity studies have also been conducted using swine. Goehring et al.

(1984b) fed young pigs up to 20 mg/kg Se as sodium selenite for 5 wk. No pigs on the

study died; however, feed intake and growth rate decreased as dietary Se concentration

increased. Whole blood Se concentrations of up to 3.5 mg/L and hair Se of more than 11

mg/kg were observed. Organic and inorganic Se was included in the diet of growing pigs

at levels of up to 20 mg/kg for 12 wk (Kim and Mahan, 2001). Feed intakes of those pigs

declined as Se level increased and daily gains were decreased when Se was fed at more

than 5 mg/kg, and inorganic Se had a more detrimental effect on performance than did








organic Se. Those authors reported plasma Se concentrations of more than 3.3 mg/L,

liver Se of more than 17 mg/kg, and hoof Se of more than 28 mg/kg and no deaths

regardless of dietary Se level. The authors concluded that the higher retention of organic

Se in tissues and blood cells may effectively reduce the amount of Se available to cause

Se toxicosis.

Tolerance of Se as selenite or selenomethionine was evaluated using yearling steers

in a 4 mo study (O'Toole and Raisbeck, 1995). Those authors observed the highest

incidence of hoof lesions in steers fed organic Se at a rate of 0.80 mg/kg BW. Likewise,

it was shown that the steer with the most severe hoof lesions also had the highest Se

concentrations in hair, liver and kidney. The findings of that study indicated that dietary

exposure of 0.8 mg Se/kg BW, in either form, for 4 mo produces subclinical to clinical

signs of Se toxicosis. The authors concluded that selenomethionine is more likely to

cause alkali disease than sodium selenite. Holstein cows were fed inorganic Se up to 100

mg/d and whole blood and liver Se concentrations of up to 4.9 mg/L and 15 mg/kg DM,

respectively, were reported (Ellis et al., 1997). It was concluded that dairy cattle could

tolerate Se intakes of up to 100 mg/d for several wk without suffering adverse affects.

It seems logical, based on previous findings, that the minimum lethal dosages and

maximum tolerable levels of Se are variable and may be affected by various factors such

as Se source (organic or inorganic), diet composition, method of Se supplementation, and

Se status of the animal. The current maximum tolerable level for dietary Se in domestic

animals is 2 mg/kg (NRC, 1980). This estimate does not consider differences in species,

source of Se, or duration of exposure. Early reports of toxicities are likely reasons for the

conservative estimate of maximum tolerable level and the notion that the range between





21

optimal and toxic level of Se is narrow. Surely, further studies which are long in duration

and use high dietary levels and different sources of Se are necessary to better estimate the

tolerance of Se for dairy and food animals.












CHAPTER 3
TOLERANCE OF INORGANIC SELENIUM IN RANGE-TYPE EWES DURING
GESTATION AND LACTATION

Introduction

Since its discovery by Berzelius in 1817, Se has had a rich and colorful history in

animal agriculture. Though much of the world is troubled with Se deficiencies

(McDowell, 2003), Se toxicities present a greater problem to control. In 1957, Se was

established as an essential nutrient and the benefits of Se supplementation to livestock

continue to be elucidated. Current estimates put the maximum tolerable level of Se at 2

mg/kg for the major livestock species (NRC, 1980) and no differentiation exists for

tolerable levels between ruminants and monogastric animals. However, the work of

Butler and Peterson (1961) and Hidiroglou et al. (1968) suggests that inorganic Se (e.g.,

sodium selenite) may be reduced to insoluble selenide by microorganisms in the rumen,

thus reducing overall absorption of Se by ruminant animals. Wright and Bell (1966)

reported that swine retained 77% of an oral dose of inorganic Se, which is nearly three-

fold the retention by sheep. Selenium toxicities have been often produced by researchers

in ruminants, but they are generally induced by Se injections (Marrow, 1968; Caravaggi

et al., 1970; Shortridge et al., 1971) or by feeding Se above maximum tolerable levels (5

to 196 ppm) to monogastric animals (Franke and Potter, 1935; Miller and Schoening,

1938; Kim and Mahan, 2001). More recently, Cristaldi et al. (2004) demonstrated that

wether sheep did not display signs of Se toxicosis after receiving up to 10 mg/kg dietary

Se for one yr. Based on these and other previous findings, it seems that the current

maximum tolerable level of Se for ruminants is underestimated. Most Se toxicity








research in ruminants has been documented in lambs or wethers. Controlled experiments

using ewes during stresses of production (e.g., gestation and lactation) are lacking in the

scientific literature. The objective of this long-term (72 wk) study were to evaluate and

compare effects of feeding Se as sodium selenite at supranutritional levels on ewe serum,

blood, wool, and tissue Se concentrations during two lambing periods and to determine

maximum tolerable level of Se.

Materials and Methods

All animal procedures were conducted within the guidelines of and approved by

the University of Florida Institutional Animal Care and Use Committee. This experiment

utilizing ewes during two lambings was conducted from December 18, 2001 to May 5,

2003 at the University of Florida Sheep Nutrition Unit located in North Central Florida.

Forty-one, four-yr-old, Rambouillet ewes, that originated from a single range flock in

Texas and had been pasture exposed to rams during October and early November 2001

(average 57 d gestation), were weighed (57.4 5.7 kg) and administered 2-ml ivermectin

dewormer s.c. (Ivomec; Merial Ltd., Iselin, NJ). Ewes were randomly assigned to one of

six dietary treatments for a 72-wk study. Six dietary treatments were 0.2, 4, 8, 12, 16, or

20 mg/kg Se as sodium selenite (as-fed basis) added to a corn-soybean meal basal diet

(Table 3-1). The basal diet was formulated to meet animal requirements for protein,

energy as TDN, vitamins, and minerals for this class of sheep (NRC, 1985). Animal

numbers per treatment were six for 0.2 (control) and seven each for 4, 8, 12, 16, and 20

mg/kg added Se treatments. Ewes were housed by treatment group in covered wooden

pens (53.5 m2) with earth floors and ad libitum water.








Diets were fed at 909 g-ewe'ld' from d 0 until lambing began, increased to 1000

g'ewe-l'dl during lambing, and again increased to 1135 g-ewe-'-d-1 during lactation. Ewes

received 909 g'ewe-l*d of their respective diets after the first lamb crop was weaned. On

August 15, 2002, ewes were pen exposed to rams for 35 d. Diets were offered at the

same increments during the second lambing and lactation as during the first. Diets were

sampled every 28 d, ground (1 mm), and frozen at 0C until analysis.

Ewe BW was recorded on d 0 and for every four wk thereafter, for the remainder

of the study. A 10-mL blood sample for serum analysis was collected using an 18-gauge

needle into a vacutainer tube with no additive (Vacutainer; Becton Dickinson, Franklin

Lakes, NJ) every four wk, via jugular venipuncture, allowed to stand for 20 min,

centrifuged at 700 x g for 25 min, and serum stored frozen at 0C until Se analysis.

Starting at wk 12, an additional 10-mL blood sample was collected into a heparinized

vacutainer tube (Vacutainer; Becton Dickinson, Franklin Lakes, NJ). This additional 10-

mL sample was collected every 12 wk for the remainder of the experiment and stored

frozen at 0C as whole blood until analysis.

The wool around the jugular was shorn initially and regrowth was collected

beginning at wk 12 and every 12 wk thereafter. The collected wool was washed with a

commercial hair shampoo (Alberto VO5; Alberto-Culver Co., Melrose Park, IL), to

remove oil and dirt, rinsed well with deionized water, dried, stored at room temperature,

and later analyzed for Se concentration.

At the termination of the experiment (wk 72), ewes were slaughtered following

approved USDA procedures at the University of Florida Meats Laboratory. An

additional 1 0-mL sample of blood was collected using an 18-gauge needle into a








vacutainer, centrifuged at 700 x g for 25 min, and serum frozen at 0C for analysis of

albumin and the following enzymes: alkaline phosphatase (Alk Phos), alanine

transaminase (ALT), aspartate transaminase (AST), creatinine phosphokinase (CK), and

gamma glutamyl transferase (GGT). Albumin and the enzymes were analyzed in order to

determine possible tissue breakdown as a result of Se toxicosis.

Samples of brain, diaphragm, heart, hoof tip, kidney, liver, and psoas major

muscle were collected, and frozen (00) until analyzed for Se. Sections (1 cm3) of liver,

heart, kidney, diaphragm, and psoas major muscle from all animals were placed in 10%

neutral-buffered formaldehyde for subsequent microscopic evaluation for evidence of Se

toxicosis.

For histopathological evaluation, the tissue samples fixed in buffered formalin

were embedded in paraffin and sectioned at 6 microns. All sections were stained with

hematoxylin and eosin, and examined under a light microscope (lOX, 20X, and 40X).

Serum albumin, Alk Phos, ALT, AST, CK, GGT were evaluated on a Hitachi 911

analyzer with reagents from Sigma (Sigma Chemical Co., St. Louis, Mo.). These

procedures were established by the Veterinary Medical Teaching Hospital at the

University of Florida.

Serum, whole blood, wool, tissue, and feed samples were analyzed for Se

concentration using a fluorometric method described by Whetter and Ullrey (1978). To

help ensure reliability of the analytical method, a certified standard (National Bureau of

Standards Bovine Liver SRM-1577a; U.S. Department of Commerce, National Institute

of Standards and Technology, Gaithersburg, MD) was frequently analyzed.








Brain, diaphragm, heart, hoof tip, kidney, liver, and psoas major muscle Se data

were analyzed for effects of treatment using PROC GLM in SAS (SAS for Windows 8e;

SAS Inst., Inc., Cary, NC) in a completely randomized design. Pre-planned orthogonal

contrast statements were used to compare means as described by Littell et al. (1998;

2000). PROC MIXED of SAS was used to analyze effects of treatment, time, and the

interaction of treatment x time on BW, serum Se, whole blood Se, and wool Se as

repeated measures with a spatial power covariance structure with respect to d and a

subplot of animal nested within treatment. Pre-planned orthogonal contrast statements

were written to determine differences in means at different sampling intervals. Means

were separated at P < 0.05 and regression analysis was used to determine relationships

between dietary Se and Se concentration of various tissues.

Results and Discussion

Performance

Ewe BW was not affected by dietary Se level (P = 0.69) or dietary Se level x time

interaction (P = 0.56). However, time did affect BW (P < 0.001). Initial BW was 57.4

5.7 kg and BW at the termination of the experiment was 61.2 15.1 kg. These findings

agree with previous studies in ruminants. Supplemental selenium fed up to 0.4 mg/kg

which is above requirement but below maximum tolerable level had no effect on rate of

gain in feedlot steers (Perry et al., 1976) and BW gains in wether sheep, fed sodium

selenite up to 10 mg/kg, was unaffected by dietary Se level (Cristaldi et al., in press).

Glenn et al. (1964a) also reported no effect of dietary Se on BW when sodium selenate

was fed to ewes as a single oral dose of up to 50 mg/d. The ewes utilized by those

authors were very similar in breed type and BW to the animals used in the present study.








Effect of time on ewe BW can be explained by changes in BW associated with gestation

and lactation over two lambings during the study.

Ten of 41 ewes died over the course of this 72-wk study. Gross necropsies were

performed on eight ewes following death. Tissues from two ewes were too severely

decomposed to allow for evaluation for pathological changes. Necropsy of eight ewes

cited causes of death as lymphadenitis associated with injury (two ewes), endoparasitism

(two ewes), ketosis (three ewes) and pneumonia (one ewe). Pathological evidence of Se

toxicosis was not found in any ewe that died before the termination of the experiment.

In the first yr, 53 lambs were born over 20 d from March 9, 2002 to March 28,

2002. Fifty-two lambs were born alive and unassisted (Table 3-2). One lamb was very

large (8 kg) and died shortly after a difficult birth. The lambs born in yr one represent a

129% lamb crop when calculated as lambs born alive per ewe exposed. In the second yr,

36 lambs were born over 34 d from January 17, 2003 to February 20, 2003 (Table 3-2).

All lambs were born alive and unassisted. Thirty-six lambs in yr two represent a 109%

lamb crop as only 33 ewes were exposed in the second yr. Number of lambs born per

ewe did not affect serum Se concentration (P > 0.54) of ewes receiving any level of

dietary Se. Glenn et al. (1964a), who fed higher levels of dietary Se than in the present

experiment, did not observe an effect of dietary Se level on reproduction in 2-yr-old

range ewes. Those researchers observed a similar number of pregnancies in each

treatment group and no malformations in lambs. In contrast, Rosenfeld and Beath (1947)

observed lamb deformities in a field study and attributed the anomalies to excess Se in

ewe diets. However, seleniferous plants were the Se source, rather than inorganic sources

used in the present experiment. Furthermore, in a grazing situation, it is possible that








lamb deformities were due to toxic elements other than Se. In both yr of our study, all

lambs were born free of congenital deformities, but the number of pregnancies were

lowest in ewes receiving 16 mg/kg dietary Se, but not 20 mg/kg. However, breeding

soundness evaluations were not performed on ewes or rams used in this study and thus, to

incriminate or exclude dietary Se level as a detriment to ewe reproduction would be

observational.

Blood

Serum Se concentrations from wk 4, 8, and 12 were analyzed together and will be

referred to throughout the results and discussion as late gestation yr 1. Lactation yr 1

includes serum Se concentrations from wk 12, 16, 20, and 24. Week 12 is included in

both late gestation and lactation for yr 1 as some ewes were lactating and some remained

in late gestation when wk 12 sampling occurred. Weeks 28, 32, 36, 40, and 48 compose

the dry, rebreeding period. Late gestation in yr 2 includes serum Se measurements from

wk 52, 56, and 60. Lactation in yr 2 includes wk 60, 64, 68, and 72. Similar to yr 1, one

sampling date (wk 60) was common to both late gestation and lactation and was included

in both periods.

During all stages of lamb production, serum Se increased in a linear fashion (P <

0.001) as dietary Se level increased (Table 3-3). This agrees with previous Se toxicity

research as Se concentrations in serum of wether sheep (Cristaldi et al., in press) also

increased linearly as dietary selenite Se was increased. All ewes had similar (P > 0.82)

serum Se at the initiation of this experiment. Initial serum Se ranged from 90 to 120

[ig/L, which is below the normal range (120 to 180 gg/L) for adult sheep (Aitken, 2001).

A cubic response within treatment (P = 0.02) was observed in serum Se across the stages

of production (time) from wk four to wk 72. Ewe serum Se, in general, was higher








during the dry, rebreeding stage. One plausible explanation for this is the lack of

placenta, fetal tissue, and milk for deposition and excretion of Se. During late gestation

in yr 1, dietary Se level affected serum Se concentration (P < 0.001), ewes receiving 8,

12, 16, and 20 mg/kg Se all had higher (P < 0.05) serum Se than did controls. Likewise,

ewes receiving 16 or 20 mg/kg Se had serum Se higher (P < 0.05) than ewes receiving 4

mg/kg Se. During lactation in yr 1, ewes receiving 16 and 20 mg/kg Se were similar (P =

0.32) and both groups were higher (P < 0.05) than controls and ewes receiving 4 and 8

mg/kg Se in serum Se concentration. During the 20 wk that ewes were not lactating and

were either open or rebreeding, ewes receiving 16 and 20 mg/kg dietary Se had similar (P

= 0.44) serum Se which was higher (P < 0.05) than from all other treatments. Ewes

receiving the intermediate levels of Se (8 and 12 mg/kg) had similar serum Se (P = 0.31)

which was higher (P < 0.05) than from controls and ewes receiving 4 mg/kg Se. Ewes in

late gestation during yr 2 generally produced numerically higher serum Se than in late

gestation the previous yr. Ewes receiving 20 mg/kg Se had serum Se which was similar

(P = 0.69) only to serum Se from ewes receiving 16 mg/kg Se and higher (P < 0.05) than

all other treatments. Ewes receiving 16 mg/kg Se produced serum Se which tended (P =

0.07) to be higher that serum Se from ewes receiving 12 mg/kg Se and was higher (P <

0.05) than serum Se from controls and ewes receiving 4 or 8 mg/kg Se. During lactation

in yr 2, ewes receiving 20 mg/kg Se had higher (P < 0.05) serum Se than serum Se from

all other treatments. Serum Se from ewes receiving 8, 12, or 16 mg/kg Se was similar (P

> 0.20) and only serum Se from ewes receiving 4 mg/kg dietary Se was similar (P = 0.21)

to controls. Throughout the experiment, serum Se concentrations in these ewes remained

below 1500 R~g/L, which is described as a toxic level in horses (Aitken, 2001) and were at








most 37% of a reported toxic level (3700 [tg/L) in swine (Aitken, 2001). Caravaggi et al.

(1970) established an LD50 for sheep at 455 Rg/kg BW. When our data are described on a

gg/kg BW basis using the highest dietary concentration (20 mg/kg), highest daily intake

(1135 g/d), and average ewe BW (60 kg), our ewes were consuming, at maximum, 378

gg/kg BW. This is 17% less than the LD50 for sheep as previously described. The ewes

in the present study were mature and maintained healthy ruminal function throughout the

study. This is contrasted with the unweaned lambs used by Caravaggi et al. (1970).

Those lambs may have received Se via i.m. injection. Administration of Se parenterally

disallows the reduction of selenite Se to insoluble selenide via ruminal microorganisms as

described by (Whanger et al., 1968). This would suggest that the LD50 for sheep could be

considerably higher than previously thought. Glenn et al. (1964a) fed sodium selenate at

high levels to range ewes that were similar in BW to ewes on the present study. Those

researchers did not induce death by Se toxicosis with daily oral doses less than 25 mg

Se/ewe. Of the 17 deaths reported in their experiment, only one was induced with a daily

dose of 25 mg Se/ewe. Eight deaths were induced with a daily dose 37.5 mg Se/ewe and

eight deaths were induced with a daily dose 50 mg Se/ewe. Those deaths were not by

acute Se toxicosis. The ewes received experimental Se doses for at least 80 d before

death by Se toxicosis was induced. In the same experiment, Glenn et al. (1964a) further

suggested an average minimum toxic level of Se for adult sheep to be 0.825 mg/kg BW

when fed for 100 d. Using this estimate, the minimum toxic level of Se for ewes of the

size used in our study would be 50.3 mg/d. Selenium consumption, at the highest dietary

level of 20 mg/kg, never reached even 50% of that previously reported level throughout

our study. Also, Blodgett and Bevill (1987) reported an LD50 for sheep, using sodium








selenite via i.m. injection, at 0.7 mg/kg BW. Other researchers (Rosenfeld and Beath,

1946) reported death in sheep with less Se (30 mg/d); however, the Se maximum intake

level used in our study was approximately 25% less. It is important to note that we used

sodium selenite as our Se source whereas previous research (Rosenfeld and Beath, 1946;

Caravaggi et al., 1970) used sodium selenate as the source of additional Se. Henry et al.

(1988) reported a higher relative bioavailability for selenate than selenite. This suggests

the possibility of a higher tolerance for sodium selenite vs selenate.

Whole blood Se was measured at wk 12, 24, 36, 48, 60, and 72 (Table 3-4).

Dietary Se level, time, and dietary Se level x time affected (P < 0.05) ewe whole blood

Se. Whole blood Se increased linearly (P < 0.001) as dietary Se increased. Response of

whole blood Se from all treatments over time was cubic (P < 0.01) which agrees with the

time response of serum Se. Maas et al. (1992) reported a strong correlation (0.88) for

whole blood Se and serum Se. Our data support this relationship, as serum Se and whole

blood Se responded to dietary Se level in a similar fashion. The cubic response of whole

blood Se over time may be attributed to ewes having no fetal tissue and producing no

milk to use as a route of excretion during the dry, rebreeding period, which encompassed

the midpoint of this study. Each dietary Se level was evaluated individually over time

and control and 8 mg/kg neither increased nor decreased with time (P > 0.20). Whole

blood Se from ewes receiving 4 mg/kg Se responded cubically (P = 0.019) and 16 mg/kg

dietary Se tended (P = 0.07) tended to respond cubically. Whole blood Se concentration

changed more sporadically over time in ewes receiving 12 or 20 mg/kg dietary Se and

each treatment produced a fifth degree polynomial (P < 0.05). At wk 12, ewes in all

treatment groups had higher (P < 0.05) whole blood Se than did controls. Ewes receiving








20 mg/kg Se had higher whole blood Se than controls and ewes receiving 4, 8, or 16

mg/kg Se and tended be higher (P = 0.13) in whole blood Se than ewes receiving 12

mg/kg dietary Se. At wk 24, ewes receiving 20 mg/kg Se had higher whole blood Se

than ewes from all other treatment groups and only ewes receiving 4 mg/kg Se had whole

blood Se similar to controls. At wk 36, ewes receiving 12, 16, and 20 mg/kg Se had

similar (P > 0.05) whole blood Se and again, only ewes receiving 4 mg/kg Se had whole

blood Se similar to controls. At wk 48, whole blood Se concentrations from ewes

receiving 16 and 20 mg/kg Se were similar (P > 0.10) and higher than (P < 0.05) from

ewes on all other treatments. Ewes receiving 8 and 12 mg/kg Se had similar (P = 0.88)

whole blood Se concentrations which were higher (P < 0.05) than those from controls

and ewes receiving 4 mg/kg Se. Only whole blood Se from ewes receiving 4 mg/kg Se

was similar (P = 0.16) to controls at wk 48. Whole blood Se concentrations at wk 60

followed a pattern similar to wk 48, in terms of differences among treatments. At wk 72,

whole blood Se from four of six dietary levels had numerically decreased from wk 60.

Whole blood from ewes receiving 20 mg/kg Se was higher (P < 0.05) in Se concentration

than in ewes from all other treatments. Ewes receiving 4, 8, 12, and 16 mg/kg dietary Se

produced similar (P > 0.10) whole blood Se and only ewes receiving 12 mg/kg Se had

higher (P < 0.05) whole blood Se than did controls. Cristaldi et al. (2004) also reported a

linear increase in whole blood Se as dietary Se was increased. Likewise, those authors

noted differences in treatment means over controls as dietary Se levels were increased up

to 10 mg/kg. Increased whole blood Se concentrations were reported in dairy cows as

their salt-based mineral mixtures were increased from 20 mg/kg to 120 mg/kg selenite Se








(Awadeh et al., 1998a). Whole blood Se increased linearly in young swine as dietary Se

was fed up to 20 mg/kg (Goehring et al., 1984b).

Wool

Selenium concentration in new growth wool was measured at wk 12, 24, 36, 48,

60, and 72 (Table 3-5). Dietary Se level, time, and dietary Se level x time affected (P <

0.001) wool Se. Wool Se increased linearly (P < 0.001) as dietary Se increased.

Response of wool Se over time was quadratic (P < 0.001) and time response for each

dietary Se level was evaluated individually. Wool Se from controls and ewes receiving 8,

12, and 16 mg/kg dietary Se responded quadratically (P < 0.03) from wk 12 to wk 72.

Wool Se from ewes receiving 4 mg/kg Se responded cubically (P < 0.05) and wool Se

from ewes receiving 20 mg/kg Se increased linearly (P < 0.01) over time. Increased Se

in hair has been reported in other livestock species. Kim and Mahan (2001) observed a

linear response in the hair of pigs as Se in their diet was increased. Goehring et al.

(1984b) reported a quadratic response in the hair of swine as dietary Se as sodium

selenite was increased up to 20 mg/kg. Likewise, Perry et al. (1976) reported increased

Se in the hair of feedlot steers as dietary selenite Se was increased. Cristaldi et al. (2004)

reported a linear increase in the wool of growing sheep as dietary Se was increased and

also observed differences in wool Se of wethers receiving 6, 8, or 10 mg/kg Se vs

controls. These authors did not report a significant treatment x time interaction.

However, wool Se in the present study was affected by time and the interaction of

treatment x time as wool Se increased and then seemed to reach a plateau around wk 48.

Kim and Mahan (2001) and Cristaldi et al. (2004) used 10 mg/kg Se as the highest

dietary level and reported linear responses in hair and wool. However, with 20 mg/kg as

the highest dietary level, the quadratic responses observed by Goehring et al. (1984b) and








in our study suggest that Se in wool and hair does not continue to increase linearly as

dietary Se is increased above 10 mg/kg. At wk 12, only ewes receiving 20 mg/kg Se had

wool Se higher (P < 0.05) than controls, however wool Se from ewes receiving 12 and 16

mg/kg Se tended (P < 0.15) to be higher than from controls. At wk 24, wool Se from

ewes receiving 16 or 20 mg/kg Se was higher than from controls and ewes receiving 4

mg/kg Se. Wool Se from ewes receiving 8 or 12 mg/kg Se tended (P < 0.07) to be higher

than wool Se from ewes receiving 4 mg/kg Se. At wk 36, wool Se from ewes on all

treatment groups was higher (P < 0.05) than from controls, and Se concentrations in wool

from ewes receiving 16 mg/kg Se were higher (P < 0.05) than wool Se from ewes

receiving 4 or 12 mg/kg dietary Se. Wool Se concentrations from ewes on all treatment

groups were similar (P > 0.15) and higher (P < 0.05) than wool Se from controls at wk

48. At wk 60, again, wool Se concentrations from ewes on all treatment groups were

higher (P < 0.05) than wool Se from controls. At the termination of the experiment, wool

Se from ewes receiving 20 mg/kg Se was higher than from ewes on all other treatments

and ewes from all treatment groups produced higher (P < 0.05) wool Se than did controls.

Some wool loss was observed in two ewes receiving 20 mg/kg dietary Se during lactation

in yr one. However, after lambs were weaned and lactation had ceased, both ewes regrew

a full fleece.

Tissues

Selenium concentrations in all tissues were affected (P < 0.001) by dietary Se

level. Selenium concentrations in brain ranged from 1.90 to 6.45 mg/kg DM and

increased linearly (P < 0.05) as dietary Se increased (Figure 3-1). Regressing brain Se

(mg/kg DM) on dietary Se concentration (mg/kg) produced the following relationship:

Brain Se = 1.89 + 1.56 Dietary Se (r2 = 0.52; P < 0.05).








Ewes consuming 12 or 20 mg/kg Se had higher (P < 0.05) brain Se than controls and

ewes consuming 20 mg/kg Se had higher (P < 0.05) brain Se than ewes consuming Se at

all levels except 12 mg/kg.

Diaphragm Se ranged from 1.27 to 4.01 mg/kg DM increased (P < 0.05) in a

linear manner as dietary Se was increased. (Figure 3-1). Regressing diaphragm Se

(mg/kg DM) on dietary Se concentration (mg/kg) produced the following relationship:

Diaphragm Se = 1.27 + 1.33 Dietary Se (r2 = 0.63; P < 0.05).

Ewes receiving 20 mg/kg Se had higher (P < 0.05) diaphragm Se than ewes receiving all

other treatments and only ewes receiving 12 or 20 mg/kg Se had higher diaphragm Se

than controls (P < 0.005).

Heart tissue Se (Figure 3-1) ranged from 1.83 to 6.24 mg/kg DM and increased in

a linear fashion (P < 0.001). Regressing heart Se (mg/kg DM) on dietary Se

concentration (mg/kg) produced the following relationship:
2
Heart Se = 1.83 + 1.99 Dietary Se (r = 0.70; P < 0.05).

Ewes receiving 12, 16, and 20 mg/kg Se had higher (P < 0.05) heart Se than controls and

ewes receiving 4 and 8 mg/kg Se tended to have higher (P < 0.12) heart Se than controls.

Heart Se concentrations from ewes receiving 20 mg/kg Se were higher (P < 0.05) than

those from ewes receiving all other dietary Se levels.

Selenium concentration in hoof ranged from 0.93 to 7.68 mg/kg DM and

increased cubically as dietary Se increased (Figure 3-2). Regressing hoof Se (mg/kg

DM) on dietary Se concentration (mg/kg) produced the following relationship:

Hoof Se = 0.93 + 1.95 Dietary Se 0.49 Dietary Se2 + 0.06 Dietary Se3 (r2 = 0.60;

P < 0.05).








Ewes receiving 16 and 20 mg/kg Se had higher hoof Se (P < 0.05) than controls.

Likewise, ewes receiving 20 mg/kg Se had higher hoof Se (P < 0.05) than ewes receiving

4, 8, and 12 mg/kg dietary Se.

Selenium concentrations in psoas major muscle (i.e. tenderloin), a muscle

commonly consumed by humans, ranged from 0.60 to 3.66 mg/kg DM and increased

linearly as dietary Se increased (Figure 3-2). Regressing psoas major muscle Se (mg/kg

DM) on dietary Se concentration (mg/kg) produced the following relationship:

Psoas major muscle Se 0.59 + 1.42 Dietary Se (r2 = 0.62; P < 0.05).

Selenium concentrations in psoas major muscle from controls were lower (P < 0.05) than

from ewes receiving 4, 12, 16, and 20 mg/kg Se and tended to be lower (P = 0.06) than

psoas major muscle Se concentrations from ewes receiving 8 mg/kg Se. Ewes receiving

20 mg/kg Se had higher (P < 0.05) psoas major muscle Se than ewes receiving all other

Se levels.

Kidney Se ranged from 5.18 to 31.61 mg/kg DM and responded to increased

dietary Se in a cubic fashion (Figure 3-2). Regressing kidney Se (mg/kg DM) on dietary

Se concentration (mg/kg) produced the following relationship:

Kidney Se = 5.18 + 6.64 Dietary Se- 2.28 Dietary Se2 + 0.32 Dietary Se3 (r2 =

0.62; P < 0.05).

Ewes receiving 20 mg/kg Se had higher (P < 0.01) kidney Se than ewes from all other

treatment groups. Ewes receiving 12mg/kg Se tended (P = 0.09) to have higher (P <

0.01) kidney Se than controls.








Liver Se concentration ranged from 4.20 to 230.36 mg/kg DM and responded

quadratically as dietary Se level increased (Figure 3-3.) Regressing liver Se (mg/kg DM)

on dietary Se concentration (mg/kg) produced the following relationship:

Liver Se = 4.19 + 26.59 Dietary Se 9.31 Dietary Se2 (r2 = 0.66; P < 0.01).

Ewes receiving 20 mg/kg dietary Se had higher (P < 0.05) liver Se than ewes from all

other treatments. No other differences (P > 0.05) existed among controls and Se

treatment groups.

Linear increases in the Se concentration of loin, liver, kidney and hoof were

reported in swine (Kim and Mahan, 2001) and sheep (Cristaldi et al., in press). Likewise,

Echevarria et al. (1988) reported linear responses of sheep liver, kidney, heart, and

muscle to dietary Se as sodium selenite Se. In our study, loin, diaphragm, heart, and

brain responded linearly, where kidney and hoof responded cubically and liver responded

quadratically. These higher degree polynomials may be due to changes in metabolism of

Se as dietary Se concentration approaches 20 mg/kg. Most previous research used 10

mg/kg Se as the highest dietary concentration.

Enzymes and Histopathology

Serum for evaluation of albumin and enzyme activities was collected at wk 72

along with samples of brain, diaphragm, heart, hoof tip, kidney, psoas major muscle, and

liver for histopathological evaluation. Concentrations of albumin and activities of Alk

phos, ALT, GGT, AST, and CK in serum were in or below the normal range for adult

sheep (Table 3-6). In instances of Se toxicosis, the activities of these enzymes would

have been increased due to tissue necrosis. Our observations agree with those reported by

Cristaldi et al. (2004) as albumin and enzyme activities in wether sheep after receiving up

to 10 mg/kg Se were in the normal ranges.








Most of the tissues collected at slaughter were free from pathological changes.

The findings of lymphocytes in the portal triads were deemed to be a background finding

and insignificant. Likewise, the findings of lymphocytic foci in the heart tissue were

determined to be associated with sarcocystic parasites. Mineral precipitations were

observed in kidney tissue of some ewes and are incidental, background findings.

Contraction bands present in the diaphragm and psoas major muscle were a result of

stunning during humane slaughter. Adipose tissue was present in the heart and psoas

major muscle, which is an indication of adequate nutrition. Hepatic lipidosis was

diagnosed in four ewes. Two cases (one severe, one moderate) were diagnosed in ewes

receiving 16 mg/kg dietary Se. In the moderate case, there was also evidence of bile

retention. Neither of these ewes lambed in either yr. This would indicate that the hepatic

lipidosis could be treatment related rather than due to metabolic changes associated with

gestation, parturition, and lactation. One ewe receiving 12 mg/kg Se and one ewe

receiving 4 mg/kg were diagnosed with mild hepatic lipidosis, however, both ewes

lambed in both yr. Thus, the hepatic lipidosis was likely due to metabolic changes

associated with lamb production. No evidence of significant pathological changes was

observed in ewes receiving 20 mg/kg dietary Se, which was the highest Se level used in

this study. Cristaldi et al. (2004) found no abnormalities after microscopic evaluation of

heart, liver, kidney, diaphragm, and muscle from wethers consuming up 10 mg/kg Se for

one yr. Likewise, only one instance of abnormal pathology was observed in ewes

consuming less than 10 mg/kg Se on our study. Furthermore, our study was

approximately 40% longer in duration, utilized treatments of up to 100% more Se, and

introduced stresses of production, all of which should have helped to induce Se toxicosis








and thus, the finding of abnormal organ pathology. However, abnormal pathological

findings were few and did not follow a pattern with respect to dietary level which would

be indicative of Se toxicosis.

No clinical signs of Se toxicosis such as abnormal hoof growth or loss of wool

were observed in ewes receiving > 16 mg/kg Se. However, some excessive hoof growth

was observed after approximately one yr in ewes receiving 16 and 20 mg/kg Se and wool

loss was observed during lactation in two ewes receiving 20 mg/kg Se. Livestock

suffering from alkali disease were reported to have hair Se concentrations of up to 45

mg/kg and whole blood Se of 4.1 mg/L, while hooves, liver, and kidney of affected

animals contained 10 mg/kg Se or more (NAS, 1983). At no time during our study did

wool Se reach even 10 mg/kg and whole blood Se remained less than 50% of the

aforementioned 4.1 mg/L concentration. Also, hoof Se remained under 8 mg/kg for all

treatments during the course of our study. Liver and kidney Se concentrations from our

study were higher than the 10 mg/kg previously reported. The elevated concentrations of

Se in the liver and kidney of ewes consuming 16 and 20 mg/kg, and the observation of

some clinical signs of Se toxicosis and limited pathological abnormalities in ewes

consuming these Se levels may indicate that some ewes were beginning to suffer from Se

toxicosis. However, definitive evidence was not observed. Therefore, it is necessary that

either dietary Se concentration or duration of experiment be increased in order to induce a

definitive Se toxicosis using inorganic Se.

Implications

The maximum tolerable level of selenium as sodium selenite for ruminants is

higher than 2 mg/kg. Feeding up to 12 ppm selenite selenium to ewes under stresses of

production (i.e., gestation and lactation) for 72 wk did not produce any clinical or








pathologic signs of selenium toxicosis. Ewes fed 16 and 20 mg/kg produced some signs

of selenium toxicosis; however, general metabolic disorders could not be ruled out and no

deaths of ewes consuming these levels of selenium were attributed to selenium toxicosis.

Further studies of this nature should further prove that the current suggested tolerable

level of Se is underestimated.

Summary

The objectives of this 72-wk study were to evaluate and compare the effects of six

dietary levels of inorganic Se on serum, whole blood, wool, and tissue Se concentrations

of mature ewes during lamb production and determine maximum tolerable level of Se

during lamb production. Forty-one range-type ewes were used in a completely

randomized design with six dietary treatments. Sodium selenite was added to a corn-

soybean meal basal diet to provide 0.2 (control), 4, 8, 12, 16, and 20 mg/kg dietary Se to

ewes during lamb production. Serum Se and ewe BW were measured at 4-wk intervals,

whole blood Se, and wool Se were measured every 12 wk, and samples of brain,

diaphragm, heart, hoof, kidney, liver, and psoas major muscle were collected at the

termination of the experiment. Dietary Se did not affect ewe BW during the study (P =

0.69). Serum Se increased linearly as dietary Se level increased (P < 0.001) and

responded cubically (P = 0.02) over time. Selenium in whole blood increased linearly (P

< 0.001) with increased dietary Se and cubically (P < 0.01) over time. Wool Se increased

linearly (P < 0.001) as dietary Se increased and response over time was quadratic (P <

0.001). Brain, diaphragm, heart, and psoas major muscle Se increased linearly as Se in

the diet increased, liver Se responded quadratically, and hoof and kidney Se responded

cubically to treatment (P < 0.05). In general, serum, whole blood, and tissue Se








concentrations from ewes receiving 12, 16, or 20 mg/kg dietary Se were higher (P < 0.05)

than from controls and ewes receiving less dietary Se. Though serum, whole blood, and

wool Se concentrations were elevated in ewes receiving increased dietary Se, at no time

did serum, whole blood, or wool Se concentrations reach levels previously reported as

toxic and a pattern of clinical signs of Se toxicosis was not observed. Microscopic

evaluation of liver, kidney, diaphragm, heart, and psoas major muscle did not reveal

evidence of Se toxicosis in ewes on any dietary Se level. Ewes under our experimental

conditions and during the stresses of production were able to tolerate up to 20 mg/kg

dietary Se as sodium selenite for 72 wk. These findings suggest that the maximum

tolerable level of inorganic Se for sheep to be much higher than 2 mg/kg as was

suggested previously. Experiments which are longer in duration and utilize higher dietary

Se concentrations may be used to clearly define the maximum tolerable level.








Table 3-1. Diet composition (as-fed) for selenite-Se supplemented ewes'
Ingredient % as-fed
Ground yellow corn 53.75
Cottonseed hulls 22.00
Soybean meal (47.5% CP) 16.00
Alfalfa meal (14% CP) 3.00
Soybean oil 3.00
Trace mineral mixb 1.00
Ground limestone 1.25
Vitamins A & D C
aSelenium levels in diet (as analyzed): 0.29, 3.77, 7.54, 11.01, 15.48, and 19.05 ppm
for Se levels 0.2, 4, 8, 12, 16, and 20 ppm, respectively.
bTrace mineral mixture supplied between 96.5% and 98.5% NaCl, and provided per kg
of diet: 1.0 mg Co (as carbonate), 5.0 mg Cu (as oxide), 0.7 mg I (as iodate), 35 mg Fe
(as oxide), 25 mg Mn (as oxide), and 35 mg Zn (as oxide).
cProvided per kg of diet: 5,000 IU of Vitamin A and 500 IU of Vitamin D3.


Table 3-2. Lamb production of ewes receiving different concentrations of dietary Se
Year 1a Year 2b
Dietary Se, ppm Ewes lambed Lambs born Ewes lambed Lambs bornd
0.2 5 9 4 5
4 7 11 7 10
8 5 6 5 7
12 5 8 6 11
16 4 5 0 0
20 7 14 2 3
Total 33 53 24 36
aEwes began receiving experimental diets at 57 d average gestation in yr 1.
bEwes were fed experimental diets continuously during breeding and gestation yr 2.
CLamb crop as lambs born (53) per ewe exposed (41) was 129% in yr 1.
dLamb crop as lambs born (36) per ewe exposed (33) was 109% in yr 2.









Table 3-3. Effect of dietary inorganic Se level on serum Se concentration of mature
ewes at various stages of lamb production a
Dietary Se, mg/kg
0.2 4 8 12 16 20
Stage of Production Serum Se, gg/L
Late Gestation, yr 1 149g 67 242gh 67 354hi 79 414 79 463 79 707 + 81
Lactation, yr 1V 151g+56 2729 54 486h63 623hi 63 718-4 j63 811- i66
Dry, rebreeding d 1629 110 298gh 95 458hi- 100 604 99 120 Y 106 1084 113
Late Gestation, yr2' 1409+ 137 313gh 124 4469b+ 142 596hi 142 986i 158 1072i + 149
Lactation, yr 2f 127 114 3259h+ 103 536hi 114 718i 114 769 136 1355 121
aData represent least squares means SE.
bLate gestation, yr 1, defined as 56 d prepartum and includes serum Se concentrations for wk 4, 8, and 12.
cLactation, yr 1, defined as 84 d postpartum and includes serum Se concentrations for wk 12, 16, 20, and 24.
dDry, rebreeding period, 168 d, includes serum Se concentrations for wk 28, 32, 36, 40, 44, and 48.
eLate gestation, yr 2, defined as 56 d prepartum and includes serum Se concentrations for wk 52, 56, and 60.
rLactation, yr 2, defined as 84 d postpartum and includes serum Se concentrations for wk 60, 64, 68, and 72.
gahijMeans within rows lacking a common superscript differ (P < 0.05).


Table 3-4. Effect of dietary inorganic Se level on whole blood Se concentration of
mature ewesa


Dietary Se, mg/kg
Week of 0.2 4 8 12
experiment Whole blood Se, gg/L
12 386h 168 839i 151 902ij 151 1241jk + 139
24C 420h 140 60lhi- 130 852ij 130 1047jk 130
36 d 438h 152 635h 149 1079i 138 1314 138
48e 378h- 154 661h -130 1117i 139 1145 130
60f 497h 154 802hi 138 1070 -151 1131i 150
729 410h 154 618hi 139 721hi 140 916 -140
aData represent least squares means SE.
n 4, 5, 5, 6, 7, and 5 for Se levels 0.2, 4, 8, 12, 16, and 20 mg/kg, respectively.
cn = 6, 7, 7, 7, 7, and 5 for Se levels 0.2, 4, 8, 12, 16, and 20 mg/kg, respectively.
dn = 5, 5, 6, 6, 6, and 5 for Se levels 0.2, 4, 8, 12, 16, and 20 mg/kg, respectively.
'n = 5, 7, 6, 7, 5, and 5 for Se levels 0.2, 4, 8, 12, 16, and 20 mg/kg, respectively.
fn = 5, 6, 5, 5, 4, and 5 for Se levels 0.2, 4, 8, 12, 16, and 20 mg/kg, respectively.
gn = 5, 6, 6, 6, 3, and 5 for Se levels 0.2, 4, 8, 12, 16, and 20 mg/kg, respectively.
hIjik lMeans within rows lacking a common superscript differ (P < 0.05).


16 20

1053'j 130 1558k 154
1312k _4 130 1822' 154
1668j 139 1373ij 154
1616 151 1951j 154
1892j 167 1796j 154
841hi 194 1855 - 154









Table 3-5. Effect of dietary inorganic Se level on wool Se concentration of mature
ewesa
Dietary Se, mg/kg
Week of 0.2 4 8 12 16 20
experment Wool Se, mg/kg (DM basis)
12 0.50' 0.58 0.71h 0.54 1.36hi 0.54 1.67h 0.54 2.00h + 0.54 2.23 0.64
24c 0.62' 0.62 143hi 0.54 2.76j 0.54 2.79j 0.54 3.58jk 0.54 4.64k 0.64
36' 1.58h 0.63 3.72' 0.54 4.86j 0.54 3.96' 0.54 5.57i 0.57 5.27'j 0.64
48' 1.18h 0.64 4.86' 0.54 4.64i 0.54 4.821 0.54 5.47 0.57 5.53' 0.64
60' 1.25' 0.64 4.06i 0.54 6.09i 0.57 5.50'j 0.57 5.63'j 0.68 5.17'j 0.64
729 0.96h 0.64 3.42i 0.57 3.67'j 0.57 5.12' 0.58 5.1 li 0.80 7.69k 0.64
aData represent least squares means SE.
bn = 6, 7, 7, 7, 7, and 5 for Se levels 0.2, 4, 8, 12, 16, and 20 mg/kg, respectively.
Cn = 5, 7, 7, 7, 7, and 5 for Se levels 0.2, 4, 8, 12, 16, and 20 mg/kg, respectively.
dn = 5, 7, 7, 7, 6, and 5 for Se levels 0.2, 4, 8, 12, 16, and 20 mg/kg, respectively.
en = 5, 7, 6, 7, 6, and 5 for Se levels 0.2, 4, 8, 12, 16, and 20 mg/kg, respectively.
fn = 5, 7, 6, 6, 4, and 5 for Se levels 0.2, 4, 8, 12, 16, and 20 mg/kg, respectively.
gn = 5, 6, 6, 6, 3, and 5 for Se levels 0.2, 4, 8, 12, 16, and 20 mg/kg, respectively.
h'ij'kMeans within rows lacking a common superscript differ (P < 0.05).


Table 3-6. Amount of albumin and tissue enzyme activities present in serum of Se
supplemented ewesa"bc
Dietary Se, mg/kg
Item Normal concentration 0.2 4 8 12 16 20
Albumin 2.4-4.0 g/dL 2.9 2.5 1.940 2.2 2.3 2.6
Alk Phos 68-387 IU/L 91.0 60.2 65.3 95.5 36.3 89.0
AST 60-280 IU/L 53.8 131.7 32.8 52.3 64.0 24.6
ALT 11-40 IU/L 15.4 34.7 2.7 13.3 17.7 24.0
GGT 15-60 IU/L 67.6 63.2 39.8 55.0 63.7 59.6
CK 0-584 IU/L 67.4 108.2 36.3 58.7 47.7 55.0
aSerum sample collected at wk 72.
bAlbumin and tissue enzyme activities presented in same units as normal concentration ranges.
CGGT and CK ranges were established by University of Florida Veterinary Teaching Hospital.








-- Brain
7
Diaphragm
6 -0- Heart
e5
4

g2

0


0 4 8 12 16 20

Dietary Se, mg/kg

Figure 3-1. Effect of dietary inorganic Se level on Se concentrations in brain, diaphragm,
and heart of ewes; SE = 0.6 to 0.9, 0.3 to 0.4, and 0.4 to 0.6 for brain, diaphragm, and
heart, respectively.








35 -
-c- Kidney

30 -Hoof
!- Loin
,. 25

20



0
15





0
5



0 4 8 12 16 20
Dietary Se, mg/kg

Figure 3-2. Effect of dietary inorganic Se level on Se concentrations in kidney, hoof, and
loin (psoas major muscle) of ewes; SE = 3.0 to 3.3, 0.8 to 1.1, and 0.3 to 0.5 for kidney,
hoof, and loin, respectively.








250 -
225 -Liver
a 200
175
S150
125
r 100
75
S50
25
0 I
0 4 8 12 16 20
Dietary Se, mg/kg
Figure 3-3. Effect of dietary inorganic Se level on liver Se concentration in ewes; SE =
27.5, 25.4, 24.5, 24.5, 34.6, and 26.9 for 0.2, 4, 8, 12, 16, and 20 mg/kg dietary Se,
respectively.













CHAPTER 4
EFFECTS OF SELENIUM LEVELS IN EWE DIETS ON SELENIUM IN MILK
AND PLASMA AND TISSUE SELENIUM CONCENTRATIONS OF LAMBS

Introduction

Selenium has long been implicated as a toxic element to livestock (Oldfield,

2002). Animals grazing seleniferous plants in certain regions of the world are subject to

Se toxicosis and conditions such as alkali disease and "blind staggers" (McDowell,

2003). The estimated maximum tolerable level of Se for ruminant livestock is 2 mg/kg

(NRC, 1980). However, recent research (Cristaldi et al., in press) has shown that sheep

may consume up to 10 mg/kg Se as sodium selenite in the total diet for one yr, without

showing signs of selenium toxicosis. Although it was concluded that these wethers were

not suffering from Se toxicity, they did have increased serum, whole blood, and tissue Se

concentrations. Like blood and tissue, milk Se is affected by dietary Se level (Conrad and

Moxon, 1979; Givens et al., 2004) and Se readily crosses the placenta to the fetus (Van

Saun et al., 1989). Furthermore, positive correlations exist between blood Se of cows and

blood Se of their calves (Kincaid and Hodgson, 1989; Enjalbert et al., 1999; Pehrson et

al., 1999). In sheep, Cuesta et al. (1995) showed increased colostrum Se from ewes

receiving supplemental Se, and that milk Se was higher after one mo of supplementation.

Thus, it seems that neonates from dams consuming high dietary levels of Se would have

increased blood Se at birth, and subsequently would be exposed to high Se intake from

increased Se in milk.








Acute Se toxicosis has been evaluated by injecting ewe lambs with sodium

selenite (Blodgett and Bevill, 1987), and Abdennebi et al. (1998) evaluated the toxic

effects of dosing lambs with extracts of milk vetch (Astragalus lusitanicus). In both

studies, weaned lambs were utilized. Newborn and pre-weaned lambs differ from older

sheep in ruminal and digestive function (Church, 1979; Goursand and Nowak, 1999), and

may respond differently than older animals to increased Se intake. We hypothesized that

when Se in gestating ewe diets is increased, colostrun Se, milk Se and plasma Se of their

lambs will increase. The objective of this experiment was to follow ewes through two

lamb crops and evaluate and compare the effects of six levels of dietary Se on ewes' milk

and the Se status of their lambs prior to weaning.

Materials and Methods

All animal procedures were conducted within the guidelines of and approved by

the University of Florida Institutional Animal Care and Use Committee. This 504-d

experiment utilizing ewes and two lamb crops was conducted from December 18, 2001 to

May 5, 2003 at the University of Florida Sheep Nutrition Unit located in southwestern

Alachua County, FL. Thirty-three, four year old, Rambouillet ewes that originated from

a single flock in Texas and were previously confirmed pregnant (average 57 d gestation)

were weighed (57.4 + 5.7 kg) and received 2-ml ivermectin (Ivomec; Merial Ltd. Iselin,

NJ). Ewes were randomly assigned to one of six dietary treatments for a 504 d (two

lambing seasons) study. Six dietary treatments were 0.2, 4, 8, 12, 16, or 20 mg/kg Se as

sodium selenite (as-fed basis) added to a corn-soybean meal basal diet (Table 4-1). The

basal diet was formulated to meet animal requirements for protein, energy as TDN,

vitamins, and minerals for this class of sheep (NRC, 1985). Animal numbers per

treatment were 5, 7, 5, 5, 4, and 7 for 0.2 (control), 4, 8, 12, 16, and 20 mg/kg added Se,








respectively. Ewes were housed by treatment group in covered wooden pens (53.5 in2)

with earth floors and ad libitum water. Diets were fed at 909 g/ewe/d from d 0 until

lambing began (d 81), increased to 1 OOOg/ewe/d during lambing (d 81 to 101), and again

increased to 1135 g/ewe/d during lactation (d 101 to 171). Diets were sampled every 28

d, ground (1 mm), and frozen at 0C until analysis.

In the first year, 52 lambs were born over 20 d from March 9, 2002 to March 28,

2002. Prior to lambing, ewes were fitted with a device to cover the udder and prevent

lambs from nursing until a blood sample could be obtained. The udder cover was crafted

from nylon pantyhose (Eeggs Products, Winston-Salem, NC) and polyester elastic (2.54

cm wide) and held in place with safety pins (Figure 4-1). A blood sample for plasma

analysis was collected from lambs immediately after birth via jugular venipuncture into

10-ml heparinized tubes (Vacutainer; Becton-Dickinson, Franklin Lakes, NJ). The udder

cover was then removed from the ewe and five ml of pre-suckled colostrum was collected

into a 15-ml plastic centrifuge tube (Fisher; Fisher Scientific, Pittsburgh, PA). Additional

blood samples were collected from lambs and milk samples from ewes at 3, 28, and 56 d

postpartum. Blood samples were centrifuged at 700 x g and the plasma then frozen at

0C. Ewe milk samples were also stored frozen at 0C for later analysis. Lambs were

weaned at 70 d of age and ewes then received 909 g/ewe/d of their respective diets until

next lambing. At 70 d, ram lambs were surgically castrated and the testes were frozen at

0C until analysis.

On August 15, 2002, ewes were pen exposed to rams for 35 d. In the second year,

36 lambs were born over 34 d from January 17, 2003 to February 20, 2003. All sampling

intervals, procedures, feeding levels, and materials used were duplicated from the first








year. Plasma, milk, testes, and feed samples from both years were analyzed for Se

concentration using a fluorometric method described by Whetter and Ullrey (1978). To

help ensure reliability of the analytical method, a certified standard (National Bureau of

Standards Bovine Liver SRM-1577a; U.S. Department of Commerce, National Institute

of Standards and Technology, Gaithersburg, MD) was frequently analyzed.

Effects of treatment on lamb testicular and colostrum Se were analyzed using

PROC MIXED in SAS (SAS for Windows 8e; SAS Inst. Inc., Cary, NC) in a completely

randomized design. Contrast statements were used to compare means as described by

Littell et al. (1998; 2000). PROC MIXED of SAS was also used to analyze effects of

treatment, d, and the interaction of treatment x d on milk Se and plasma Se as repeated

measures with a spatial power covariance structure with respect to d and a subplot of

animal nested within treatment. Contrast statements were written to determine

differences in means for different sampling d. PROC CORR was used to determine

correlations of ewe milk Se to lamb plasma Se.

Results

In year one, 11 of 52 lambs were removed from the study before 56 d of age. Five

lambs were born to ewes which produced little or no milk, one ewe had extremely

enlarged or "bottle" teats and her two lambs were unable to suckle, two lambs died of

weakness/dehydration, and two died of "joint ill." There were no apparent signs of

selenium toxicosis in any lambs regardless of dietary Se level of their dams.

In year one, colostrum Se was affected by Se concentration of the ewes' diet (P =

0.008) and increased linearly (P < 0.001) as dietary Se increased (Table 4-2). Ewes

receiving 16 or 20 mg/kg dietary Se produced higher (P < 0.05) colostrum Se than did

controls. Ewes receiving 20 mg/kg dietary Se also produced higher (P < 0.05) colostrum








Se than did those ewes receiving 4 mg/kg dietary Se and tended (P < 0.12) to produce

higher colostrum Se than ewes receiving 8 and 12 mg/kg dietary Se. Likewise, colostrum

Se from ewes receiving 16 mg/kg tended to be higher (P = 0.052) than colostrum Se from

ewes receiving 4 mg/kg dietary Se. Colostrum Se from ewes receiving 8, 12, or 16

mg/kg dietary Se was similar (P > 0.20).

In year two, 12 of 36 lambs were removed from the study before d 56. Seven

lambs were removed due to their dams having either no milk or enlarged teats that were

unable to be suckled, four lambs were lost to predation, and one lamb was removed due

to physical injury. No lambs were lost or removed from the study due to dietary Se in the

diet of their dam. As in year one, colostrum Se was affected by dietary Se (P < 0.05) and

increased linearly (P < 0.01) as dietary Se increased (Table 4-2). Ewes receiving 8, 12,

or 20 mg/kg dietary Se produced colostrum with similar (P > 0.19) Se concentrations,

which were higher (P < 0.05) than colostrum Se from controls. Ewes consuming 8

mg/kg dietary Se produced colostrum Se higher (P < 0.05) than those ewes consuming 4

mg/kg dietary Se. Likewise, ewes consuming 20 mg/kg dietary Se tended to produce

colostrum Se higher (P = 0.10) than ewes consuming 4 mg/kg dietary Se. No ewes

receiving 16 mg/kg dietary Se lambed in year two and are not represented in these

comparisons.

Ewe milk Se collected at 3, 28, and 56 d postpartum increased linearly (P <

0.001) as dietary Se increased in year one (Table 4-3). Day of sampling also had an

effect (P = 0.002), but there was no treatment x d interaction. At d 3 postpartum, ewes

receiving 4 and 8 mg/kg dietary Se produced similar (P = 0.60) milk Se and milk Se from

ewes receiving 8 mg/kg dietary Se tended (P = 0.06) to be higher than from controls.








Ewes consuming 12, 16, and, 20 mg/kg Se produced milk Se higher (P < 0.05) than

controls. Milk Se from ewes consuming 16 mg/kg Se tended to be higher (P = 0.09) than

that from ewes consuming 20 mg/kg Se and was higher (P < 0.05) than milk Se from all

other treatments. At d 28 postpartum, ewes consuming 20 mg/kg dietary Se produced

milk Se concentrations higher (P < 0.05) than did controls or ewes consuming 4 mg/kg

dietary Se. Likewise, milk Se from ewes consuming 20 mg/kg dietary Se tended to be

higher (P < 0.075) than milk Se from ewes receiving 8 or 12 mg/kg Se. Milk Se from all

other treatment groups was similar (P > 0.20). At the final milk collection in year one (d

56), milk Se concentrations from controls and ewes consuming 4, 8, and, 12 mg/kg Se

were similar (P > 0.17). Ewes consuming 16 and 20 mg/kg Se produced similar milk Se

(P = 0.43), which was higher (P < 0.05) than milk Se from all other treatments. A linear

increase (P < 0.01) in milk Se as dietary Se increased was observed at each sampling d as

well as over all sampling d. Milk Se concentrations in year one remained below 1000

gg/L from d 3 to d 56.

In year two, dietary Se concentration had an effect on milk Se (P < 0.05), as did

the interaction of dietary Se concentration x sampling d (P <0.05). Milk Se

concentrations from ewes consuming 8, 12, and 20 mg/kg Se were similar (P > 0.38) to

each other and higher (P < 0.05) than controls at d 3 (Table 4-4). Ewes consuming 12

mg/kg Se had higher (P < 0.05) milk Se than did those consuming 4 mg/kg Se. Ewes

consuming 20 mg/kg Se produced milk Se that tended (P = 0.09) to be higher than milk

Se from ewes consuming 4 mg/kg Se. There were no differences (P > 0.05) in milk Se

among treatment groups at d 28. However, ewes consuming either 12 or 20 mg/kg Se

had milk Se which tended to be higher (P < 0.12) than control. At d 56, milk Se








concentrations from ewes consuming 20 mg/kg Se were higher than from all other

treatments (P < 0.05). Ewes consuming 12 mg/kg Se had milk Se which was higher (P <

0.05) than controls and tended to be higher than from ewes receiving 4 mg/kg Se (P =

0.07). Milk Se concentrations, at d 56, from all other treatment groups were similar (P >

0.18). Milk Se concentrations increased linearly (P < 0.001) as dietary Se increased over

all sampling d.

Lamb plasma Se was affected by dietary Se concentration of their dams (P <

0.001) and increased linearly as dietary Se of dams increased (P < 0.001) in year one

(Table 4-5). Likewise, d of sampling affected lamb plasma Se concentration (P < 0.01).

On d 3 to 56, lamb plasma Se was positively correlated to ewe milk Se (r = 0.29; P <

0.00 1). At birth, lambs suckling ewes consuming 20 mg/kg Se had higher plasma Se than

controls (P < 0.05) and lambs suckling ewes consuming 16 mg/kg Se tended to have

higher plasma Se than controls (P = 0.15). All other lambs had similar plasma Se (P >

0.25). At 3 d of age, lambs from ewes consuming 20 mg/kg Se had higher plasma Se (P

< 0.01) than all other treatment groups. Plasma Se concentrations from control lambs

were lower (P < 0.05) than plasma Se from lambs suckling ewes consuming 8, 12, or 16

mg/kg Se. At 28 d of age, lambs suckling ewes receiving 12, 16, or 20 mg/kg Se had

higher plasma Se than did controls (P < 0.01). Likewise, lambs suckling ewes receiving

4 or 8 mg/kg Se tended to have higher plasma Se than controls (P < 0.14). Plasma Se

from lambs suckling ewes receiving 20 mg/kg Se was higher than from lambs suckling

dams that received 4, 8, or 12 mg/kg Se (P < 0.05) and tended to be higher than from

lambs suckling dams that received 16 mg/kg Se (P 0.09). At 56 d, plasma Se from

lambs suckling ewes receiving 4, 12, 16, or 20 mg/kg Se was higher than controls (P <








0.05) and lambs suckling ewes that received 8 mg/kg Se tended to have higher plasma Se

than controls (P = 0.067). Plasma Se from lambs suckling ewes receiving 16 or 20

mg/kg Se was higher than plasma Se from lambs suckling ewes receiving 4 mg/kg Se (P

< 0.05) and tended to be higher than plasma Se from lambs suckling ewes receiving 8

mg/kg Se (P: <0.08).

Lamb plasma Se, in year two (Table 5-6), was affected by the concentration of Se

in the diet of their dams (P < 0.001) and increased linearly as Se concentration in dams'

diet increased (P < 0.001). Day of sampling and the interaction of dietary Se

concentration x d of sampling also affected lamb plasma Se (P < 0.01). At birth, lambs

from ewes receiving 12 mg/kg Se had higher plasma Se than did controls (P < 0.05).

Likewise, lambs from ewes receiving 20 mg/kg Se had higher (P < 0.05) plasma Se than

did controls and lambs from dams receiving 4 or 8 mg/kg Se. From 3 to 56 d of age,

lamb plasma Se was positively correlated to ewe milk Se (r = 0.44; P < 0.001). At 3 d of

age, lambs from all treatment groups had higher plasma Se than did controls (P < 0.05)

and lambs suckling ewes receiving 8, 12, or 20 mg/kg Se had higher plasma Se than did

those suckling ewes receiving 4 mg/kg Se (P < 0.05). At d 28, lamb plasma Se from all

treatment groups was higher than controls (P < 0.05). Lambs suckling ewes receiving 20

mg/kg Se had plasma Se higher than all other treatment groups (P < 0.05). Also, lambs

suckling ewes receiving 12 mg/kg Se had plasma Se higher than lambs suckling ewes

receiving 4 or 8 mg/kg Se (P < 0.05). At d 56, lambs suckling dams receiving 4, 8, or 12

mg/kg Se had higher plasma Se than did controls (P < 0.05) and lambs suckling dams

receiving 20 mg/kg Se tended to have higher plasma Se than did controls (P = 0.067).








Plasma Se from lambs suckling ewes receiving 12 mg/kg Se was higher than (P < 0.05)

than plasma Se from lambs suckling ewes receiving 4 and 8 mg/kg.

Selenium concentration in testis (dry basis) taken from ram lambs at 70 d of age

(weaning) increased linearly (P < 0.00 1) as dams' dietary Se concentration increased

(Figure 4-2) in year one. Testicular Se from lambs suckling ewes receiving 20 mg/kg Se

was higher than testis Se from controls and lambs suckling ewes receiving 4 or 8 mg/kg

Se (P < 0.05). Lambs suckling ewes receiving 16 mg/kg Se had testicular Se which was

higher (P < 0.05) than testicular Se from controls and lambs suckling ewes receiving 4

mg/kg Se. Lambs suckling ewes receiving 12 mg/kg tended to have higher testicular Se

than did controls or lambs suckling ewes receiving 4 mg/kg Se (P < 0.11). Likewise,

lambs suckling ewes receiving 8 mg/kg tended to have higher testicular Se than did

controls (P = 0.14). There was no effect of treatment on testicular Se in year two (P -

0.70). Testicular Se concentrations were 2.05, 3.16, 2.96, and 3.24 mg/kg for controls

and lambs suckling ewes receiving 4, 8, and 12 mg/kg Se, respectively.

Discussion

Colostrum Se increased with dietary Se level in both years. Cuesta et al. (1995)

reported higher colostrum Se from vitamin E + Se supplemented ewes versus their

unsupplemented counterparts. These findings are further supported by Mahan (2000),

who demonstrated that colostrum Se was increased by increasing Se in prepartum and

postpartum sow diets. Colostrum Se from Se supplemented crossbred ewes was

increased over unsupplemented controls (Norton and McCarthy, 1986), however, those

researchers used injectable vitamin E + Se as the supplemental Se source rather than

dietary Se. Overnes et al. (1985) also reported an effect on colostrum Se from ewes

receiving Se fed via free-choice salt and mineral mixtures. In the present study, ewe








colostrum Se concentrations from controls in year one were lower at 257 Pig/L than

values in cow colostrum from Romania reported by Serdaru et al. (2004). However, in

year two, after our ewes had been receiving their respective diets for approximately 13

mo, colostrum Se from controls had more than doubled to 705 gg/L. The increase in

colostrum Se after a longer duration of Se supplementation is substantiated by Maus et al.

(1980). Those authors reported that Se in cows' milk increased with time when fed at 0.2,

0.3, 0.4, and 0.7 mg/kg in a corn-brewers' grain dairy diet. As dietary Se was increased

by increments of 4 mg/kg from 4 mg/kg up to 20 mg/kg, colostrum Se increased by 45.3,

8.8, 55.2, and 10.1%, respectively in year one. Colostrum Se was numerically higher in

year two when Se was fed at 0.2, 4, 8, and 12 mg/kg. This furthers the idea that

colostrum Se, when Se is supplemented at equivalent concentrations, may be increased as

animals are supplemented for an extended period of time. The use of increased Se in

gestating animals may prove beneficial to their offspring as it provides greater

antioxidative protection through increased colostrum Se and thus provides greater

phagocytic and microbicidal activity (Wuryastuti et al., 1993).

As with colostrum Se, subsequent milk Se also increased as dietary Se increased.

Givens et al. (2004) reported increased milk Se as selenite Se increased from 0.38 to 1.14

mg/kg and that a strong positive correlation exists between milk Se and dietary Se (r =

0.979). Gardner and Hogue (1967) reported up to five-fold increases in milk Se when

sodium selenite was added to ewe diets at 1 mg/kg. In the present study, a five-fold or

greater increase was observed in milk Se as dietary Se was increased from control (0.2

mg/kg) to 8 mg/kg for d 3, from control (0.2 mg/kg) to 20 mg/kg for d 28, and from

control (0.2 mg/kg) to 12 mg/kg for d 56 in year one. In year two, a four-fold or greater








increase in milk Se was observed as dietary Se increased from control to 8 mg/kg on each

sampling d. Few increases of that magnitude were observed between groups receiving Se

at high concentrations. This indicates that the proportion of Se transferred to milk

decreases as dietary Se concentration increases. Waite et al. (1975) suggest that Se is

subject to a bioreducing process as it is transferred from plasma to milk. These authors

report that only 1.5% of dietary selenite Se appeared in milk. Givens et al. (2004)

observed increases of more than four-fold in cows' milk Se when dietary Se was doubled

and tripled using Se yeast. However, in our study, increases of such magnitude were not

observed when dietary Se was doubled and tripled using sodium selenite. These

observations would indicate that, though milk Se can be increased using an inorganic Se

source, a greater proportion of organic Se is transferred to milk. This concept is supported

by several studies using cattle (Knowles et al., 1999; Ortman and Pehrson, 1999; Pehrson

et al., 1999).

One objective of this study was to quantitate the effects on lambs that were

suckling ewes that received dietary Se above the maximum tolerable level of 2 mg/kg

(NRC, 1980). No lambs were born with congenital deformities or abnormalities, nor did

any lamb display signs of selenium toxicosis (e.g., wool loss, hoof malformation,

anorexia) from birth to weaning. Selenium included in ewe diets has previously been

shown to be transmitted to the lamb via the placenta and milk (Jacobsson et al., 1965). In

our study, plasma Se in lambs increased as Se concentration in their dams' diet increased

and was positively correlated to milk Se. Lambs from ewes receiving the control diet had

plasma Se at birth that averaged 81 4g/L in year one and 85[tg/L in year two. These

values are only slightly above the normal range (50-80 gg/L) for neonate lambs (Aitken,








2001) and more than double the plasma Se concentration suggested by Bostedt and

Schramel (1990) for normal growth and health in newborn calves. Lamb plasma,

collected before nursing, increased in Se as Se concentration in the diet fed to ewes

during gestation increased in both years. Ewes receiving 12 mg/kg dietary Se gave birth

to lambs with up three-fold higher plasma Se than did controls. Likewise, ewes receiving

20 mg/kg dietary Se, which is ten fold higher than the established maximum tolerable

level for Se, gave birth to lambs with only approximately four-fold higher plasma Se than

did controls. These results indicate that Se does cross the placenta to the fetus. Koller et

al. (1984) demonstrated maternal transfer of Se in beef cattle and Kim and Mahan (2001)

reported elevated serum and tissue Se in neonate pigs when dietary Se levels of sows

were increased. This does not concur with (Wright and Bell, 1964) who reported no

increase in lamb plasma Se when their dams were fed increased Se and demonstrated a

defined placental barrier for Se.

Plasma Se remained elevated in lambs which were suckling ewes receiving

increased dietary Se and from d 3 to d 56 ranged from 196 to 648 gg/L in year one and

244 to 775 jug/L in year two. These plasma Se concentrations were much higher than the

> 70 gg/L suggested as adequate by Zachara et al. (1993). However, at no time did any

lamb have plasma Se near or above 1400 jug/L which has been suggested as the plasma

level when signs of Se toxicosis appear in sheep (Glenn et al., 1964c) and swine (Kim

and Mahan, 2001). Marrow (1968) reported that death occurred within 16 hours in 35%

of nursing lambs which were dosed with 10 mg of sodium selenite orally in an attempt to

prevent nutritional muscular dystrophy. Smyth et al. (1990) observed death as soon as

six hours after an oral dose of 5 mg Se/kg BW. Contrarily, Lagace et al.(1964) dosed








lambs from two to 14 wk of age with 5 mg of sodium selenite via subcutaneous injection

every two wk and did not induce Se toxicity. Lambs on our study did not receive nearly

the amount of Se that others reported to be deadly, even from nursing dams supplemented

with Se up to 20 mg/kg during gestation and lactation. However, our lambs were

subjected to elevated milk Se concentrations. Based on data from Mellor and Murray

(1986) and Wohlt et al. (1984) milk intake in lambs from birth to 56 d ranges from 866-

1246 g/d. Given those intake estimates and the colostrum and milk Se concentrations

from the present study, lambs consuming the colostrum or milk with the highest Se

concentration at the highest intake would ingest 4.39 mg of Se/d. In newborn lambs (3

kg BW), that amount of Se would translate to 1.46 mg Se/kg BW and to 0.29 mg Se/kg

BW in 8 wk old lambs (15 kg BW). These levels are considerably less than levels

previously reported to cause death in young lambs.

Testes taken from ram lambs at 70 d were evaluated for Se concentration.

Selenium is implicated in sperm quality and reproductive function of livestock

(Hidiroglou, 1982; Marin-Guzman et al., 2000) and concentrations in testes are less than

in liver and generally greater than in heart, spleen, and pancreas. As with plasma Se in

the suckling lambs, testicular Se of lambs increased as Se increased in the ewe diets and

ranged from 1.67 mg/kg in controls to 4.25 mg/kg in lambs whose dams received 20

mg/kg dietary Se. These Se concentrations lie between those concentrations found in the

liver and heart of wethers consuming up to 10 mg/kg Se as sodium selenite for one year

(Cristaldi et al., in press). Those wethers were reported to also have elevated

concentrations of Se in serum, whole blood, wool, and other organs. However, they

displayed no clinical signs of selenium toxicosis








Implications

Feeding Se to gestating and lactating ewes above the current maximum tolerable

level (2 mg/kg) does increase the Se concentration in colostrum and subsequently

produced milk. However, this practice does not increase milk Se concentrations to a

level at which their nursing lambs suffer from Se toxicosis. Likewise, feeding increased

Se to ewes does increase plasma and tissue Se in lambs but not to a concentration above

those previously found in sheep determined not to be suffering from Se toxicity.

Moreover, data from other species even suggests that feeding increased Se to gestating

and lactating animals may produce colostrum of higher quality that may be beneficial to

their offspring. Data from this and other recent research has now established that the

maximum tolerable level of Se, as selenite, for sheep to be considerably higher than the

previously suggested 2 mg/kg.

Summary

The objective of this 504-d experiment was to evaluate and compare the effects of

six levels of dietary selenium (Se) on ewes' milk and the Se status of their lambs prior to

weaning. Sodium selenite was added to a basal diet to provide 0.2 (control), 4, 8, 12, 16,

and 20 mg/kg dietary Se for ewes during gestation and lactation over two lambings.

Colostrum Se ranged from 257 to 3542 gg/L and increased linearly as dietary Se

increased (P < 0.001) in both years. Ewe milk Se ranged from 75 to 2228 ptg/L and also

increased linearly as dietary Se increased (P < 0.01). In general, ewes receiving > 12

mg/kg Se produced higher milk Se than controls. Blood samples were collected from

lambs before nursing and at 3, 28, and 56 d of age to evaluate plasma Se concentrations.

At birth, lamb plasma Se ranged from 74 to 775 gg/L and was affected (P < 0.001) by the

Se concentration of the ewe diets, which indicates placental transfer of Se. Lambs from








ewes receiving dietary Se at 20 mg/kg had higher (P < 0.05) plasma Se than controls at

birth and 3, 28, and 56 d of age in both years. Selenium concentration in lamb testes

collected at 70 d of age was also affected by Se content of ewe diets. In year one, lambs

whose dams received 16 or 20 mg/kg Se had higher (P < 0.05) testicular Se than controls,

and no differences in testicular Se were observed in year two. No signs of Se toxicosis

were observed in lambs regardless of dietary Se concentration of the ewes' diet. These

results suggest that ewes consuming up to 20 mg/kg inorganic Se may give birth to

normal lambs, and that the lambs may not suffer from Se toxicosis before weaning.

Selenium as sodium selenite may be fed to ewes at concentrations greater than the current

maximum tolerable levels (2 mg/kg) without adversely affecting their offspring.








Table 4-1. Diet composition (as-fed) for Se (selenite) supplemented ewes
Ingredient % as-fed
Ground yellow corn 53.75
Cottonseed hulls 22.00
Soybean meal (47.5% CP) 16.00
Alfalfa meal (14% CP) 3.00
Soybean oil 3.00
Trace mineral mixb 1.00
Ground limestone 1.25
Vitamins A & D C
aSelenium levels in diet (as analyzed): 0.29, 3.77, 7.54, 11.01, 15.48, and 19.05 ppm
for treatments 0.2, 4, 8, 12, 16, and 20 ppm, respectively.
bTrace mineral mixture supplied between 96.5% and 98.5% NaCl, and provided per kg
of diet: 1.0 mg Co (as carbonate), 5.0 mg Cu (as oxide), 0.7 mg I (as iodate), 35 mg Fe
(as oxide), 25 mg Mn (as oxide), and 35 mg Zn (as oxide).
CProvided per kg of diet: 5,000 IU of Vitamin A and 500 IU of Vitamin D3.


Table 4-2. Colostrum selenium concentrations ([tg/L) of ewes receiving different levels
of selenium supplementation as sodium selenitea
Year of experiment
Added Se, mg/kg 11 22
0.2 257b 624 705' 517
4 1300bc 543 1452 bd 421
8 1889bcd 767 3256c 480
12 2072bcd 704 2373cd 462
16 3216c 767
20 3542 d 537 2925 d 741
aData represent least squares standard errors.
b, dMeans within columns lacking a common superscript differ (P < 0.05)
in = 4, 5, 3, 3, 3, and 6 for Se levels 0.2, 4, 8, 12, 16, and 20 mg/kg, respectively.
2n = 4, 6, 5, 5, and 2 for Se levels 0.2, 4, 8, 12, and 20 mg/kg, respectively.








Table 4-3. Milk Se concentrations ( tg/L) from ewes receiving different levels of
dietary Se as sodium selenite in year one'


31
75 b 117
312 117
400bc 117


12 490c 144
16 920d 117
20 628ra 117
aData represent least squares means + SE.
b'c'dMeans within columns lacking a common
1n = 3, 3, 3, 2, 3, and 3 for Se levels 0.2, 4, 8,
2n 5, 5, 3, 3, 2, and 5 for Se levels 0.2, 4, 8,
3n = 5, 6, 3, 3, 3, and 6 for Se levels 0.2, 4, 8,


Table 4-4. Milk Se concentrations ([tg/L)
dietary Se as sodium selenite in year twoa


Days postpartum
28
660 91
121b91
163bc 117
189bc- 117
253bc 144
466c 91


56'
32 b91
165' - 83
160b 117
241' 117
653c 117
538c 83


superscript are different (P < 0.05).
12, 16, and 20 mg/kg, respectively.
12, 16, and 20 mg/kg, respectively.
12, 16, and 20 mg/kg, respectively.


from ewes receiving different levels of


Added Se, mg/kg
0.2
4


31
57b 4- 238
574 b 194
1442d 213


Days postpartum
282
81b 238
646b 180
462b 213


56 3
69b 238
339bc 194
493 b 4- 213


12 895cd 238 689b 238 914c 238
20 933cd 337 923b 475 2228d 476
aData represent least squares means SE.
b,'cdMeans within columns lacking a common superscript are different (P < 0.05).
in = 4, 5, 5, 5, and 2 for Se levels 0.2, 4, 8, 12, and 20 mg/kg, respectively.
2n = 4, 6, 5, 4, and 1 for Se levels 0.2, 4, 8, 12, and 20 mg/kg, respectively.
3n = 4, 5, 5, 4, and 1 for Se levels 0.2, 4, 8, 12, and 20 mg/kg, respectively.


Added Se, mg/kg
0.2
4
8






65


Table 4-5. Plasma Se concentrations ([tg/L) of lambs suckling ewes receiving different
levels of dietary Se as sodium selenite in year onea
Age of lamb, d
Se in ewe diet,mg/kg 01 32 28' 564
0.2 81b 66 111b 66 76b 52 92b 52
4 127be 60 204bc 56 196b 49 246c 49
8 131be 85 330c 73 209bc 73 258bed 73
12 188be 66 333c 85 374 d 66 354cd 66
16 238bc 85 387c 73 297c 73 431 d 66
20 2940 47 648d 60 508d 44 419d 47
aData represent least squares means SE.
b'cdMeans within columns lacking a common superscript are different (P < 0.05).
in 5, 6, 3, 5, 3, and 6 for Se levels 0.2, 4, 8, 12, 16, and 20 mg/kg, respectively.
2n = 5, 7, 4, 3, 4, and 6 for Se levels 0.2, 4, 8, 12, 16, and 20 mg/kg, respectively.
3n = 8, 9, 4, 5, 5, and 11 for Se levels 0.2, 4, 8, 12, 16, and 20 mg/kg, respectively.
4n = 8, 9, 4, 5, 5, and 10 for Se levels 0.2, 4, 8, 12, 16, and 20 mg/kg, respectively.


Table 4-6. Plasma Se concentrations (gg/L) of lambs suckling ewes receiving different
levels of dietary Se as sodium selenite in year twoa
Age of lamb, d
Se in ewe diet,mg/kg 01 32 283 564
0.2 85' 62 74b - 55 81b 55 86b 55
4 182be 41 325c 44 244c 47 287c 47
8 186be 47 601d 47 314cd 55 263c 51
12 253 d 41 686d 51 553e 62 430d 55
20 353' 71 737d 71 775f 87 3401cd 124
aData represent least squares means SE.
b' 'd'e' Means within columns lacking a common superscript are different (P < 0.05).
in = 4, 9, 7, 9, and 3 for Se levels 0.2, 4, 8, 12, and 20 mg/kg, respectively.
2n = 5, 8, 7, 6, and 3 for Se levels 0.2, 4, 8, 12, and 20 mg/kg, respectively.
3n = 5, 7, 5, 4, and 2 for Se levels 0.2, 4, 8, 12, and 20 mg/kg, respectively.
4n = 5, 7, 6, 5, and I for Se levels 0.2, 4, 8, 12, and 20 mg/kg, respectively.



























Figure 4-1. Rambouillet ewe in late gestation fitted with a device to cover the udder and
prevent lambs from suckling until a blood sample was obtained. Device was made from
nylon pantyhose and elastic straps, and held in place with safety pins. (Device courtesy of
Dr. Donald J. Davis, Crossville, TN)


5.5

)5
0E 4.5

4

_3.5
'-,

0 3
2.5

S1.2
E
1.5-

1


0.2 4 8 12 16 20


Se content of ewe diet, mg/kg

Figure 4-2. Effects of Se concentration of ewe diet on testicular Se concentration of their
lambs in year one. Testicular selenium concentrations were 1.67, 1.83, 2.88, 3.26, 3.76,
and 4.25 mg/kg (dry matter basis) for ram lambs suckling ewes receiving 0.2, 4, 8, 12,
16, and 20 mg/kg dietary Se, respectively. SE = 0.56.













CHAPTER 5
COMPARATIVE EFFECTS AND TOLERANCE OF VARIOUS DIETARY LEVELS OF
SE AS SODIUM SELENITE OR SE YEAST ON BLOOD, WOOL, AND TISSUE SE
CONCENTRATIONS OF WETHER SHEEP

Introduction

Selenium was first implicated as an essential nutrient for animals by Schwarz and

Foltz (1957). Prior to that, Se was viewed primarily as a detriment to livestock which

was documented by Franke (1934) and Moxon (1937). Selenium deficiency is far more

prevalent worldwide than toxicity. However, Se toxicity is a greater concern to livestock

producers and nutritionists, as toxicities are more difficult than deficiencies to control.

Current estimates put the maximum tolerable level of Se at 2 mg/kg for the major

livestock species (NRC, 1980) and no differentiation exists for tolerable levels between

ruminants and monogastric animals. However, previous research suggests that inorganic

Se (e.g., sodium selenite) may be reduced to insoluble selenide by microorganisms in the

rumen, thus reducing overall absorption of Se by ruminant animals (Butler and Peterson,

1961; Hidiroglou et al., 1968). Wright and Bell (1966) reported that swine retained 77%

and sheep retained 29% of an oral dose of inorganic Se. The NRC makes no distinction

between inorganic and organic (e.g., Se yeast or seleno-methionine) forms of Se for

current maximum tolerable levels, though the chemical form of dietary Se leads to

markedly different physiological responses of livestock (Knowles et al., 1999; Pehrson et

al., 1999; Gunter et al., 2003). Kim and Mahan (2001) reported more accumulation of Se

in the plasma and tissues of swine fed high dietary levels of Se as Se yeast compared to

the same Se levels as sodium selenite, and that Se toxicity occurred sooner and its clinical








signs were more severe when inorganic Se was used as the dietary source. In concluding

that > 5 mg/kg dietary Se, regardless of source, did produce signs of Se toxicity in

growing swine, those authors postulated that the greater tissue retention of organic Se

may reduce the incidence of Se toxicity. Based on these findings and the increasing use

of organic forms of Se for supplementation to livestock, an experiment was conducted to

evaluate and compare effects of feeding Se as sodium selenite or Se yeast at high dietary

levels on serum, whole blood, wool, and tissue Se concentrations of wether sheep.

Materials and Methods

All animal procedures were conducted within the guidelines of and approved by

the University of Florida Institutional Animal Care and Use Committee. This experiment

was conducted from June 4, 2002 to July 29, 2003 at the University of Florida Sheep

Nutrition Unit located in southwestern Alachua County, FL. Twenty-eight, 2-yr-old,

Rambouillet-crossbred wethers were weighed (62.3 8.5 kg) and received 2-ml

ivermectin dewormer s.c. (Ivomec; Merial Ltd. Iselin, NJ). Wethers were randomly

assigned to one of eight dietary treatments for a 60-wk study. Dietary treatments were

arranged as a 2 x 4 factorial with 0.2, 20, 30, and 40 mg/kg Se (as-fed) as four dietary

levels and Se yeast and sodium selenite as two Se sources added to a corn-soybean meal-

cottonseed hull basal diet (Table 5-1). Feed-grade yeast was used as a carrier for the

sodium selenite in order to alleviate differences in the palatability and protein content of

the diets. The basal diet was formulated to meet animal requirements for protein, energy

as TDN, vitamins, and minerals for this class of sheep (NRC, 1985). Animal numbers

per treatment were three for 0.2 (control) and 20 mg/kg Se, and four each for 30 and 40

mg/kg Se treatments for both Se sources. Wethers were housed by treatment group in

covered wooden pens (53.5 m2) with earth floors and ad libitum water.








Diets were fed at 909 g'wether''d- throughout the experiment. Samples of each

diet were taken every 28 d, ground (1 mm), and frozen at 0C until analysis.

Wether BW was recorded on d 0 and for every eight wk thereafter, for the

remainder of the study. A 10-mL blood sample for serum analysis was collected using an

18-gauge needle into a vacutainer tube with no additive (Vacutainer; Becton Dickinson,

Franklin Lakes, NJ) every 12 wk, via jugular venipuncture, allowed to stand for 20 min,

centrifuged at 700 x g for 25 min, and serum stored frozen at 0C until Se analysis. An

additional 10 mL of blood was collected into a heparinized vacutainer tube (Vacutainer;

Becton Dickinson, Franklin Lakes, NJ). This additional 10-mL sample was also

collected every 12 wk for the remainder of the experiment and stored frozen at 0C as

whole blood until analysis.

The wool around the jugular was shorn initially and regrowth was collected

beginning at wk 12 and every 12 wk thereafter. The collected wool was washed with a

commercial hair shampoo (Alberto V05; Alberto-Culver Co., Melrose Park, IL), to

remove oil and dirt, rinsed well with deionized water, dried, stored at room temperature,

and later analyzed for Se.

At the termination of the experiment (wk 60), wethers were slaughtered by

stunning and exsanguination, following USDA procedures at the University of Florida

Meats Laboratory. Immediately prior to slaughter, a 10-mL sample of blood was

collected using an 18-gauge needle into a vacutainer, centrifuged at 700 x g for 25 min,

and serum frozen at 0C for analysis of albumin and the following enzymes: alkaline

phosphatase (Alk Phos), alanine transaminase (ALT), aspartate transaminase (AST),

creatinine phosphokinase (CK), and gamma glutamyl transferase (GGT).








Samples of brain, diaphragm, heart, hoof tip, kidney, liver, and psoas major

muscle were collected, and frozen (00C) until analyzed for Se. Sections (1 cm3) of liver,

heart, kidney, diaphragm, and psoas major muscle from all animals were placed in 10%

neutral-buffered formaldehyde for subsequent microscopic evaluation for evidence of Se

toxicosis.

For histopathological evaluation, the tissue samples fixed in buffered formalin

were embedded in paraffin and sectioned at 6 microns. All sections were stained with

hematoxylin and eosin, and examined under a light microscope (1OX, 20X, and 40X).

Serum albumin, Alk Phos, ALT, AST, CK, GGT were evaluated on a Hitachi 911

analyzer with reagents from Sigma (Sigma Chemical Co., St. Louis, Mo.). These

procedures were established by the Veterinary Medical Teaching Hospital at the

University of Florida.

Samples of kidney, heart, and liver were evaluated for cell structure changes using

transmission electron microscopy. Tissues were transferred to Trump's fixative, pH 7.2,

for 2 h at room temperature (240 C). Samples were then rinsed in O.1M sodium

cacodylate buffer at room temperature for 1 h. After three 15-min rinses with deionized

water, the tissues were placed in a 1% aqueous uranyl acetate solution for 45 min.

Samples were then dehydrated through a graded ethanol-acetone series at room

temperature. Tissue samples were then infiltrated with and embedded in Spurr's resin.

Silver sections (0.06 um) were cut using an RMC MT 6000XL ultramicrotone (RMC

Products; Boeckeler Instruments, Inc.,Tuscon, AZ) and mounted on formvar-coated

copper mesh grids. Sections were stained with 5% acidic uranyl acetate and Reynold's

lead citrate and examined in a Zeiss 100 microscope (Carl Zeiss, Inc, Thornwood, NY).








Serum, whole blood, wool, tissue, and feed samples were analyzed for Se

concentration using a fluorometric method described by Whetter and Ullrey (1978). To

help ensure reliability of the analytical method, a certified standard (National Bureau of

Standards Bovine Liver SRM-1577a; U.S. Department of Commerce, National Institute

of Standards and Technology, Gaithersburg, MD) was frequently analyzed.

Brain, diaphragm, heart, hoof tip, kidney, liver, and psoas major muscle Se data

were analyzed for effects of treatment using PROC GLM in SAS (SAS for Windows 8e;

SAS Inst., Inc., Cary, NC) in a 2 x 4 factorial arrangement. Pre-planned orthogonal

contrast statements were used to compare means as described by Littell et al. (1998;

2000). PROC MIXED of SAS was used to analyze effects of treatment, time, and the

interaction of treatment x time on BW, serum Se, whole blood Se, and wool Se as

repeated measures with a spatial power covariance structure with respect to d and a

subplot of animal nested within treatment. Pre-planned orthogonal contrast statements

were written to determine differences in means at different sampling intervals.

Regression analysis was used to determine relationships between dietary Se and Se

concentration in serum, whole blood, wool, and tissues.

Results and Discussion

Wether BW was affected by dietary Se level (P < 0.05), source of dietary Se (P <

0.05), time (P < 0.05), and average BW decreased linearly (P < 0.10) as dietary Se level

increased (Table 5-2). Body weights of wethers receiving 30 or 40 mg/kg dietary Se as

Se yeast decreased from wk 0 to wk 60, whereas wethers receiving all other dietary Se

treatments gained weight from wk 0 to wk 60. Previous studies have reported no effect

of Se fed above requirements on BW of feedlot cattle (Perry et al., 1975) and no effects

on BW when included up to 10 mg/kg in the diets of wether sheep (Cristaldi et al., in








press). Likewise, Ullrey et al. (1977) reported that lamb BW were unaffected by dietary

Se level in feeds containing differing proportions of organic and inorganic Se. However,

Kim and Mahan (2001) reported a quadratic decrease in final BW of swine as dietary Se

level was increased using sodium selenite or Se yeast. Those authors observed the most

drastic decreases when selenite Se was added above 10 mg/kg and when Se yeast was

added at 20 mg/kg. Our results differ from the findings with swine, as organic Se had a

more dramatic deleterious affect on BW than did selenite Se. This could be explained by

organic Se not being subject to reduction to selenides by rumen microorganisms as

suggested by previous research (Whanger et al., 1968; van Ryssen et al., 1989; Whanger,

2002) and thus, being more available to cause toxic effects in ruminant livestock.

Seven of 28 wethers died during the course of our study and were subjected to

gross necropsy by pathologists at the University of Florida Veterinary Teaching Hospital.

All wethers were described in good physical condition at time of necropsy with adequate

adipose tissue. Causes of death were determined to be spinal cord compression trauma,

endoparasitism, pulmonary edema, and unknown. One wether from the 20 mg/kg

organic Se group had mild hepatic lipidosis and one wether from the 30 mg/kg organic Se

group exhibited signs of mild myocarditis. However, definitive evidence of death due to

Se toxicosis was not found and the gross lesions seemed to be due to metabolic changes

or were merely background findings.

Serum selenium values (105 23 gg/L) were below the normal range (120 to 150

gg/L) for adult sheep (Aitken, 2001) and did not differ among treatment groups (P =

0.36) at the initiation of the experiment. Serum Se concentrations measured at wk 12, 24,

48, and 60 ranged from 110 to 3922 ptg/L and increased linearly (P < 0.05) as dietary Se








level increased, while a quadratic response (P < 0.05) was observed at wk 36 (Table 5-3).

Serum Se concentrations of wethers were affected by dietary Se level, Se source, and the

interaction of dietary Se level x Se source interaction (P < 0.05). Likewise, over the

entire trial serum Se increased quadratically (P < 0.05) as dietary Se level increased.

Wethers receiving organic Se had higher (P < 0.001) serum Se than did selenite treated

animals throughout the study. Kim and Mahan (2001) reported a linear increase in

plasma Se of swine as dietary Se was increased as organic or inorganic Se. Likewise,

those authors reported an effect of source of Se, with pigs supplemented with organic Se

having higher plasma Se than their inorganically supplemented counterparts. Cristaldi et

al. (2004) reported a linear increase in serum Se as dietary Se was increased, however

those authors used a maximum level of 10 mg/kg dietary Se as selenite. The quadratic

response observed in the present study suggests homeostatic regulation of Se in blood as

dietary levels exceed 30 mg/kg. Serum Se in wethers fed up to 10 mg/kg selenite Se for

52 wk reached 870 gg/L (Cristaldi et al., in press). Wethers in the present experiment

exceeded 870 jig/L when receiving either Se source at 20, 30, or 40 mg/kg and at wk 24

wethers receiving 30 mg/kg organic Se had more than four-fold higher serum Se than the

maximum serum Se reported by Cristaldi et al. (2004). Our data show that at most

collections organic Se produced serum Se of more than double the concentration

produced by feeding selenite Se at the same level. Wethers receiving 20, 30 or 40 mg/kg

organic Se had serum Se above 1500 gg/L throughout the experiment. Aitken (2001)

reported serum Se of 1500 gg/L as a level at which signs of toxicity appear in horses.

Likewise, Aitken (2001) reported that serum Se of 3700 gg/L was evident of Se toxicosis





74

in swine. At wk 24, wethers receiving 30 mg/kg organic Se had serum Se of 3922 gg/L.

At no other time during our study did wether serum Se exceed 3700 gg/L.

Whole blood Se was measured in addition to serum Se because of the possibility

of a more accurate Se measurement since use of whole blood eliminates the possibility of

falsely high Se readings in serum due to hemolysis (Maas et al., 1992). Whole blood Se

responded to Se supplementation in much the same fashion as did serum Se (Table 5-4).

This response supports the high correlation between serum Se and whole blood Se

previously described by Maas et al. (1992) and Cristaldi et al. (2004). Whole blood Se

concentrations of wethers were affected by dietary Se level, Se source, and the interaction

of dietary Se level x Se source interaction (P < 0.05) ranged from 392 to 6259 gg/L, and

overall increased quadratically (P < 0.01) as Se concentration of the wether diets was

increased. Whole blood Se concentrations measured at wk 12, 24, 48, and 60 increased

linearly (P < 0.10) as dietary Se level increased. Whole blood Se in swine increased

linearly as dietary Se as sodium selenite was increased from 0 to 20 mg/kg (Goehring et

al., 1984b) and whole blood in sheep responded linearly to increased dietary Se (Cristaldi

et al., in press). Likewise, those authors reported a strong correlation between Se

concentrations of serum and whole blood. In the present study, whole blood Se

responded in neither a linear nor quadratic manner at wk 36 (P > 0.15). It seems that

organic Se was used in place of inorganic Se for the selenite control diet during that

feeding period, which created a whole blood Se concentration of 1004 gg/L for the

selenite control group at wk 36. On average, whole blood Se responded quadratically (P

< 0.05) as dietary Se level was increased, again suggesting the influence of homeostatic

regulation when dietary Se is increased above 30 mg/kg. Wethers receiving organic Se








had higher (P < 0.001) whole blood Se than did wethers receiving inorganic Se

throughout the study. The maximum whole blood Se concentration observed during our

study was 6259 gg/L. This concentration is well above the range of 2000 to 4000 Ptg/L

for whole blood Se, where clinical signs of Se toxicosis should appear (Rosenfeld and

Beath, 1945; 1946) and likewise is greater than a whole blood Se concentration of

4000gg/L, that Maag and Glenn (1967) described as the blood concentration above which

steers became depressed and inactive. However, wethers on the present study, with the

highest whole blood Se concentrations (> 6200 pg/L) did not exhibit signs of Se toxicosis

(e.g. wool loss, anorexia, abnormal hoof growth). Glenn et al. (1964a) fed sodium

selenate at high levels to range ewes that were similar in BW and breed type to the

wethers on the present study. Those researchers did not induce death by Se toxicosis

with daily oral doses less than 25 mg Se/ewe. Of the 17 deaths reported in their

experiment, only one was induced with a daily dose of 25 mg Se/ewe. Eight deaths were

induced with a daily dose 37.5 mg Se/ewe and eight deaths were induced with a daily

dose 50 mg Se/ewe. Those reported deaths were not by acute Se toxicosis; rather the

ewes received experimental Se doses for at least 80 d before death by Se toxicosis was

induced. In the same experiment, Glenn et al. (1964a) further suggested an avg minimum

toxic level of Se for adult sheep to be 0.825 mg/kg BW when fed for 100 d. Using this

estimate, the minimum toxic level of Se for sheep of the size used in our study would be

51.4 mg/d. Selenium consumption of wethers receiving the highest dietary Se level (40

mg/kg) was 78% of the aforementioned minimum toxic level for sheep. Blodgett and

Bevill (1987) reported the LD50 for sheep, using sodium selenite via i.m. injection, to be

0.7 mg Se/kg BW. Wethers of avg BW, on our study, receiving 1 kg of diet containing








40 mg/kg Se received 91.1% of that LD50 for sheep throughout the experiment.

Furthermore that LD50 for sheep (Blodgett and Bevill, 1987) was established using

injectable Se. Administration of Se parenterally disallows the reduction of selenite Se to

insoluble selenide via ruminal microorganisms as described by (Whanger et al., 1968).

This would suggest that the LD50 for Se in sheep could be considerably higher than

previously thought.

Selenium concentration in new growth wool was measured at wk 12, 24, 36, 48,

and 60 (Table 5-5). Dietary Se level, Se source, time, dietary Se level x Se source, and

dietary Se source x time affected (P < 0.05) wool Se. Wool Se ranged from 1.19 to 39.09

mg/kg and increased linearly (P < 0.001) as dietary Se increased. Wool Se from wethers

receiving organic Se was often more than three-fold higher (P < 0.001) than from wethers

receiving selenite Se at the same dietary level. Increased Se in hair has been reported in

other livestock species. Kim and Mahan (2001) observed a linear response in the hair of

pigs as Se in their diet was increased. Goehring et al. (1984b) reported a quadratic

response in the hair of swine as dietary Se as sodium selenite was increased up to 20

mg/kg. Likewise, Perry et al. (1976) reported increased Se in the hair of feedlot steers as

dietary selenite Se was increased. Cristaldi et al. (2004) reported a linear increase in the

wool of wether sheep as dietary Se was increased. Those authors did not report a

significant Se level x time interaction. However, wool Se of sheep, on the present study

was affected by time and the interaction of Se source x time as wool Se continued to

increase from wethers fed selenite Se and wool Se from wethers receiving organic Se

increased then seemed to reach a peak around wk 48. This suggests that wool Se may

reach a plateau when animals are fed high dietary concentrations of organic Se. Wool Se








concentrations in the present study were more than ten-fold higher than concentrations of

2 to 2.5 mg/kg in wool from wethers fed up to 10 mg/kg dietary Se as selenite (Cristaldi

et al., in press), but never exceeded 40 mg/kg which is less than 45 mg/kg which was

described as the Se concentration in hair of animals suffering from alkali disease

(National Academy of Sciences [NAS], 1971).

Selenium concentrations in brain, diaphragm, heart, kidney, and loin muscle were

affected (P < 0.05) by dietary Se level, Se source, and dietary Se x Se source interaction.

Hoof Se concentration was affected by source (P < 0.05) and liver Se was affected (P <

0.05) by dietary Se level and dietary Se x Se source interaction, and tended to be affected

(P = 0.11) by Se source. Selenium concentrations, on a DM basis, were highest in liver

followed by kidney, heart, hoof, brain, loin, and diaphragm (Table 5-6). This pattern is

similar to a ranking of Se concentrations in tissues of farm animals by (Combs and

Combs, 1986) with the exception of liver and kidney being reversed. However, in

animals fed Se at or below requirements, kidney generally has a higher concentration of

Se than does the liver, but when dietary Se is increased, liver Se quickly becomes higher

in Se than kidney. This supported by the work of Ewan et al. (1968) and the findings of

Cristaldi et al. (2004) in sheep, and McDowell et al. (1977) in swine. Those authors

reported higher liver Se than kidney Se when dietary Se was increased. Unlike minerals

such as Zn and Mn, the status of Se is reflected in many tissues (McDowell, 2003). Brain

Se concentrations ranged from 1.28 to 32.3 mg/kg and brain Se concentrations from

wethers receiving organic Se were higher (P < 0.001) than brain Se from wethers

receiving selenite Se. These results suggest that Se does cross the blood-brain barrier and

that brain Se is influenced by dietary Se. Previous research using sheep supports our








findings of increased Se in brain as dietary Se is increased (Yeh et al., 1995;1997;

Ouazzani et al., 1999). Diaphragm Se ranged from 0.82 to 26.34 mg/kg and tended to

increase linearly (P = 0.089) as dietary Se increased. Diaphragm Se was higher (P <

0.001) in wethers receiving organic Se than from wethers receiving selenite Se. Heart Se

ranged from 1.59 to 33.93 mg/kg and, like brain and diaphragm Se was higher (P <

0.001) in wethers receiving organic Se than from wethers receiving selenite Se.

Selenium concentrations in the hoof tip ranged from 3.44 to 29.20 mg/kg and increased

linearly as dietary Se increased (P < 0.05). Selenium concentrations of hoof tip taken

from wethers receiving organic Se tended (P = 0.07) to be higher than from wethers

receiving inorganic Se. Both Se sources produced hoof Se concentrations higher than 10

mg/kg which was previously reported in animals with alkali disease (NAS, 1971).

Kidney Se tended (P = 0.07) to respond linearly to increased dietary Se and ranged from

8.43 to 77.61 mg/kg. Kidney Se concentrations from wethers receiving organic Se were

higher (P < 0.01) than from wethers receiving selenite Se. Kidney Se concentrations

from the present study are much higher than those reported by Maag and Glenn (1967)

where death due to Se toxicosis was produced in 245 kg Hereford steers. However, the

calves used by those authors received approximately 270 mg Se-steerl-d-1 as sodium

selenite and death was induced within 6 wk. Liver Se concentrations ranged from 2.66

to 132.73 mg/kg and increased linearly (P < 0.001) as dietary Se level increased.

Selenium concentrations in liver from wethers receiving organic Se were not different (P

= 0.34) than liver Se concentrations from wethers receiving selenite Se. Selenium

concentrations in the loin muscle (psoas major), which is often consumed by mankind

ranged from 0.71 to 26.87 mg/kg and tended (P = 0.12) to increase linearly as dietary Se








was increased. Organic Se was more effective (P < 0.001) at increasing Se

concentrations in edible tissue than was selenite Se. As daily intake of Se by humans

declines in some parts of the world, increasing the Se content of foods for human

consumption by manipulating source and level of Se supplementation to livestock is now

of interest to food scientists. Givens et al. (2004) suggested that the Se content of cows'

milk could be increased through the use of Se yeast as the supplemental form of Se to

dairy cows. Our findings indicate that Se content of muscle and organ tissue can be

influenced by source and level of Se supplementation to food animals. In general, Se

concentrations of brain, heart, kidney, liver, and muscle were much higher than those

reported in studies with cattle (Maag and Glenn, 1967) and sheep (Glenn et al., 1964c).

Deaths due to Se toxicosis were induced in both species. In contrast, Se death due to Se

toxicosis was never produced during our study. It is important to note that in the two

previous studies that animals were fed Se at higher levels and for a shorter period of time.

Our findings further agree with Smith et al. (1937) who found that the effects of

continued dosing of Se were cumulative and that Se from organic sources was

accumulated in higher quantities in tissues than Se from inorganic sources.

Most of the heart, diaphragm, loin, liver, and kidney tissues subjected to

histopathological evaluation were free from pathological changes. The findings of

lymphocytes in the portal triads were deemed to be a background finding and

insignificant. Three instances of vacuolic degeneration associated with the cytoplasm of

hepatocytes suggesting fatty degeneration were noted. However, no pattern associating

abnormal pathology to either dietary Se level or source could be established. Therefore,

lesions could not be definitively linked to treatment and could have been metabolic in








nature. Cristaldi et al. (2004) found no abnormalities after microscopic evaluation of

heart, liver, kidney, diaphragm, and muscle from wethers consuming up 10 mg/kg Se for

one yr. Examination of kidney, heart, and liver tissues by transmission electron

microscopy did not reveal any apparent changes in cell structure as related to Se toxicosis

and no differences in tissue cells from controls and wethers receiving 20, 30, or 40 mg/kg

Se were observed.

Concentrations of albumin and activities of Alk phos, ALT, GGT, AST, and CK

in serum collected at the termination of the experiment were, in general, in or below the

normal range for adult sheep (Table 5-7). In instances of Se toxicosis, the activities of

these enzymes would have been increased due to tissue necrosis. Our observations agree

with those reported by Cristaldi et al. (2004) as albumin and enzyme activities in wether

sheep after receiving up to 10 mg/kg Se were in the normal ranges. The lack of elevated

enzymes, which are suggestive of tissue necrosis, further indicates that the wethers on our

study were not suffering from Se toxicosis.

Throughout this 60-wk experiment clinical signs of Se toxicosis (e.g., lameness,

wool loss, and abnormal hoof growth) were not observed, though serum and whole blood

Se concentrations were frequently higher than those described in livestock diagnosed

with hyperselenosis. However, wool Se concentrations from wethers on our study never

reached the levels previously reported in the hair of livestock suffering from alkali

disease. Loin muscle and diaphragm showed no gross lesions at slaughter and no

abnormality was observed with microscopic evaluation. Abnormal pathology in the

kidney, heart, and liver was rare and could, in each case, be attributed to a cause other

than Se toxicosis. No pale focal areas were observed in the myocardium, though








previous research (Glenn et al., 1964b; Smyth et al., 1990) has shown the heart to be the

target organ in instances of Se toxicosis. No abnormalities were prevalent enough to

establish a treatment-related pattern and no wethers receiving the maximum level of

dietary Se on our study (40 mg/kg) showed any abnormal tissue lesions. Further

evaluation of kidney, heart, and liver using transmission electron microscopy also

revealed no cellular abnormalities and enzymes, suggestive of tissue necrosis, were in or

below normal ranges at the termination of the experiment. Without the presence of tissue

damage and clinical signs, it seems that Se toxicosis was not induced in wether sheep fed

up to 40 mg/kg dietary Se as Se yeast or sodium selenite. However, Se concentrations in

serum, blood, wool, and tissues from wethers receiving organic Se indicate that Se

toxicity is dependent on Se source and that much inorganic dietary Se is reduced to

insoluble forms. The work of Cousins and Caimey (1961), Whanger et al. (1968), and

Koenig et al. (1997) support our findings.

Implications

The current estimate of the maximum tolerable dietary level of selenium for sheep

(2 mg/kg) seems to be grossly underestimated. Selenium, whether organic or inorganic in

form, can be fed as high as 40 mg/kg for up to 60 wk without inducing Se toxicosis.

Previously the range between optimal and toxic levels of selenium was reported as

narrow; however data from the present study would suggest that the range is relatively

wide. Increasing dietary selenium level, regardless of source, is an effective means of

increasing selenium in blood and tissues. Organic selenium is more greatly accumulated

by organs and tissues. Manipulation of dietary selenium source and level is an effective

way to change the selenium content of animal tissues commonly consumed by mankind.








Summary

The objectives of this 60-wk experiment were to evaluate and compare effects of

feeding Se as sodium selenite or Se yeast at high dietary levels on serum, whole blood,

wool, and tissue Se concentrations in wether sheep and determine maximum tolerable

level of Se. Twenty-eight, 2-yr-old, Rambouillet-crossbred wethers (62.3 8.5 kg initial

BW) were utilized in a 2 x 4 factorial arrangement with 0.2, 20, 30, and 40 mg/kg dietary

Se (as-fed) from sodium selenite or Se yeast added to a corn-soybean meal basal diet.

Wethers were weighed at 8-wk intervals, serum Se, whole blood Se, and wool Se were

measured every 12 wk, and samples of brain, diaphragm, heart, hoof, kidney, liver, and

loin muscle and serum samples for evaluation of albumin and enzyme activities were

collected at the termination of the experiment. Wether BW was affected by dietary Se

level (P < 0.05), source of dietary Se (P < 0.05), and time (P < 0.05). Average BW

decreased linearly (P < 0.10) as dietary Se level increased, though most wethers gained

BW. Serum Se, whole blood Se, and wool Se concentrations were affected (P < 0.05) by

dietary level of Se and source of Se. Serum Se and whole blood Se ranged from 110 to

3922 gg/L and 392 to 6259 gg/L, respectively, and increased in a quadratic fashion as

dietary Se level increased (P < 0.05) and wool Se ranged from 1.19 to 39.09 mg/kg and

responded linearly (P < 0.05) to increased dietary Se. Serum Se, whole blood Se, and

wool Se concentrations from wethers receiving organic Se were higher (P < 0.01) than

those from wethers receiving inorganic Se. Selenium concentrations in brain, diaphragm,

heart, hoof, kidney, liver, and loin muscle were affected (P < 0.05) by dietary Se level,

with higher Se concentrations generally observed in tissues from wethers receiving

organic Se. Though Se concentrations in serum, blood, wool, and major organs at most

times exceeded concentrations previously reported in livestock suffering from Se








toxicosis, a pattern of clinical signs of Se toxicosis was not observed in this experiment.

Microscopic evaluation of liver, kidney, diaphragm, heart, and psoas major muscle did

not reveal definitive evidence of Se toxicosis in wethers on any dietary Se treatment.

Wethers under our experimental conditions tolerated up to 40 mg/kg dietary Se for 60

wk, though differences in Se source were observed. Contrary to previous thought, the

range between optimal and toxic dietary level of Se is not narrow. The maximum

tolerable level of dietary Se, regardless of source, is much higher than the current

estimate of 2 mg/kg.









Table 5-1. Diet composition (as-fed) for Se supplemented wethersa
Ingredient % as-fed
Ground yellow corn 58.00
Cottonseed hulls 30.00
Soybean meal (47.5% CP) 6.50
Soybean oil 3.00
Trace mineral mixb 1.00
Ground limestone 1.00
Ammonium chloride 0.50
Vitamins A & D C
aSelenium levels in diet (as analyzed): 0.48, 20.48, 30.86, and 38.10 mg/kg for Se levels 0.2, 20, 30, and
40 mg/kg from sodium selenite, respectively; 0.54, 20.26, 30.71 and 37.65 mg/kg for Se levels 0.2, 20,
30, and 40 mg/kg Se from Se yeast, respectively
bTrace mineral mixture supplied between 96.5% and 98.5% NaC1, and provided per kg of diet: 1.0 mg
Co (as carbonate), 5.0 mg Cu (as oxide), 0.7 mg I (as iodate), 35 mg Fe (as oxide), 25 mg Mn (as oxide),
and 35 mg Zn (as oxide).
'Provided per kg of diet: 5,000 IU of Vitamin A and 500 IU of Vitamin D3.


Table 5-2. Effects of four dietary levels of Se as sodium selenite or Se yeast on BW of
wethersa
Se source
Sodium selenite- Se yeast
-Dietary Se level, mg/kg
0.2 20 30 40 0.2 20 30 40
Week Wether BW, kg SEM
0 61.2 65.0 65.8 58.9 57.0 55.5 67.9 64.3 4.5
8 59.1 63.6 59.8 49.3 56.8 51.5 54.3 50.6 4.5
16 61.8 68.2 57.9 48.2 57.9 55.1 54.8 50.0 5.70
24 65.1 70.0 59.2 52.4 60.0 57.3 53.3 48.2 7.5e
32 70.9 76.7 63.6 56.1 68.2 59.5 51.1 52.7 9.0e
40 70.5 61.8 61.8 57.3 74.7 51.8 47.3 50.9 8.4
48 77.6 81.8 69.8 62.9 70.5 65.5 38.6 54.5 9.0c
60 83.3 85.6 76.5 67.9 78.2 61.8 50.2 54.5 10.4
Avg 68.7 71.6 64.3 56.6 65.4 57.3 52.2 53.2 7.2bcde
aData represent least squares means and pooled SE.
bDietary Se level response (P < 0.05).
'Selenium source response (P < 0.05).
dTime response (P < 0.05).
e Dietary Se level linear response (P < 0.10).






85


Table 5-3. Serum Se concentrations of wethers fed four dietary levels of Se as sodium
selenite or Se yeast
Se source
-Sodium selenite Se yeast
Dietary Se level, mg/kg
0.2 20 30 40 0.2 20 30 40
Week Serum Se, tg/L SEM
12 157 548 788 1000 412 2583 3210 2458 249bcdc
24 130 1683 1487 1724 354 2639 3922 1585 826be
36 444 851 960 1083 540 3283 2086 1409 250bdf
48 110 822 1219 1496 292 2428 2076 1831 253bce
60 119 610 886 1250 424 1699 2712 2549 331bee
Avg 192 903 1068 1311 404 2526 2801 1966 395bedf
aData represent least squares means and pooled SE.
bDietary Se level response (P < 0.05).
'Selenium source response (P < 0.05).
dDietary Se level x Se source interaction (P < 0.05).
eDietary Se level linear response (P < 0.05).
fDietary Se level quadratic response (P < 0.05).


Table 5-4. Whole blood concentrations of wethers fed four dietary levels of Se as
sodium selenite or Se yeasta
Se source
Sodium selenite-- Se yeast
Dietary Se level, mg/kg-
0.2 20 30 40 0.2 20 30 40
Week Whole blood Se, gtg/L SEM
12 392 1172 1484 1315 1183 4344 4290 5484 475bcde
24 420 1551 2228 2353 1661 4521 6259 5780 415bee
36 1004 1021 1708 2406 1549 5018 4841 1972 635c
48 393 1772 1977 2416 1068 5061 5220 4929 298bcde
60 402 1258 1621 2043 1500 1759 3629 4914 435bee
Avg 522 1355 1804 2107 1392 4028 4803 4408 534bcdf
aData represent least squares means and pooled SE.
bDietary Se level response (P < 0.05).
cSelenium source response (P < 0.05).
dDietary Se level x Se source interaction (P < 0.05).
e Dietary Se level linear response (P < 0.10).
fDietary Se level quadratic response (P < 0.05).






86


Table 5-5. Wool Se concentrations of wethers fed four dietary levels of Se as sodium
selenite or Se yeast
Se source
Sodium selenite Se yeast
Dietary Se level, mg/kg
0.2 20 30 40 0.2 20 30 40
Week Wool Se, mg/kg SEM
12 1.37 3.27 6.69 4.15 3.78 12.67 21.09 24.26 3.80ce
24 1.47 3.57 5.72 11.92 7.04 31.58 35.69 37.30 2.87bd
36 1.68 6.02 9.85 10.85 5.70 18.99 22.79 21.29 4.72ce
48 1.19 3.15 5.64 7.23 6.39 24.81 39.09 29.65 2.22bcd
60 1.29 3.90 5.01 6.23 4.38 23.22 25.65 25.99 2.01bcd
Avg 1.40 3.98 6.58 8.08 5.46 22.25 28.87 27.70 3.38bcdefg
aData represent least squares means and pooled SE.
bDietary Se level response (P < 0.05).
cSelenium source response (P < 0.05).
dDietary Se level x Se source interaction (P < 0.05).
'Dietary Se level linear response (P < 0.10).
fTime response (P < 0.05).
gTime x Se source interaction (P < 0.05).


Table 5-6. Effects of four dietary levels of Se as sodium selenite or Se yeast on tissue
Se of wethersa
Se source
Sodium selenite- Se yeast
Dietary Se level, mg/kg
0.2 20 30 40 0.2 20 30 40
Tissue Se concentration, mg/kg SEM
Brain 1.28 4.22 4.74 6.87 6.12 21.90 32.30 18.71 0.99bcd
Diaphragm 0.82 4.74 3.33 7.81 5.28 10.30 26.34 20.71 2.69bdc
Heart 1.59 3.80 5.13 6.23 6.35 23.77 28.71 33.93 2.43cd
Hoof 3.44 8.79 9.68 13.78 6.26 12.53 29.20 23.66 5.52ce
Kidney 8.43 19.94 27.93 27.89 22.26 33.96 77.61 36.28 6.87bcde
Liver 2.66 31.72 41.42 78.18 15.67 23.42 132.73 41.24 18.17de
Loin 0.71 3.13 4.41 5.13 5.73 14.69 23.51 26.87 1.05bcd
aData represent least squares means and pooled SE.
bDietary Se level response (P < 0.05).
'Selenium source response (P < 0.05).
dDietary Se level x Se source interaction (P < 0.05).
eDietary Se level linear response (P < 0.10).






87


Table 5-7. Amount of albumin and tissue enzyme activities present in serum of Se
supplemented wethers'ab
Se source
Sodium selenite-- Se yeast
Dietary Se level, mg/kg
0.2 20 30 40 0.2 20 30 40 Normal
Enzyme Se concentration, mg/kg .. range
Albumin, g/dL 3.0 3.0 2.9 2.7 2.8 2.0 2.5 2.9 2.4 4.0
AlkPhos, IU/L 127.3 141.7 128.8 153.5 105.7 32.0 50.5 172.0 68-387
ALT, IU/L 12.0 12.0 10.8 9.3 9.7 3.0 2.5 6.0 11 -40
AST, IU/L 81.7 100.7 108.8 81.8 94.7 83.0 65.0 53.0 60-280
GGT, IU/L 49.3 51.7 61.8 60.5 54.3 42.0 46.5 59.0 15 -60
CK, IU/L 123.7 139.7 75.0 103.0 109.0 48.0 50.0 51.0 00-584
aSerum sample collected at wk 60.
bGGT and CK ranges were established by University of Florida Veterinary Teaching Hospital.













CHAPTER 6
EFFECTS OF FORM OF PARENTERAL OR DIETARY SELENIUM
SUPPLEMENTATION ON BODY WEIGHT AND BLOOD, LIVER, AND MILK
CONCENTRATIONS IN BEEF COWS

Introduction

Many areas of the United States have selenium deficient soils (McDowell, 2003)

and may produce forages and grains which are unable to provide adequate Se to

livestock. Selenium deficient brood cows may give birth to calves which are stillborn,

premature, weak, or afflicted with nutritional muscular degeneration (Maas, 1983; Corah

and Ives, 1991). Likewise, even with adequate blood Se at birth, calves suckling Se

deficient dams are susceptible to becoming Se deficient (Pehrson et al., 1999). Without

adequate dietary or parenteral Se supplementation, brood cows may suffer from

infertility, retained placentas, ovarian cysts, metritis, silent estrus periods, and/or poor

weight gains (Dargatz and Ross, 1996).

In cattle, it has been well established that Se crosses the placenta (Koller et al.,

1984; Van Saun et al., 1989), that dietary Se is transferred to milk (Conrad and Moxon,

1979), and that positive correlations exist between blood Se of cows and blood Se of their

calves (Kincaid and Hodgson, 1989; Enjalbert et al., 1999; Pehrson et al., 1999). The

chemical form of Se affects its metabolism and previous research has shown differences

in blood, milk, and liver Se concentrations due to form (organic vs inorganic) of

supplemental Se (Knowles et al., 1999; Gunter et al., 2003; Valle et al., 2002).

Selenium is often supplemented as sodium selenite and included in free-choice

livestock mineral mixtures. However, Se may be supplemented through subcutaneous








injection of barium selenate, sodium selenate or sodium selenite and, in ruminants, with

slow-release, long lasting ruminal Se boluses or pellets. With the recent Food and Drug

Administration approval of Se yeast for use in ruminant diets, livestock producers now

have more choices of form and method of Se supplementation. The objective of this

experiment was to evaluate and compare effects of form and method of Se

supplementation on blood, liver, and milk Se concentrations in beef cows.

Materials and Methods

All animal procedures were conducted within the guidelines of the University of

Florida Institutional Animal Care and Use Committee. Animals were housed at the

University of Florida Boston Farm-Santa Fe Beef Unit located in Northern Alachua

County, Florida. On August 6, 2002, 43 Angus cows, aged 2-3 yr, (mean age = 2.67 yr)

were palpated to diagnose pregnancy and estimate d in gestation. All cows were

determined pregnant and gestation estimates ranged from 115 to 130 d. Each animal

received a chemically altered modified live 4-way viral + vibriosis and leptospirosis

vaccination (Cattlemaster 4+VL-5; Pfizer Animal Health, Exton, PA) and fly control

(Permectrin 10% EC pour-on; Boehringer Ingelheim Vetmedica, Inc., St. Joseph, MO)

according to manufacturer directions. Cows were weighed (average initial BW = 417

46 kg), stratified by age and assigned to one of five treatment groups for a 365 d study.

The five treatments were 1) no Se supplementation, control group, 2) one subcutaneous

injection of 9 mL (50 mg Se'mL1) barium selenate (Deposel Multidose; Novartis New

Zealand, Ltd., Auckland, NZ) at the initiation of the experiment, 3) three subcutaneous

injections of 5 mL (5 mg SemL1) of sodium selenite + 68 IU vitamin E as dl-alpha

tocopheryl acetate (Mu-Se; Schering-Plough Animal Health, Union, NJ), one at the

initiation of the experiment and one every four mo thereafter, 4) free-choice access to a








mineral mixture containing 26 mg Se/kg as sodium selenite (Southeastern Minerals, Inc.,

Bainbridge, GA), or 5) free-choice access to a mineral mixture containing 26 mg Se/kg

as Se yeast (Sel-Plex; Alltech, Inc, Nicholasville, KY). All cows grazed bahiagrass

(Paspalum notatum) pastures and were supplemented with bahiagrass (Paspalum

notatum) hay, molasses-based liquid supplement ad libitum, and whole cottonseed and

pelleted citrus pulp at rates of 0.68 kg-cow-ld1 and 1.81 kg-cowl-d1, respectively, from

November, 2002 through March, 2003.

Treatment groups receiving no Se or injectable Se were housed together and had

access to a free-choice mineral mixture containing no Se (Table 6-1). Cows receiving Se

via free-choice mineral mixtures were housed in separate groups and had access to the

same mineral mixture with added Se as sodium selenite or Se yeast for treatments 4 and

5, respectively (Table 6-1). All free-choice mineral mixtures were offered in wooden

mineral feeders and protected from rain.

Blood samples for plasma analyses were collected via jugular venipuncture into

10-mL heparinized tubes (Vacutainer; Becton-Dickinson, Franklin Lakes, NJ) at the

initiation of the study (d 0) and at d 365. Calving occurred over a 24 d span between

December 31, 2002 and January 23, 2003 and whole blood samples were collected in the

same manner from all cows immediately after parturition and at 30, 90, and 205 d

postpartum. A colostrum or milk sample was also collected on those days into a 15-mL

plastic centrifuge tube. Forty-one calves, 24 male, 17 female, were born alive and

unassisted. One cow in the free-choice selenite group had a stillbirth and one control cow

aborted very early in the experiment and both were removed from the study. Liver

biopsies were performed on all cows using the technique described by Chapman et al.