, exaggzumz.
.5 4“...»sz
$1,

#3.. "a“.
_
\ v

5?.
it? I." 7.!
. .l ‘14
nu! as». I;

‘ 2
3.... ans.
3...... r;

1. 3:4
i.

 

 

 

..
. I 4..
I.
i . . .¥...5,3..,.s.
-: \O' . . .
IV -2 3
ix! .9.

 

fame

2502'

. LIBRARY
M1chigan State
University

This is to certify that the
dissertation entitled

IMPACT OF NUTRITIONAL SODIUM ZEOLITE A
SUPPLEMENTATION IN THE EQUINE AND BOVINE

presented by

KARI KRICK TURNER

has been accepted towards fulfillment
of the requirements for the

Ph.D. degree in Animal Science

 

 

:Vé/V-x {fl //\(3fl1~

Major Professor’s Signature

Date

 

MSU is an Affirmative Action/Equal Opportunity Institution

 

PLACE IN RETURN BOX to remove this checkout from your record.
To AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE

DATE DUE

DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

2/05 p:/ClRC/DateDue.indd-p.1

IMPACT OF NUTRITIONAL SODIUM ZEOLITE A SUPPLEMENTATION IN THE
EQUINE AND BOVINE

By

Kari Krick Turner

A DISSERTATION
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
DOCTOR OF PHILOSOPHY
Department of Animal Science

2005

ABSTRACT
IMPACT OF NUTRITIONAL SODIUM ZEOLITE SUPPLEMENTATION IN THE
EQUINE AND BOVINE
By
Kari Krick Turner
This group of experiments focused on determining the effects of dietary sodium

zeolite A (SZA) supplementation on animal mineral concentrations (plasma, milk or
tissue) and joint and bone physiology. Sodium zeolite A is broken down into silicon and
aluminum in the digestive tract and contains approximately 650 mg Si/kg SZA and
130,000 mg Al/kg SZA. Eight two-year-old Standardbreds were used to evaluate SZA’s
effect on osteochondrotic lesions in the metacarpo-, metatarsophalangeal and tarsocrural.
To determine alterations in plasma and milk mineral concentrations following SZA
supplementation, 20 lactating Holsteins were utilized. Twenty Holstein bull calves were
used to determine the effects of SZA supplementation on plasma and tissue mineral
concentrations, mineral retention and absorption, qualities of the fused third and fourth
metacarpal bones, synovial fluid hyaluronic acid (HA) concentration, and cartilage
glycosaminoglycan (GAG) content. Bone metabolism markers were analyzed in both
cows and calves. There were no effects of SZA on osteochondrotic lesions in
Standardbreds. Feed intake and milk production in the cows were decreased by SZA
(P<0.003). Plasma phosphorus concentrations were decreased in the SZA-treated cows
and plasma calcium concentrations were increased (P<0.0002). Milk aluminum
concentration was increased, and phosphorus concentration was decreased in the SZA-

treated cows (P<0.03). Plasma magnesium concentration was decreased by SZA

supplementation in the calves, and plasma phosphorus concentration tended to be
decreased (P<O. 1). Aluminum retention was increased in the calves (P=0.001), but
plasma aluminum concentrations did not change. Aluminum content was substantially
increased in all analyzed tissues except bone (P<0.05). There was little effect of SZA on
bone or joint physiology, with an increase in deoxypyridinoline in the calves as the only
difference (P<0.05). The results from the series of experiments described here suggest
that SZA has an impact on mineral metabolism. The use of SZA as a dietary supplement
to animals should be reconsidered. The substantial accumulation of tissue aluminum and
therefore potential adverse consequences would preclude any benefits, regardless of
whether or not the animal is intended for human consumption. The possibility of excess

aluminum in human diets is a situation which should be avoided.

Copyright by
KARI KRICK TURNER
2005

ACKNOWLEDGEMENTS

First, I would like to thank Dr. Brian Nielsen. Thank you for believing in me and
giving me the opportunity to pursue my doctoral degree (and the elusive big buck).
Luckily I was able to obtain one of those. Thanks for all the encouragement in whatever
I was doing at the time — research, teaching, applying for jobs, ice skating, wrapping legs
etc. You helped me grow from a Masters student to a PhD student, and now from a PhD
student into a faculty member. Thanks also for all the laughter and good times. But I
must admit, I will not be missing the “Does that thing have a Hemi?”

I would like to thank Dr. Mike Orth for his guidance and editing. Thanks for all
the “‘2?” and “what?” throughout my dissertation -— they spurred more thoughts and
helped point me in the right direction. I also enjoyed all of our sports talks, even though
you did get a jab or two in on the Gators.

Thanks also to the rest of my committee. Dr. Diana Rosenstein — I thoroughly
enjoyed our talks in the car and in your office. Thanks for all the hours driving around to
the Standardbred farms in the snow and ice. I still owe you a tow out of the snow bank,
don’t I? Dr. Chris Womack — Your classes were my absolute favorite. Thanks so much
for allowing a horse girl (and a VaTech one at that!) into your human classes, I appreciate
your willingness to let me adapt the assignments to my interests. Dr. Hal Schott — I
appreciated your guidance and insight into the projects. Thanks for questioning things as

we went along — it definitely helped develop the projects. I also appreciated your help in

obtaining supplies for the calf project — we definitely couldn’t have done it without your
help!

I also would like to thank Dr. Gretchen Hill. I enjoyed picking your brain and
working with you in the 110 labs. Thanks for all the encouragement, you made me
believe in myself as a teacher. I also appreciate the help you gave me during the job
search and application process.

Dr. Dave Hawkins also deserves thanks. Thank you for giving me the
opportunity to try out my teaching wings with all the 110 students. I learned a lot by
watching you, and the other great teachers (Dr. Hill, for example) that taught sections of
the class. I learned much more than just “how to teach”, but also how to care for the
students and treat them as individuals, not just numbers.

Adrienne Woodward, you are next on my list. When I heard you were coming in
Spring 2003, I was excited at getting a new grad student because I was lonely as the only
one. But then you finally got here (after a long drive through the wintery weather —
welcome to M1) and I realized right away you weren’t just a new grad student, you were
a new friend. Thanks for taking the time to teach me the basics of horse judging. Thanks
for all the help on the calf project, even though it almost cost you a thumb. You amaze
me with all your abilities, and I am so proud of the great things you are accomplishing
with the Equestrian Team. I am glad that you are considering a career in academia — I’m
looking forward to having you as a colleague.

Cara O’Connor. I think this is the hardest one to write since you and I don’t use
words when we communicate. And a big hug just won’t sum it all up. I never expected

when I started at MSU to make such a wonderful friend. I absolutely treasure every

vi

moment I spent with you, even when you ripped off my windshield wiper. I have learned
so much from you it is unbelievable. Everything from quilting to baking to assembling
skeletons to lab techniques to harnessing and driving a team. Thanks for helping with all
my projects, even after you had almost lost a thumb (although it was a boring way to
almost lose it - were you and Adrienne trying to get me to pass out???) I just have one
more thing to say, and I know you’re waiting for it (Adrienne please join me): We have a
maroooooooon single cart!!

I would like to thank the Turner family. Thank you for welcoming me into your
family. I also appreciate all your support throughout the years — you didn’t have to do
what you did (are doing), but you did, and that warms my heart.

And thanks to two of the most special people in my life - my parents. Without
your love, support and guidance I would not be where I am today. Thanks for nurturing
my love of animals, particularly that of horses. Thanks for instilling in me a value system
that has led me to be successful thus far in life. I also appreciate your support of all my
schools — but remember that the Orange and Blue come first.

And lastly, to my most treasured person, Gregory Turner. Thank you. For
everything. I could not have done this “he-rowing” journey without you. You have been
so selfless throughout this entire process. You always put me and my goals first. When
it came down to accepting a job, you supported and encouraged me when you knew it
would mean that once again we would be living in different states. Even though I know
you really wanted me there so you could get FL-GA tickets. I love you for it, regardless.
Hurry up and move to Georgia.

And with that, I’m outta here. The world is Gator bait, and it’s time I took a Chomp.

vii

TABLE OF CONTENTS

List of Tables ...................................................................................... x
List of Figures ..................................................................................... xii
Chapter 1: Introduction ........................................................................... 1
Chapter 2: Review of Literature ................................................................. 4
Silicon ...................................................................................... 4
Silicon essentiality ........................................................................ 7
Bone formation and mineralization ........................................... 8
Glycosaminoglycan synthesis and structure ................................. 9
Osteochondrosis .............................................. l 2
Silicon toxicity ........................................................................... 13
Supplemental bioavailable silicon .................................................... 14
Sodium zeolite A ............................................................... l6
Silicon and aluminum interactions ................................................... 19
Aluminum in livestock ................................................................. 20
Aluminum toxicity ...................................................................... 25
Analyses .................................................................................. 27
Silicon ........................................................................... 27
Glycosaminoglycan concentration ........................................... 28
Rationale for experiments .............................................................. 30

Chapter 3: Silicon supplementation and osteochondrotic lesions in two-year-old

Standardbreds ............................................................................ 31
Summary .................................................................................. 3 1
Introduction ............................................................................... 32
Materials and Methods .................................................................. 33
Results ..................................................................................... 36
Discussion .................................................................................. 37
Chapter 4: Sodium zeolite A supplementation to lactating Holsteins ...................... 42
Summary ................................................................................... 42
Introduction ................................................................................ 43
Materials and Methods ................................................................... 45
Results ...................................................................................... 50
Discussion .................................................................................. 53
Chapter 5: Sodium zeolite A supplementation to newborn Holstein calves ............. 59
Summary ................................................................................. 59
Introduction .............................................................................. 60
Materials and Methods ................................................................. 62

viii

Results .................................................................................... 73

Discussion ............................................................................... 84
Chapter 6: Conclusions and Implications ..................................................... 98
Appendix A ....................................................................................... 104

Summary ................................................................................. 104

Introduction .............................................................................. 105

Materials and Methods ................................................................. 106

Results .................................................................................... 109

Discussion ............................................................................... 1 l 1
Appendix B ....................................................................................... 117
Appendix C ....................................................................................... 121
Literature Cited ................................................................................... 122
Vita ................................................................................................. 134

ix

LIST OF TABLES

Table 1. Responses of tissue minerals to high dietary Al (P < 0.10) ......................... 24

Table 2. Detailed horse information with trainer, sex, gait, joints affected, treatment, and
study duration ............................................................................... 36

Table 3. Total mixed ration (TMR) formulation and mineral analyses for control (CO)
and SZA-supplemented (SS) diets ...................................................... 47

Table 4. Treannent*time effects (P < 0.05) on feed intake and milk production in control
(CO) and SZA-supplemented (SS) cows ................................................ 52

Table 5. Overall effect of treatment between control (CO) and SZA-supplemented (SS)
cows .......................................................................................... 52

Table 6. Treatrnent*time effects (P < 0.1) on mineral intake and plasma and milk mineral
concentrations in control (CO) and SZA-supplemented (SS) cows ................. 53

Table 7. Overall effect of treatment on mineral intake and plasma and milk mineral
concentrations in control (CO) and SZA-supplemented (SS) cows ................. 53

Table 8. Overall effect of treatment on markers of bone metabolism in control (CO) and
SZA-supplemented (SS) cows ............................................................ 54

Table 9. Mineral concentrations (ug/ g) of milk replacer and SZA fed to calves ........... 64

Table 10. Body weights and average daily gains (ADG) of control (CO) and SZA-
supplemented (SS) calves ................................................................ 74

Table 11. Digestibility variables in control (CO) and SZA-supplemented (SS) calves. . .75

Table 12. Length, diameters, and cortical widths of MC III&IV from control (CO) and
SZA-supplemented (SS) calves .......................................................... 76

Table 13. Bone mechanical properties of MC III&IV from control (CO) and SZA-
supplemented (SS) calves ................................................................ 77

Table 14. Fecal and urine mineral concentrations in control (CO) and SZA-supplemented
calves ........................................................................................ 78

Table 15. Metabolism of minerals in control (CO) and SZA-supplemented (SS)
calves ........................................................................................ 79

Table 16. Plasma mineral concentrations (pg/ml) in control (CO) and SZA-supplemented
(SS) calves ................................................................................. 80

Table 17. Percentage of body weight (%) of organs from control (CO) and SZA-
supplemented (SS) calves ................................................................ 81

Table 18. Organ and feces dry matter (%) from control (CO) and SZA-supplemented (SS)
calves ........................................................................................ 81

Table 19. Mineral concentrations (ug/ g dry weight) of various tissues from control (CO)

and SZA-supplemented (SS) calves ..................................................... 82
Table 19 (cont’d) .................................................................................... 83
Table 19 (cont’d) .................................................................................... 84
Table 19 (cont’d) .................................................................................... 85
APPENDIX TABLES
Table 1A. Mineral concentrations (mg/kg dry wt) of the diet ............................. 108

Table 2A. Plasma, fecal and urine concentrations in control (CO) and SZA-supplemented
(SS) horses .............................................................................. 110

Table 3A. Metabolism of minerals in control (CO) and SZA-supplemented (SS)
horses ..................................................................................... l 11

Table 1B. Differences in bone dimensions and mechanical properties between two highly
active calves and the other calves .................................................... 120

Table ZB. Comparisons of osteocalcin and deoxypyridinoline between two highly active
calves and the other calves ............................................................ 121

xi

LIST OF FIGURES

Figure l. Lesion length (cm) in control (CO) and silicon—supplemented (SS) horses pre-
and post-supplementation of placebo or silicon ....................................... 40

Figure 2. Lesion height (cm) in control (CO) and silicon-supplemented (SS) horses pre-
and post-supplementation of placebo or silicon ....................................... 41

Figure 3. Lesion area (cmz) in control (CO) and silicon-supplemented (SS) horses pre-
and post-supplementation of placebo or silicon ....................................... 42

xii

CHAPTER 1

INTRODUCTION

Supplemental silicon, primarily in the form of sodium zeolite A (SZA), has been
gaining popularity in the equine industry as a supplement for bone health. The interest
arises predominantly from a study conducted at Texas A&M University (Nielsen et al.,
1993). Researchers found SZA to be associated with both decreased skeletal injuries and
increased training distance until injury. Moreover, plasma silicon concentrations were
correlated with training distance before an injury occurred. However, the mechanism by
which SZA was associated with the decreased injuries is unknown.

Trainers feeding silicon to their horses have reported a decrease in
osteochondrosis in the joints of their horses. Although anecdotal, the underlying results
of osteochondrosis - disturbed endochondral ossification and glycosaminoglycan loss —
may be remedied by supplementation of silicon. Chicks supplemented with silicon had
increased cartilage glycosaminoglycan concentrations (Carlisle, 1976), hastened bone
mineralization (Carlisle, 1982) and improved endochondral bone growth (Carlisle, 1980)
compared to controls on a silicon-deficient diet. The incidence of osteochondrosis in the
horse population, specifically in Standardbreds, is quite high. Although surgery is a
viable option, the inherent risks and costs of surgery mean it would be beneficial to have
a non-invasive method that reduces osteochondrosis in horses.

Sodium zeolite A is being used in the feed industry as an anti-caking supplement,
with allowable levels up to 2% SZA. However, there are many questions surrounding the

feeding of SZA. Studies using SZA have found conflicting results in feed intake,

digestibility, and mineral metabolism, both within and between species. For example,
feeding 2% SZA to horses has not been reported to affect feed intake (Frey et al., 1992;
Nielsen et al., 1993; Lang et al., 2001b,c), but cows fed the same level reduced their feed
intake (Johnson et al., 1988; Thilsing—Hansen et al., 2002). Alternatively, less than 2%
SZA to cows increased feed intake (Roussel et al., 1992). Research needs to address this
issue. Additionally, most studies have overlooked the effects of aluminum. As SZA is
an aluminosilicate that breaks down in the digestive tract to aluminum and orthosilicic
acid (bioavailable silicon), aluminum cannot be ignored as is can cause toxicities in
animals and humans (Goodman, 1986; Nayak, 2002). The ramifications of feeding a
supplement high in aluminum should be investigated, particularly in food producing
animals. Additionally, beneficial effects have been seen when supplementing silicon. By
identifying tissues that accumulate silicon following supplementation, future research can

be planned accordingly.

The primary objectives of this project were to:

1. determine if supplementation of SZA decreases osteochondrotic
lesion size in Standardbreds.

2. determine if supplementation of SZA to lactating dairy cows
would affect plasma mineral metabolism as well as bone
metabolism.

3. attempt to elucidate a mechanism by which SZA was associated
with decreased skeletal injuries in racing horses by evaluating

bone mechanical properties, turnover, and architecture.

evaluate the tissue mineral composition following SZA
supplementation to determine the extent of tissue mineral

accumulation in young calves.

CHAPTER 2

REVIEW OF LITERATURE

Silicon

Silicon (Si) is the second most abundant mineral found in the earth’s crust
(Carlisle, 1984). As silicon is not found free in nature, it exists mainly as silicates and the
oxide, silica. Silicon is relatively inert, but can be reactive to halogens and dilute alkali.
Most acids do not affect it. Silicon has three stable isotopes — 2E’Si, 29Si, and 3(Si, and two
radioisotopes that can be used as tracers — 3 1Si and 32Si (Carlisle, 1984). Silicon is found
throughout the environment, particularly in glass and sand. Forages and grains also
contain silicon.

Silicon is found throughout the mammalian body in various tissues. Early data on
the silicon content of mammalian tissues have reported variable results. This could be
due to the use of glass in experiments that would result in leaching of silicon from glass
Lang et al., 2001c).

Silicon is present in the blood as free soluble monosilicic acid, and is found in
similar concentrations (~ 100 ug/dl) in monkey, rat, and human whole blood (Carlisle,
1984). Silicon appears to be freely diffusible throughout tissue fluids as other examined
body fluids had similar silicon concentrations as the blood serum (Carlisle, 1984). Blood
silicon levels remain fairly constant except in instances in which additional silicon is
added to the diet (Carlisle, 1984). Supplemental silicon increases plasma silicon
concentrations in horses (Frey et al., 1992; Nielsen et al., 1993; Lang et al., 2001b,c;

Mazzella et al., 2005).

Various tissues throughout the body, particularly the connective tissues such as
the aorta, trachea, tendon, bone, and skin are unusually rich in silicon (Carlisle, 1972).
Carlisle (1972) suggested that the high silicon content of connective tissue is mainly from
its presence as an integral component of the glycosaminoglycans and their protein
complexes that contribute to the structural framework of the tissues. Furthermore, these
tissues show a decline in silicon concentration as the animal ages. While the heart, liver,
and muscle silicon concentrations remained unchanged over several years, concentrations
in the aorta and skin decreased significantly (Carlisle, 1974). Moreover, a decrease in
total glycosaminoglycan synthesis seems to occurs in human skin (Vuillermoz et al.,
2005), lending some support to the hypothesis that the majority of silicon in connective

tissue is associated with glycosaminoglycans.

Rat kidney, liver, bone, skin, spleen and lung were found to accumulate the
greatest amounts of 3"-labeled silicic acid one hour afier intracardiac injection of the

labeled element

Few studies have investigated tissue silicon accumulation following silicon
administration. One such study found that silicon was increased in several rat tissues
post-intracardiac injection of 31-labeled silicic acid (Adler et al., 1986). Initially, the
skin, bone, and muscle contained the largest amounts of silicon, with 85 % of the total
3 1Si in those three organs. However, as time progressed, the 31Si concentration of the
three organs decreased as the concentrations in the other organs did not change or
increased. After four hours, 56% of the total 31Si was found in the bone, muscle and skin.
The decreases in those tissues and the subsequent increases in the other tissues could be a

result of equilibrium. The same study also reported that the 31Si was found almost

entirely in a nonprotein bound form in the plasma (Adler et al., 1986). Additionally, the
blood-brain barrier excluded 31Si from the brain as it was present only in negligible
concentrations. This appears to be the only study that investigated the effects of
supplemental silicon on tissue uptake. Silicon supplemented through dietary means, and
not via injection into circulatory system, might have different effects. Also, a livestock
species such as the horse or cow, might have dissimilar tissue responses to supplemental
silicon as the tissues might have different concentrations than the rat due to the livestock
diet of grains, forage, and oftentimes, sand. The effects of long-term supplementation of
silicon are unknown as well.

Reported silicon concentrations in cow milk have been conflicting.
Concentrations have been reported to be 0.81 ppm in normal cows (Parantainen et al.,
1987), 0.39 ppm in mastitic cows (Parantainen et al., 1987), or absent entirely (Anderson,
1989). Differences may come from the use of glass labware, processing procedures, or
simply cow differences due to diet or environment. Supplemental silicon, in the form of
sodium silicate (N a2Si03) and at a dosage rate of 1 gm/d, did not consistently or
significantly increase the silicon content of milk in Holstein, Ayrshire, and Guernsey
cows (Archibald and Fenner, 195 7). However, the form of silicon may not be as
bioavailable as other forms, such as sodium zeolite A, and the low dosage rate may not be
sufficient to cause an increase in milk concentration. Finally, the study was conducted in
only six cows, two per breed, which may explain the variability in results.

Unlike cows, an increase in milk silicon concentration has been reported in mares
(Lang et al., 2001b). Within 24 hr of foaling, mares were supplemented with sodium

zeolite A at a dosage rate of 2% dietary intake, for 45 days. Milk silicon concentration

was higher (P < 0.01) on d 45 than (I 0 in treated mares, and on d 45, milk silicon
concentration was higher (P < 0.001) in treated mares compared to control mares (Lang
et al., 2001b). Additionally, plasma silicon concentration in treated mares’ foals was
higher (P < 0.05) on d 45 than (1 0 and plasma silicon concentration on d 45 was higher
(P < 0.01) in treated mares’ foals compared to control mares’ foals. This study suggests
that supplemental silicon can increase silicon concentrations in milk, and that the milk
silicon can be consumed and absorbed. Therefore, studies need to be conducted with a
bioavailable source of silicon, and at a higher dose than in a previous study, to determine

if milk silicon concentration can be increased in supplemented cows.

Silicon essentiality

Silicon has been suggested to be an essential element for several biological
processes including bone formation and mineralization, glycosaminoglycan synthesis and
composition, collagen formation, and brain function. As a result of the research
implicating silicon as an essential element, the American Institute of Nutrition added
silicon to the purified rodent diet (Reeves et al., 1993). Numerous studies have been
conducted in rats and chicks, but no studies on silicon essentiality in other animals have
been found. Additionally, the studies were conducted primarily by one research group
(EM. Carlisle) using extreme conditions. Essentiality was determined by the finding of
adverse effects when silicon was deficient in the diet, as compared to when adequate to
high levels of silicon were supplied in the diet. To obtain the deficient diets, extreme
measures such as washing ingredients like casein to remove silicon (Carlisle, 1980a),

were taken so that the highly purified diets induced responses to the low intakes.

Furthermore, the crystalline amino acid diets used in studies by Carlisle prior to 1980
were admittedly inadequate diets that produced less than optimum grth even in the
control animals (Carlisle, 1981). Silicon deficiencies have not been reported under
normal management circumstances, and requirements for humans and animals have not
been established (Nielsen, 1991). However, the requirements for silicon are probably
fairly low as diets used to induce adverse effects typically must be below 2 mg Si/kg diet
(Nielsen and Poellet, 2004). Rats fed 4.5 mg Si/kg diet did not differ from rats fed 35 mg
Si/kg diet, and both prevented the deprivation signs exhibited by rats fed 1 mg Si/kg diet
(Seaborn and Nielsen, 1993). Additionally, essential elements are normally found in
adequate quantities in milk, so that young animals may receive nutrition to enhance
proper growth. Silicon has been found to be absent (Anderson, 1989) or present in low
quantities (0.81 mg Si/kg milk) (Parantainen et al., 1987) in cow milk. Horse milk also
contained low quantities of silicon (Lang et al., 2001b). There is some support to the
suggestion that silicon is an essential element; however, a more appropriate term might

be “probably essential”, as suggested by Georgievskii (1982).

Bone formation and mineralization. Some evidence exists that suggests silicon
may be an essential element for bone formation and mineralization. Silicon deprivation
in chicks caused depressed long bone skeletal development, and also caused the skulls to
be abnormally shaped and smaller (Carlisle, 1972). Dietary silicon deficiencies in rats
caused depressed grth and skull abnormalities (Schwarz and Milne, 1972). .
However, the diets used in the previous studies were inadequate to maintain optimum

growth in the control group as well. Subsequent studies using a more appropriate diet

seem to support the earlier postulates that silicon is essential for bone formation. A
study conducted in chicks revealed a difference between chicks consuming a silicon-
supplemented diet and those consuming a silicon-deficient diet (Carlisle, 1980a). The
matrix of the deficient chicks’ skulls lacked the normal striated trabecular pattern. The
deficient chicks also showed a nodular pattern of bone arrangement, indicative of a
primitive type of bone. Additionally, collagen content was reduced in the deficient
chicks. Silicon deficient chicks also had gross abnormalities in skull architecture, with
fewer trabeculae and less calcification (Carlisle, 1981).

A study conducted in vitro showed that silicon was present in active growing
areas of bones in young rats and mice (Carlisle, 1970). In the process of mineralization,
initially silicon and calcium concentrations rose congruently in osteoid tissue. In the
more advanced stages of mineralization, silicon concentrations fell markedly while
calcium concentrations approached proportions found in mature bone (Carlisle, 1970).
Silicon appears to be involved in calcification through some effect on preosseous matrix.
In vivo studies show that silicon hastens the rate of bone mineralization. The tibia of rats
fed a high (250 ppm) silicon diet reached a higher degree of mineralization in a shorter
time than the tibia from a medium (25 ppm) and low (10 ppm) silicon diet (Carlisle,

1970).

Glycosaminoglycan synthesis and structure. Formerly called
mucopolysaccharides, glycosaminoglycans (GAGs) are isolated from vertebrate
connective tissue. The structures of the different families of GAGs are similar, as each

are hydrophilic, unbranched single chain polymers with repeating disaccharide units

(Lovekamp et al., 2005). Glycosaminoglycans, with the exception of hyaluronic acid,
contain sulphate groups (Fraser et al., 1997). They are synthesized in the endoplasmic
reticulum and Golgi bodies. Hyaluronic acid is the major GAG of synovial fluid (Scott et
al., 2000). Unlike other GAGs, hyaluronic acid’s primary structure does not contain a
protein portion, a reflection of plasma membrane synthesis origin, and not Golgi body.
Hyaluronic acid gives synovial fluid its viscosity, and is synthesized by cells in the
synovial membrane. It appears to have several functions, which include lubrication,
water homeostasis, filtering effects, and plasma protein distribution. Common thought
has been that hyaluronic acid itself was the active ingredient in boundary lubrication of
the joint, allowing the joint surfaces to reduce friction, thereby preventing wear.
However, hyaluronic acid is not necessarily the lubricant of the joint, but rather the
carrier of the lubricant to the binding site to facilitate boundary lubrication, as lubrication
was not changed when hyaluronidase was applied to synovial fluid, even though viscosity
was decreased (Hills and Monds, 1998). Furthermore, hyaluronic acid acts as a wetting
agent, reducing the surface energy of the interface between synovial fluid and the
hydrophobic articular surface. This could have an important role in promoting
hydrodynamic lubrication, which is favored where load permits because it involves no
contact on the sliding surfaces and, therefore, no wear (Hills and Monds, 1998).
Hyaluronic acid concentration in the synovial fluid varies with species, with higher
concentrations found in rabbits (3.6 mg/ml) (Price et al., 1996) than adult horses
(approximately 0.5 mg/ml) (Persson, 1971). Bovine metatarso-phalangeal synovial fluid

also contains approximately 0.5 mg/ml hyaluronic acid (Coleman et al., 1999). Age of

10

the subjects appears to play a role in hyaluronic acid concentration as the young human
concentration of 3.8 mg/ml falls to 2.5 mg/ml in the elderly population (Balazs, 1982).
Results from several studies suggest that silicon may be important in
glycosaminoglycan synthesis and composition. The total amount of hexosamines and the
proportion of hexosamine in the comb were found to be greater in combs of chicks
supplemented with silicon, as compared to combs of chicks on a silicon deficient diet
(Carlisle, 1976). Additionally, articular cartilage from chicks supplemented with silicon
contained greater amounts of GAG than cartilage from silicon deficient chicks (Carlisle,
1974). The same supplemented chicks also had greater water content in their tibial
bones, which the author concluded was due to an increased GAG concentration as GAGs
attract and bind water (Carlisle, 1976). I
Fractionation procedures reveal that connective tissues yield high amounts of
silicon (Carlisle, 1984). No less than 330 to 554 ppm bound silicon has been detected in
chondroitin 4-sulfate, heparin sulfate and hyaluronic acid from umbilical cord (Schwarz,
1973). This could correspond to one atom of silicon to about 0.3 molecules of hyaluronic
acid. However, hyaluronic acid from bovine vitreous humor contained negligible
amounts of bound silicon. The silicon found in the GAGS exists not as free silicate ions
or silicic acid, but as a firmly bound component of the polysaccharide matrix. Therefore
the bound silicon could function in several ways. It could modulate molecular shape and
establish a secondary structure by connecting different portions of the same chain
(Schwarz, 1973). Instead of connecting the same chain, it could connect different
polysaccharide chains together, thereby contributing to the high molecular size of GAGs

(Schwarz, 1973). Additionally, silicon could connect polysaccharides and collagen

ll

(Schwarz, 1973). Silicon could link the binding sites through a R; — O — Si — O — R2
group. Thus silicon could aid in the architecture and stability of GAGS. As the Schwarz
(1973) study appears to be the only study pinpointing silicon as an actual component of
GAGS, a more current study utilizing current technologies should be conducted to
confirm the findings.

Osteochondrosis. Supplemental silicon might impact osteochondrosis.

 

Osteochondrosis (0C) is defined as a disturbance of cell differentiation in joint cartilage
leading to altered endochondral ossification (Jeffcott, 1991). No specific cause of DC
has been identified yet, but there are multiple factors that have been pinpointed including
growth rates, nutrition, endocrinological factors, genetics, and trauma (J effcott, 1991).
Regardless of the factor(s), the result remains the same — disturbed endochondral
ossification. Key cellular events in the normal developing epiphysis include chondrocyte
proliferation, extensive extracellular matrix creation, chondrocyte differentiation,
vascular invasion and matrix calcification. In OC, vascular penetration of the distal
region of the proliferative zone fails, which leads to a failure of the final stages of
cartilage maturation and modification of the surrounding matrix, resulting in the
accumulation of small rounded chondrocytes apparently trapped in the predifferentiation
stage within the cartilage (Jeffcott and Henson, 1998). The early lesion of OC develops
as a small retained core of cartilage extending into the subchondral bone.

Cartilage extracellular matrix alterations have been identified in naturally
occurring DC in horses (Lillich et al., 1998). These included a loss of
glycosaminoglycans, including chondroitin sulfate, in cartilage from DC lesions when

compared to normal (Lillich et al., 1998). Proteoglycan production from chondrocytes

12

was decreased in explant cultures of cartilage from equine OC lesions (van den Hoogen
et al., 1999). The authors hypothesized that the chondrocytes had lowered metabolism
and decreased vitality, and that spontaneous regression of the lesions could not occur at
that point. The disturbed proteoglycan synthesis was more likely to be a consequence

than a primary cause of the disturbed ossification (van den Hoogen et al., 1999).

Silicon toxicity

The main problems with silicon toxicity occur in the lungs and kidneys.
Inhalation of dust and silicates can cause silicosis, which stimulates a severe reaction in
the lungs (Carlisle, 1984; Rimal et al., 2005). A subsequent increase in immunoglobulin
production and lymphocyte production occurs (Moseley et al., 1988). After phagocytosis
of the silica particle by the macrophages, the macrophages are killed, reingested by other
macrophages, and the cycle continues. Collagen synthesis is then stimulated by the
macrophage death (Carlisle, 1984). Additionally, inflammatory cytokines such as TNF-
alpha are increased, causing damage to the lungs (Rimal et al., 2005).

In some instances, silica is deposited in the urinary tract, forming calculi (Carlisle,
1984). Eventually the calculi may become large enough that urine excretion is blocked,
which could lead to death. There is little understanding of the formation of calculi, as
efforts to create them in sheep and cattle, such as increasing silicon intake, have been
unsuccessful. However, calculi were created in rats and dogs fed high levels of a silicon

supplement, not silica (Carlisle, 1984).

13

Supplemental bioavailable silicon

Although abundant in the environment, silicon is usually not readily absorbed by
animals. The naturally occurring form is silica (8102) (Carlisle, 1982) which is present in
grasses and cereal grains. However, silica is considered to be largely insoluble, and thus
relatively unavailable. Silica would need to be broken down into soluble parts.
Orthosilicic acid (Si(OH)4) is formed by hydration of silica. Orthosilicic acid is the
major form of silicon in drinking water and other liquid sources, including beer (Reffitt et
al., 1999). Studies suggest that orthosilicic acid might be the only detectable, absorbable
form of silicon (Jugdaohsingh et al., 2000) and has an approximate absorption rate of
53% in humans. Therefore, studies interested in supplementing silicon should use the
bioavailable form, orthosilicic acid.

A study conducted on 60 calves demonstrated the high bioavailability of
orthosilicic acid (Calomme and Vanden Berghe, 1997). The treated group received 377.5
mg of silicon, including basal diet silicon content and supplemental orthosilicic acid,
while the control group received 360 mg of silicon through their basal diet. With just a
4.9% increase in dietary silicon intake from supplemental orthosilicic acid, the treated
group had a 70% higher serum silicon concentration than control calves after 23 wks.
Additionally, the treated calves had higher levels of collagen in the skin, indicating that
not only was there increased absorption as evidenced by increased serum silicon, but that
the silicon was able to illicit changes in extracellular matrix formation, suggesting
bioavailability.

Increasing dietary silicon may have beneficial effects on bone health, as suggested

by in vitro and in vivo work. Orthosilicic acid (absorbable form of silicon) at

14

physiological concentrations stimulates in vitro collagen type 1 synthesis in human
osteoblast-like cells and enhances osteoblastic differentiation (Reffitt et al., 2003).
Osteocalcin was used to determine osteoblastic differentiation, and was found to be
increased by both 10 uM and 20 uM orthosilicic acid. Gene expression of osteocalcin
was also increased in the presence of orthosilicic acid. Furthermore, alkaline
phosphatase, a marker of bone formation, was increased by supplementation of
orthosilicic acid (Reffitt et al., 2003). The precise method by which silicon enhances
collagen type I synthesis is unknown.

The population sector that may reap the most benefits of silicon supplementation
is that of post-menopausal women as well as people afflicted with osteoporosis. Silicon
appears to alleviate the loss of bone mineral content or mass, common symptoms of
osteoporosis and ones that normally accompany menopause. Rats that underwent an
ovariectomy, and thereby experienced menopausal conditions such as a decrease in
estrogen, did not show a loss of bone mass when supplemented with silicon, as compared
to sham-operated controls (Rico et al., 2000). Not only did silicon appear to inhibit bone
loss, it also stimulated bone formation in the ovariectomized rats. Silicon has also had
positive effects in humans as well. Silicon supplementation in the form of monomethyl
trisilanol at a dosage rate of 50 mg twice a week for a period of 4 months was associated
with increased femoral bone density in post-menopausal, osteoporotic women (Eisinger
and Clairet, 1993). Silicon may have a therapeutic effect on osteoporosis by the

prevention of bone loss and stimulation of bone formation.

15

Sodium zeolite A. Sodium zeolite A (SZA) is an aluminosilicate that is
hydrolyzed at low pH into silicic acid, amorphous aluminum silicates and aluminum
(Thilsing-Hansen et al., 2002). Thus, by supplying SZA to the diet, an animal’s
gastrointestinal tract can break down the zeolite into orthosilicic acid. The orthosilicic
acid is absorbed, thus providing the body with supplemental silicon. The extent of
absorption of Al released from the zeolite is unknown. Zeolites are crystalline structures
that have a high attraction for water and a large number of positively charged ions, such
as K+, NH4+, Ca2+, and Mg”, which can be reversibly bound or released (Mumpton and
Fishman, 1977). Multiple studies have been conducted in mammalian species with
sodium zeolite A supplementation (Frey et al., 1992; Roussel et al., 1992; Nielsen et al.,
1993; Lang et al., 2001b,c; Thilsing-Hansen et al., 2003).

An in vitro study conducted with avian osteoclasts concluded that sodium zeolite
A can inhibit bone resorption (Schutze et al., 1995). When osteoclasts were treated with
100 ug/ml of SZA, the number of pits per osteoclast was reduced 3-fold at 24 hours after
treatment. Thus, osteoclast-mediated resorption activity was reduced. Additionally,
secreted cathepsin B enzyme activity was reduced 3-fold. The conclusion was that the
actual structure of SZA is responsible for the effects as compounds used to mimic
constituents of SZA failed to illicit the same responses (Schutze et al., 1995). Thus,
inferences on the effect of SZA on bone resorption in vivo are limited, as the compounds
in SZA are separated in the digestive tract, and therefore may be ineffective.

Studies conducted in horses produced inconclusive results on bone metabolism.
Pyridinoline crosslinks (PYD) and carboxy—terminal pyridinoline cross-linked telopeptide

region of type-I collagen (ICTP), markers of bone resorption, were measured in mares

l6

from l-d post-parturition to one month post-parturition (Lang et al., 2001b). Sodium
zeolite A tended to decrease PYD in supplemented mares on d 30, but by d 45 the
differences were gone. However, ICTP was not affected. Pyridinoline is not a bone
specific marker, and can be affected by collagen breakdown in other tissues, such as the
uterus (Christenson, 1997). The more specific bone marker, ICTP, was unaffected,
implicating that SZA has little to no effect on bone resorption in broodmares.
Alternatively, PYD was not affected in SZA supplemented yearlings, while ICTP was
decreased on d 45 (Lang et al., 2001c). Not only are bone resorption results inconclusive,
but bone formation results as well. Osteocalcin, a marker of bone formation, was not
different between control and SZA supplemented yearlings (Lang et al., 2001c).
However, supplemented broodmares tended to have higher concentrations of osteocalcin
compared to control horses (Lang et al., 2001b). The effects of SZA on bone metabolism
in vivo are uncertain, and should be clarified. Sodium zeolite A has been associated with
reduced skeletal injuries in race horses. F ifi‘y-three race horses entering race training at
18 months of age were divided into four groups at 6 mo of age: 0, 0.92, 1.86 or 2.00 %
dietary SZA (Nielsen et al., 1993). The medium and high treatment groups had greater
distances traveled before injury, and the medium group also had more cycles (strides)
until injury. Fewer injuries occurred in the treated groups as well (Nielsen et al., 1993).
Of thirteen control horses, 8 became injured during the study. Five horses (out of 13) in
the 0.92% SZA group became injured, two horses (out of 9) in the 1.86 SZA% group
were injured, and 4 (out of 12) horses in the 2.00% SZA became injured during the study.

Furthermore, plasma Si concentrations correlated with training distance to failure

17

(Nielsen et al., 1993). However, the mechanisms by which SZA is associated with fewer
injuries is unknown.

Sodium zeolite A has also been used in several ruminant studies, with marked
effects on mineral metabolism, particularly that of calcium and phosphorus. Decreases in
plasma phosphorus concentrations occurred in cows fed varying levels of SZA, from 0.5
to 1.0 kg daily (Thilsing-Hansen et al., 2003). Additionally, the incidence of parturient
hypocalcemia (milk fever) was decreased in cows fed SZA (Thilsing-Hansen and
Jorgensen, 2001; Thilsing-Hansen et al., 2003). Serum calcium concentrations were
higher in treated cows than control cows following SZA supplementation, thereby
preventing hypocalcemia. Cows also had increased serum calcium when fed 1.0 and
1.5% SZA of a diet consisting of 50% alfalfa hay as well as 1.5% SZA of a diet
containing 50% corn silage (Roussel et al., 1992). An increase in calcium absorption has
been shown in chickens fed 1.0% zeolite A for 7 d (Ballard and Edwards, 1988).

A single study investigating the effects of sodium zeolite A on tissue mineral
composition was found. Swine fed 0.50% SZA had increased liver and bone zinc, but no
accumulation of Ca, P, or Al in the bone (Ward et al., 1991). The pigs also had decreased
serum Ca and P concentrations. However, silicon concentrations were not determined.
As SZA appears to alter mineral metabolism, both in the blood and tissue, in—depth
studies should be conducted to determine the extent of alteration, particularly that of

silicon in tissues.

18

Silicon and aluminum interactions

There is a dearth of information pertaining to silicon and its interactions with
other minerals, with the exception of aluminum. There appears to be a strong interaction
between silicon and aluminum. Silicon may be the natural antidote to aluminum toxicity
(Reffitt et al., 1999). An inverse relationship exists between silicon and aluminum in
water supplies (Birchall and Chappell, 1989). Alzheimer’s disease appears to be
prevalent in areas with high concentrations of Al in drinking water supplies. The low
incidence of Alzheimer’s in areas with low aluminum concentration, and thus high
silicon concentration, may be due to the excess silicon forming hydroxyaluminosilicates,
and thus prevention of A1 uptake. Supplementation of aluminum increased brain Al
concentrations when rats were fed a low silicon diet, yet the increase was negated when
rats were fed a silicon supplemented diet (Carlisle and Curran, 1987). Silicon has been
hypothesized to be essential for collagen formation (Seaborn and Nielsen, 2002a,b), in
which silicon deprivation decreased the amount of hydroxyproline. Aluminum appears to
inhibit prolylhydroxylase activity; however, this inhibition is relieved when silicon is
present.

Orthosilicic acid reduces Al absorption in humans (Edwardson etal., 1993).
Silicate did not affect gastrointestinal uptake of 26A], but rather increased renal clearance
almost two-fold over the first 24-hr post supplementation in humans (King et al., 1997).
Bellia et al. (1996) found an increase in aluminum renal clearance in response to
monosilicic acid in beer, and suggested that silicon and aluminum may form
hydroxyaluminosilicates in the kidney, thereby decreasing aluminum reabsorption.

Alternatively, Reffitt et al. (1999) found no increase in urinary A1 in humans

19

supplemented with orthosilicic acid, and suggested that perhaps something in the beer
besides silicon in the Bellia et a1. (1996) study may be responsible for excretion of
aluminum from body stores. The form of silicon also plays a role on its effect on
aluminum metabolism. Monomeric silicon (orthosilicic acid) failed to prevent an
increase in serum aluminum following ingestion of an aluminum supplement, while
oligomeric silicon reduced serum aluminum by approximately 67% (J ugdaohsingh et al.,
2000). Moreover, the oligomeric silicon also reduced urinary excretion of Al. To lend
further evidence that oligomeric silicon interacts with aluminum in the gastrointestinal
tract and prevents aluminum absorption, oligomeric silicon itself was not absorbed
(Jugdaohsingh et al., 2000). They determined that 280 umol Si/L is needed to chelate 8
umol Al/L. Alternatively, monomeric silicon was easily absorbed and had no effect on
aluminum metabolism. The form of silicon needs to be taken into account when
designing studies to investigate the effects of silicon on aluminum metabolism.

The SizAl ratio may also play a role in aluminum toxicity. When the Si:Al ratio
dropped to 3.7 from 13.0 in a fish tank there was a significant increase in aluminum
levels in the fish (Birchall et al., 1989). Aluminum levels in the fish continued to rise as
the SizAI ratio dropped below 1.0. An excess of silicon over aluminum appears to be

necessary to reduce the effects of aluminum toxicity.

Aluminum and livestock
Excess exposure to aluminum is significantly associated with phosphate binding
in the digestive tract, phosphate deficiency, and interference with phosphate metabolism

(Krueger et al., 1985). The fact that high dietary aluminum interferes with phosphorus

20

metabolism in animals by forming unabsorbable complexes in the digestive tract has been
known for a long time (Deobald and Elvehjem, 1935). A significant decrease in plasma
phosphorus in response to supplementation of aluminum has been seen in lambs (Rosa et
al., 1982), beef cows (Allen et al., 1986) and chicks (Hussein et al., 1990). The decrease
can occur within 7 days post supplementation.

An increase in plasma calcium concentrations occurred when chicks were fed
high amounts of aluminum (0.196% of the diet) (Hussein et al., 1990). This effect was
seen as early as 7 days post supplementation. Increases in serum calcium concentrations
were also observed in lactating beef cows receiving 1,730 mg A1/kg DMI (Allen et al.,
1986). The increase in calcium may be as a result of homeostatic measures in response to
the decreased phosphorus. Decreases in calcium concentration in the bone have been
reported after high aluminum supplementation (Allen, 1984). When additional
phosphorus was added to the diets, the increase in plasma calcium was less dramatic
(Hussein et al., 1990). Changes in calcium and phosphorus metabolism have also been
noted in humans with high aluminum intakes, including hypophosphaturia,
hypophosphatemia, and hypercalciuria (Spencer et al., 1982). Hypercalciuria may be due
to increased calcium absorption. An increase in calcium retention and availability was
seen in non-lactating cows that were fed an additional 1000 mg Al/kg DMI (Robinson et
al., 1984).

Aluminum interacts not only with phosphorus and calcium, but magnesium as
well. Aluminum added to the rumen of steers decreased serum magnesium (Allen and
Robinson et al., 1980), as did dietary aluminum (Allen etal., 1984; Allen et al., 1986).

Chicks fed 0.392% A1 for 21 days also had decreased magnesium blood concentrations

21

(Hussein et al., 1990). The same response was seen in sheep fed 2,000 mg Al/kg feed for
56 d (Valdivia et al., 1982) and tended to be seen in lambs fed 1,450 mg Al/kg feed for
76 d (Rosa et al., 1982). However, dairy calves approximately 64-d-old given 0.20% Al
did not show a decrease in serum magnesium concentrations (N eathery et al., 1990a).
The mechanisms by which aluminum interacts with magnesium are not as clearly defined
as with phosphorus (Allen, 1984). Unlike with phosphorus, the effect of aluminum on
magnesium does not appear to be from the formation of unabsorbable complexes (Allen
and Fontenot, 1984). Although serum magnesium levels are depressed, there are no
changes in magnesium absorption (Valdivia et al., 1982; Allen and Fontenot, 1984).

Very few studies exist on supplying aluminum to horses. One study found that
there were no ill effects of feeding 1,370 mg Al/kg DMI but 4,500 mg Al/kg DMI
decreased plasma phosphorus concentration, and increased plasma and urinary calcium
concentration (Schryver et al., 1984). In the same study, the ponies on the high
aluminum diet were found to have increased plasma hydroxyproline, indicating higher
bone turnover perhaps due to altered calcium and phosphorus metabolism. Roose et a1.
(2000) found that feeding mature Thoroughbreds either 160 mg Al/kg DMI or 930 mg
Al/kg DMI produced no adverse effects on phosphorus or calcium metabolism.
Additionally, there were no changes in nutrient digestibilities such as dry matter, crude
protein, fiber and mineral digestibilities.

There appears to be a threshold at which Al has negative effects on feed intake.
Lambs fed 810 mg Al/kg DMI maintained feed intake (Thompson et al., 1959) as did
steers fed 1,200 mg Al/kg DMI (Valdivia etal., 1978). Supplementation of 1,450 mg

Al/kg DMI (Rosa et al., 1982) and 2,000 mg Al/kg DMI (V aldivia et al., 1982) decreased

22

feed intake in lambs. Young Holstein calves reduced their feed intake by 17% and had a
47% decrease in body weight gain after consuming a diet consisting of 0.20 % A1 (Crowe
et al., 1990). The decreases in feed intake may be a result of the binding of aluminum to
phosphorus in the digestive system, essentially reducing dietary phosphorus to the
animal. The animals then begin to exhibit classic signs of phosphorus deficiency, the
most common deficiency in ruminants. These signs include decreases in both feed intake
and milk production (Call et al., 1987), and unchanged concentrations of minerals in the
milk (Underwood, 1981).

The negative effects of high aluminum intakes can be alleviated by supplementing
dietary phosphorus. Increases in previously decreased plasma inorganic phosphorus
concentration has been observed when phosphorus was supplied to the animals (Lipstein
and Hurwitz, 1982; Hussein et al., 1990). Decreases in gain and feed intake in Al—
supplemented lambs were reversed when the animals were fed additional phosphorus
(Valdivia etal., 1977; Rosa etal., 1982). Lipstein and Hurwitz (1982) suggest that 0.76 g
of supplemental phosphorus is required to offset the effect of 1 g of aluminum. Animals
that are borderline phosphorus deficient would seem to be most at risk of the negative
aluminum effects on phosphorus metabolism.

While the addition of dietary aluminum caused a decrease in plasma phosphorus
concentration, and an increase in plasma calcium concentration, plasma aluminum
concentration remained unchanged (Hussein et al., 1990). This might be a reflection on
the thought that dietary aluminum is poorly absorbed and has “little effect on animal life”
(McCollum et al., 1982). King et a1. (1997) found that apparent gastrointestinal uptake of

aluminum is between 01-03%, which was probably slightly underestimated. As plasma

23

aluminum levels are not indicative of tissue storage status (Krueger et al., 1985), it might
be possible that plasma aluminum concentrations do not change because the aluminum is
rapidly transported to the tissues. Several studies have investigated the effects of high
aluminum on tissue mineral composition. Rosa et a1. (1982) used 1,450 mg Al/kg DMI
in lambs while Valdivia et a1. (1982) used a slightly higher level of 2,000 mg Al/kg DMI,
also in lambs. For ease of reporting results, Table 1 displays information on the organ
responses to high aluminum intakes. Valdivia et a1. (1982) found that at least a portion of
dietary aluminum is absorbed and transported as there were increases in the liver, kidney
and muscle. Therefore the opinion that dietary aluminum in essentially non-absorbable

should be re-evaluated.

Table 1. Responses of tissue mineral composition to high dietary
aluminum (P < 0.10)

 

 

Organ Effect Reference
Liver Increased Al Valdivia et al., 1982
Increased Zn Valdivia et al., 1978

Decreased Mn Valdivia et al., 1982
Kidney Decreased Mg Rosa et al., 1982; Valdivia et al., 1982

Decreased P Rosa et al., 1982
Increased Al Valdivia et al., 1982
Increased Cu Rosa et al., 1982; Valdivia et al., 1982
Increased Zn Valdivia et al., 1978; Rosa et al., 1982
Muscle Increased A1 Valdivia et al., 1982
Increased Ca Rosa et al., 1982
Increased Cu Rosa et al., 1982; Valdivia et al., 1982
Increased Mg Rosa et al., 1982
Bone Decreased P Rosa et al., 1982
Decreased Mg Rosa et al., 1982
Increased Mn Valdivia et al., 1982

 

24

Aluminum toxicity

Aluminum is common in soils, and tends to cause problems with excess, rather
than deficiency. Besides changes in feed intake and phosphorus absorption, high
aluminum intakes can be toxic to both animals and humans. In livestock, the most
common results of aluminum toxicity are phosphorus deficiency and grass tetany.
Aluminum has been linked to grass tetany in cattle, a metabolic disorder characterized by
hypomagnesaemia (Robinson et al., 1984). High intakes of aluminum depress serum
magnesium levels (Rosa et al., 1982; Valdivia etal., 1982; Hussein et al., 1990).

Patients with chronic renal failure are often dosed with high levels of aluminum to
reduce hyperphosphatemia (Krueger etal., 1985). However, oftentimes the patients
develop vitamin D- and vitamin D metabolite-resistant osteomalacia, which is thought to
develop from the aluminum (Krueger et al., 1985). Aluminum accumulates in the
calcification portion of the bone, preventing calcium from entering the bone (Boyce et al.,
1982). Subsequently, osteomalacia develops, and hypercalcemia occurs. Decreased bone
remodeling, with decreased numbers of osteoclasts and osteoblasts, and high plasma and
bone aluminum concentrations are signs of osteomalacia (Chazan et al., 1991). Besides
osteomalacia, high aluminum intakes can cause adynamic bone disease, in which bone
turnover ceases (Jeffery et al., 1996). Renal patients are not the only ones subject to bone
problems following aluminum administration. Rats developed osteomalacia afier
accumulating aluminum in the bone (Goodman, 1986).

Chronic or high intakes of aluminum can cause anemia, mainly through the
reduction of hemoglobin synthesis (Abreo et al., 1990) via iron deficiency (Jeffery et al.,

1996). Rats supplemented with aluminum had decreased hematocrit, hemoglobin and

25

plasma iron concentrations (Guo et al., 2004). Aluminum likely interferes with iron
metabolism. Aluminum can reduce intestinal iron uptake (Cannata et al., 1991).
Additionally, the reduction of Fe3+ to Fey, mediated by ceruloplasmin, is important for
translocation of iron and for the release of iron from ferririn. Ceruloplasmin is inhibited
by aluminum (Jeffery et al., 1996). However, the exact mechanisms by which aluminum
interferes with iron metabolism are unclear, but it is likely through competition of
binding sites.

Aluminum can bring about neurotoxic effects as well. In humans, aluminum
toxicity has been associated with memory loss, tremor, jerking movements, impaired
coordination, sluggish motor movement, loss of curiosity, and ataxia (Nayak, 2002).

Rats fed 0.3% aluminum in their drinking water displayed specific cognitive impairment
(J ope and Johnson, 1992). However, locomotor skills were not impaired. Aluminum has
also been linked to Alzheimer’s disease (AD) (N ayak, 2002). Elevated levels of
aluminum have been found in the brains of AD patients (Yoshimasu et al., 1981), and AD
incidence has been linked to aluminum in drinking water (Martyn et al., 1989). However,
a more recent study concluded that aluminum in drinking water is not related to AD
(Wettstein et al., 1991). Additionally, the paired helical filaments characteristic of AD
were not found in dementia patients with elevated brain aluminum levels (Arieff, 1990).
The link between aluminum and AD remains unclear, but ingestion of additional
aluminum should be avoided just in case.

Aluminum can be toxic to the cardiovascular system as well as the hepatobiliary

system, as both the heart and liver can accumulate aluminum (N ayak, 2002). Aluminum

26

is associated with arrhythmia and sudden death in hemodialysis patients (Birchall, 1991).

Hepatocytes can be destroyed by large deposits of aluminum (N ayak, 2002).

Analyses

Silicon. Analysis of samples for silicon concentration is often regarded as being
one of the most difficult procedures in trace mineral research for several reasons. The
first is the presence of silicon in the environment. As the second most abundant element
in the earth’s crust, silicon is found almost everywhere. In the laboratory, contamination
of samples by silicon-containing dust is a common problem. An even larger problem is
that of silicon leaching from glassware into samples. This is a problem not only in the
laboratory but also in the field, as blood collection into standard glass Vacutainers is
discouraged due to contamination.

A second potential problem in silicon analysis is the ability to digest the samples
so that total silicon is obtained, as silicon is found in tightly bound forms. It is resistant
to most digestion substances, as it has a low solubility in acid media (Schwarz, 1973).
Currently, there are two accepted methods for silicon digestion. One requires the use of
hydrofluoric (HF) acid, the other tetra methyl ammonium hydroxide (TMAH)
(Hauptkom etal., 2001). The advantage of using the HF method is that it is strong
enough to break silicon bonds, thereby allowing total silicon recovery. However, the HF
is so strong it will dissolve glass (hence the reason for breaking silicon bonds), which
could potentially harm analysis instruments. Therefore, a masking agent such as boric
acid is needed to prevent the etching of glassware. In addition, the hazards to humans

that come into contact with the HF are great. Lastly, HF mixtures are highly volatile,

27

thus care is needed to prevent silicon loss into the environment. Therefore, digestion
with an alkaline mixture such as TMAH is beneficial, as losses to volatization are
minimized and a masking agent is not needed (Hauptkom et al., 2001). Additionally,
biological samples need to undergo microwave digestion and the way silicon is
embedded in the tissue sample dictates the amount of heat needed. For example, ascorbic
acid and pork liver samples gave consistent results after heating to 60° C, while bovine
liver and muscle needed to be heated to 80 and 125° C, respectively (Hauptkom et al.,

2001).

GAG concentration. In the past, glycosaminoglycan concentration of cartilage
was determined in vitro using biochemical and histological analyses. Recently a method
was developed that utilizes magnetic resonance imaging (MRI) to determine cartilage
GAG content using the charged contrast agent gadolinium diethylene triamine
pentaacetic acid (Gd(DTPA)2') (Burstein and Gray, 2003). This method, delayed
gadolinium-enhanced magnetic resonance imaging of cartilage (dGEMRIC), is a
sensitive and specific measure of GAG content that can be used for both in vivo and in
vitro analyses. The dGEMRIC method works on the principle that cartilage GAGS have
negatively charged carboxyl and sulfate groups that create a negatively fixed charged
density to the cartilage matrix. Thus, ions that are distributed throughout the cartilage
will respond to the local fixed charge density of the surrounding cartilage area. The
Gd(DTPA)2' has a negative fixed charge as well. Thus, when the contrast agent is
distributed throughout the cartilage it is repelled by the negatively charged GAGS

(Burstein and Gray, 2003). Hence, there will be less Gd(DTPA)2' in areas of higher GAG

28

concentration, and greater amounts of Gd(DTPA)2' in regions of low GAG content. A
quantitative number for GAG content in a cartilage region can be calculated by obtaining
the Gd(DTPA)2' concentration of that particular region (Bashir et al., 1999).

The dGEMRIC procedure has been validated in vitro in bovine (Bashir et al.,
1996) and human (Bashir et al., 1999) tissue. Cartilage samples were first placed in
saline. After equilibration, T. of the sample was measured, and then the cartilage was
placed into a solution containing saline with added Gd(DTPA)2' (Bashir et al., 1999) and
allowed to equilibrate. T1 was again measured on the cartilage plug. Tissue Gd(DTPA)2'
concentration was calculated, followed by fixed density charge (FCD).
Glycosaminoglycan concentration was calculated using the FCD and assuming -2 moles
of charge per mole of disaccharide and a molecular weight of 502.5 g/mole (Bashir et al.,
1999)

The samples were then digested and analyzed for GAG content using the standard
biochemical method, the dimethylmethylene blue (DMMB) assay. There was a high
correlation between using dGEMRIC and DMMB assay (r2 = 0.95; slope = 1), thereby
validating the use of dGEMRIC to quantify GAG concentration in cartilage samples
(Bashir et al., 1999). The dGEMRIC results were also compared to a histological
analysis, Toluidine blue staining. Similar to the biochemical comparison, there was a
high agreement between dGEMRIC and histological analysis (Bashir et al., 1999).

The dGEMRIC procedure was further validated in vivo in human subjects. Gd(DTPA)2'
was injected intravenously into the subjects at a rate of 0.2 mM/kg BW (Bashir et al.,
1999). The subjects then exercised for 10 minutes, to ensure adequate distribution into

the tissues, as it was previously determined that exercise increased uptake into the joints

29

(Drape et al., 1993; Winalski et al., 1993). Approximately two hours post injection the
subjects underwent MR imaging of their knee joints. The GAG concentration was then
calculated. Total knee replacement surgery was performed on the subjects, and the
removed patella and tibial plateau were imaged after equilibration with Gd(DTPA)2'.
There was good correlation between the in vivo and in vitro results, validating the use of
dGEMRIC in vivo (Bashir et al., 1999). The dGEMRIC procedure appears to be an
innovative and valid method of quantifying GAG content of cartilage, and would be of

particular interest in in vivo studies.

Rationale for experiments

The major aim of these experiments was to explore the impact of sodium zeolite
A on the physiology of different animals and further clarify several uncertainties
surrounding the use of SZA. Specifically, the investigation of SZA’s effect on
osteochondrotic lesions will clear up anecdotal reports from the equine industry.
Additionally, the issue of SZA and bone metabolism will be addressed not only by
measuring bone markers of metabolism, but also by conducting mechanical stress tests
and evaluating bone architecture. Finally, the determination of changes in mineral
retention and absorption, and plasma and tissue mineral concentrations will provide a
more complete picture on the effects of SZA on mineral metabolism. The impact of
aluminum on the animals will also be investigated, as it has been largely ignored as a
component of SZA. We hypothesize that SZA supplementation will decrease
osteochondrotic lesion size in horses, enhance bone metabolism, and impact mineral

metabolism, specifically that of calcium, phosphorus, silicon and aluminum.

30

CHAPTER 3
SILICON SUPPLEMENTATION AND OSTEOCHONDROTIC LESIONS IN
TWO-YEAR-OLD STANDARDBREDS
SUMMARY
The objective of the study was to evaluate the effects of supplemental silicon on

osteochondrotic (OC) lesion size in Standardbreds. Initial radiographs were taken on
two-year-old Standardbreds (n=44) on private facilities to identify OC defects in the
distal third metacarpus/tarsus or osteochondral fragments at the dorsal aspect of the joint
and defects/fragments at the distal tibia and on the trochlear ridges of the talus. Only
clinically sound horses were included. If lameness occurred, follow-up radiographs were
taken and the horse was removed from the study. For sound horses, follow-up
radiographs of lesions were taken at 120 days. Radiographs were digitized and the length
and height of the OC defects/fragments were measured. Horses meeting study-inclusion
requirements (n=8) were pair-matched by facility and affected joint(s) and assigned to a
group: control (CO, receiving 200 g whole-grain flour) and Si supplemented (SS,
receiving 200 g bioavailable Si source). Treatments were top dressed on the feed for 120
days. Four horses had lesions in one or more fetlock(s) and four had lesions in one or
more hock(s) for a total of 12 affected joints. Due to the onset of lameness, four horses
did not complete the 120-d study duration. There was no effect of treatment, time, or
treatment*time on lesion length (P>0.22), height (P>0.51), or area (P>0.67). Silicon

supplementation did not alter the radiographic appearance of OC lesions.

31

INTRODUCTION

Osteochondrosis (0C) is defined as a disturbance of cell differentiation in joint
cartilage leading to altered endochondral ossification (J effcott, 1991). No specific cause
of DC has been identified. Key cellular events in the normal developing epiphysis
include chondrocyte proliferation, extensive extracellular matrix creation, chondrocyte
differentiation, vascular invasion and matrix calcification. In OC, the vascular
penetration of the distal region of the proliferative zone appears to fail, which disrupts the
final stages of cartilage maturation and modification of the surrounding matrix, resulting
in the accumulation of small rounded chondrocytes apparently trapped in the
predifferentiation stage within the cartilage (J effcott and Henson, 1998). The early lesion
of OC develops as a small retained core of cartilage extending into the subchondral bone.
Cartilage extracellular matrix alterations have been identified in naturally occurring DC
in horses (Lillich et al., 1998). These included a loss of glycosaminoglycans in cartilage
from DC lesions when compared to normal (Lillich et al., 1998). Proteoglycan
stimulation by chondrocytes was decreased in cartilage from equine OC lesions,
suggesting that the chondrocytes had lowered metabolism and decreased vitality (van den
Hoogen et al., 1999). It was suggested that spontaneous regression of the lesions could
not occur at that point, and that the disturbed proteoglycan synthesis was more likely to
be a sequence than a primary cause of the disturbed ossification.

Horse trainers have suggested a relationship between supplemental dietary silicon
and spontaneous healing of osteochondrotic lesions. However, no scientific proof has
been obtained. Results from several studies indicate that silicon is important in cartilage

and bone metabolism. Chicks supplemented with silicon had increased

32

glycosaminoglycan concentrations in cartilage (Carlisle, 1976). Additionally, silicon
deficiencies caused the amount of articular cartilage and glycosaminoglycan
concentrations to be considerably less in chick tibias (Carlisle, 1980). Joints were less
well formed and smaller when there was a silicon deficiency (Carlisle, 1976). Silicon has
also been associated with hastened bone mineralization (Carlisle, 1982). Besides
mineralization, the largest effect of silicon is on formation of the organic matrix (Carlisle,
1982). Silicon deficiency produced disturbed epiphyseal cartilage sequences resulting in
defective endochondral bone growth in chicks (Carlisle, 1980). Therefore, supplemental
silicon may reduce OC lesions by increasing glycosaminoglycan concentration, hastening
bone mineralization, and improving altered endochondral bone growth.

We hypothesized that supplemental silicon would decrease the size of the lesions
as they appear on radiographs. Our objective was to determine the effects of
supplemental silicon on osteochondrotic lesions in the tarsi and fetlock joints of two-
year-old Standardbred horses.

MATERIALS AND METHODS
Horses

Standardbred trainers in Southern Michigan with two-year-old horses were
contacted via phone calls and farm visits to determine their interest in participating in the
study. The horses remained on their respective facilities under the routine care of their
caretakers during the duration of the study, and continued with their respective training
programs as determined by their trainers. This study was approved by the Michigan State

University All-University Animal Care and Use Committee.

33

Radiographs

Initial radiographs were taken to identify osteochondrosis lesions in the
metacarpophalangeal, metatarsophalangeal (fetlock) and tarsocrural (hock) joints of the
horses. Radiographs were taken by a veterinary radiologist, using a portable x-ray
machine (MinXRay® HF100, Northbrook, IL) and conventional film-screen cassettes.
Lateromedial projection images were taken of all four fetlocks to identify OC defects in
the midsagittal ridge of the distal third metacarpus and metatarsus or osteochondral
fragments at the dorsal aspect of the joint (60 KVp, 1.8 mA, 0.060 sec, 80 cm focal-film
distance). Lesions in the palmar and plantar tubercles of the proximal first phalanx were
excluded due to the controversial nature of the origin of these lesions (Dalin et al., 1993).
Dorsaomedial to lateroplantar oblique projection images were acquired of both tarsi to
identify defects or osteochondral fragments at the intermediate ridge of the distal tibia
and on the trochlear ridges of the talus (65 KVp, 1.9 mA, 0.065 sec, 80 cm focal-film
distance). These views were selected based upon frequency of the specific lesions of
interest reported in previous literature (Alvarado et al., 1989; Grondahl, 1992). A
metallic marker of known size was included on each radiograph to calculate image
magnification. If horses had undergone prior joint surgery, the treated joint was excluded
from the study while other joints on the same horse were included. Only clinically sound

horses were included.

Treatments

Horses were assigned to two treatment groups, pair-matched by facility and

affected joint(s). The control (CO) horses were fed a placebo of 100 g whole-grain flour,

34

top-dressed on their normal feed twice daily, so that each horse received 200 g per day
for 120 days. The supplemental silicon (SS) horses received 100 g of sodium zeolite A
(SZA), a bioavailable source of silicon, top dressed on their feed twice daily.
Owners/trainers and the radiologist were blinded to the treatment groups. Horses were
not to be fed additional supplements for the prevention or treatment of lameness (i.e.,

chondroitin sulfate, glucosamine).

F allow-up radiographs and lesion analysis

If lameness became apparent to the owner/trainer during the study, the date and
signs of lameness were recorded, follow-up radiographs were taken, and the horse was
removed from the study so that treatment could be sought. For all sound horses, follow-
up radiographs of joints with OC lesions were taken at 120 days. The radiographs were
digitized and an imaging software program was used to measure the length and width of
the OC defects and osteochondral fragments. Area of the lesion was calculated using the
formula for area of an ellipse. Using the metallic markers on the radiographs,

adjustments were made on these measurements to account for image magnification.

Statistics
Due to the small numbers, lesions were grouped together and not analyzed by
joint. The general linear model was used in SAS 8.2. The main effects were treatment,

time and treatment*time with horse(treatment) as the error term.

35

RESULTS

F orty-four horses were radiographed to identify the presence of 0C lesions.
Thirteen horses had affected joints; however, five of those horses had plantar
osteochondral fragments and were excluded. The remaining eight were placed on study.
Of these eight horses, four horses had 0C lesions in one or more fetlock(s) and four had
lesions in one or more hock(s) for a total of 12 OC lesions in 12 joints (Table 2). At the
onset of the study it was our intention to only use a trainer if they had more than one
horse on study so that the horses could be pair-matched. However, due to the limited
availability of horses, two trainers with one horse each were included. Due to either
lameness or trainer reluctance during the trial period, four horses (three SS, one CO) did

not complete the 120-d duration of the study.

Table 2. Detailed horse information with trainer, sex, gait, joints affected,
treatment, and study duration.

 

 

 

Jointls) Affected
Trainer # Sex Gait Hock Fetlock Treatment Duration (d)
1 Gelding Trot RF SS 99
2 Horse Trot RH, LH CO 120
3 Horse Pace RH CO 120
2 Marc Pace RH, LH SS 120
4 Horse Trot Both, DT SS 62
1 Horse Pace Both, DT CO 99
2 Horse Pace Left, LT CO 120
5 Marc Trot Both, Lt SS 53

 

RF = right fore; RH = right hind; LH = lefl hind; Both = lefi and right hock; DT = lesion on distal
intermediate ridge of tibia; LT = lesion on lateral trochlear ridge

There was no effect of treatment (P = 0.32), time (P = 0.83), or treatment*time (P
= 0.22) on length of the OC lesions (Figure l). Lesion height (Figure 2) was also not

affected by treatment (P = 0.96), time (P = 0.74) or treatrnent*time (P = 0.51).

36

Additionally, lesion area was not affected by treatment (P = 0.81), time (P = 0.67) or

treatment*time (P = 0.87).

DISCUSSION

Silicon supplementation did not cause alterations in the radiographic appearance
of the OC lesions in these two-year-old Standardbred horses. Fifty percent (4/8) of the
trial horses did not complete the study. Two SS horses became lame and the trainers
elected to remove the horses to seek treatment. The cause of lameness in the two SS
horses was not determined; therefore the association between silicon supplementation and
lameness was not investigated. The other two horses (1 SS, 1 CO) were fi’om the same
barn and were removed due to trainer reluctance. At the onset of the study, palatability of
the SZA was a concern for several trainers; however, after a few days the horses became
accustomed to it and the concerns dissipated. No negative effects of the SZA
supplementation were observed.

Osteochondrosis is the disturbance of endochondral ossification by altered cell
differentiation (J effcott, 1991). Additionally, cartilage matrix alterations such as a loss of
glycosaminoglycans occur (Lillich et al., 1998). Silicon has been hypothesized to be
necessary for proper endochondral ossification and glycosaminoglycan synthesis
(Carlisle, 1976; 1980). A more appropriate study may have been investigating the effects
of supplemental silicon on preventing lesions, rather than reducing the size of lesions that
are already present.

Horses with plantar osteochondral fragments (POF) were not included in the

study because there is debate over the etiology of these lesions. Many factors point

37

towards the fragments resulting from trauma, rather than OC. One such factor is that the
classic characterization of OC, disturbance of endochondral ossification resulting in
cartilage retention, is not present in the area near POF (Dalin et al., 1993). Additionally,
the POF fragments are usually present in one joint of one type of Standardbred: the hind
fetlock in trotters. Ninety-six percent of POF are found in the hindlimb, with 75% of
them occurring in the medial region (Sandgren et al., 1993). If POF were the result of
systemic mechanisms they would be found in multiple joints. A more likely cause of
POF is trauma. More than 80% of Standardbreds show some degree of outward rotation
of the hind legs (Dalin et al., 1993), increasing the strain on the medial proximal
sesamoid bone. Minor tears in the insertion of the medial short sesamoidean ligament
may occur, resulting in formation of an osteochondral fragment from osteogenic repairs
(Dalin et al., 1993). Since it appears that trauma is the main cause of POF, and not
failure of endochondral ossification, silicon will most likely not have an effect, so these
lesions were excluded from this study.

Five out of 44 (11.4 %) horses were found to have POF. This is similar to the
11.5 % (89 out of 753) incidence reported in yearling Swedish trotters (Grondahl, 1992),
but lower than the 21.5% (145 out of 674) incidence previously reported in Swedish
Standardbred trotters around 18 months of age (Sandgren et al., 1993). As conformation
and use appears to play a large role in the formation of POF, different genetics between
Swedish Standardbreds and Michigan Standardbreds could account for the difference in
incidence in the 18-month-old horses, as well as differences in sample size. Additionally,
pacers were included in the current study, not just trotters as in the Swedish studies. Four

of the 44 (9.1%) horses had lesions on the sagittal ridge of either the third

38

metacarpus/metatarsus, and only one horse had a lesion in more than one joint. This is
close to the 15.7 % incidence found in previous literature (Grondahl, 1992). Of the five
lesions found in the fetlocks, four were in the hindlimbs. This high prevalence in the
hindlimbs was found in other Standardbreds (Grondahl, 1992). Of the 44 horses
radiographed, 9.1% had lesions in their books. The majority of total lesions found were
in the hooks (58.3%). These numbers are comparable to previous literature, in which
Alvarado et a1. (1989) found more lesions in the hook than in the fetlock and stifle in both
yearling and adult horses. Additionally, most of the lesions were on the distal
intermediate ridge of the tibia, compared to the lateral trochlear ridge (Alvarado et al.,
1989). This was also the case in the current study as 57.1% of the lesions found in the
hock were on the distal intermediate ridge of the tibia. Overall, 70.4% of the horses
radiographed did not have radiographic evidence of flattening of the sagittal ridge, POF,
or lesions on either the distal intermediate ridge of the tibia or lateral trochlear ridge.
This is comparable to the 64.5% incidence found in both yearling and adult
Standardbreds in Quebec (Alvarado et al., 1989).

Orally supplemented silicon did not alter the radiographic appearance of the OC
lesions in two-year-old Standardbred horses. Silicon supplementation may be more
appropriate at a younger age to prevent the development of osteochondrosis, rather than
healing lesions in young adult horses in active training. Additionally, the small number
of horses used in the study makes finding differences difficult. However, there is nothing
in this study to support the anecdotal reports by horse trainers of lesion regression. The
incidence rates found in the current study agree with previous literature in Standardbreds

in Sweden and Quebec.

39

Mpre
finest

Lesion Length (cm)

 

Figure 1. Lesion length (cm) in control (CO) and silicon-supplemented (SS) horses pre-
and post-supplementation of placebo or silicon.

40

m pre
Epost

    
  

O

    
  

Lesion Height (cm)

0.09

CO SS
Treatment

Figure 2. Lesion height (cm) in control (CO) and silicon-supplemented (SS) horses pre-
and post-supplementation of placebo or silicon.

41

CHAPTER 4
SODIUM ZEOLITE A SUPPLEMENTATION TO LACTATING HOLSTEINS
SUMMARY
Twenty Holstein dairy cows in either their first or second parity were used to
determine the effects of sodium zeolite A (SZA) on plasma and milk mineral
concentrations. The cows were pair-matched by parity and milk production and placed
into one of two groups — SZA supplemented (SS) or control (CO). Individual cow milk
production and feed intake were recorded daily. All cows received their normal total
mixed ration, but SS cows also were supplemented with SZA at a dosage level of 2% of
the diet. Milk, blood, and feed samples were taken on d 0, 15, and 30 for mineral
analysis. Blood was also used for osteocalcin and deoxypyridinoline determination.
Treated cows decreased feed intake and milk production from d 15 to d 30 (P<0.007) and
had lower feed intakes and milk production than CO cows on both (1 15 and d 30
(P<0.003). On (1 15, SS cows had higher plasma Si concentrations than CO cows
(P=0.01) but there was no difference on d 30 (P=0.71). Overall, plasma Al concentration
was increased by SZA (P=0.002). SS-cows had higher milk Al concentrations than CO
cows (P=0.02). Milk P concentration decreased from d 15 to d 30 in SS cows (P=0.007).
On d 15 and 30, milk P concentrations were lower in SS cows than CO cows (P<0.03).
Treated cows had higher plasma Ca concentrations, and lower plasma P concentrations
than CO cows (P<0.0002). There was no treatrnent*day effect on osteocalcin, DPD, or
DC to DPD ratio (P>0.24). Sodium zeolite A decreased feed intake, milk production,
and plasma phosphorus concentration, possibly via a phosphorus deficiency caused by

the aluminum intake. Bone metabolism was not affected by the SZA.

42

INTRODUCTION

One method of preventing milk fever is by feeding low calcium diets prior to
parturition; however, it is difficult to lower calcium in the diet. Alternatively, sodium
zeolite A (SZA) supplementation to the diet may reduce or prevent milk fever by
increasing plasma calcium levels prior to parturition (Thilsing-Hansen and Jorgensen,
2001). Several studies have reported changes in calcium metabolism due to SZA
supplementation in dairy cows (Roussel et al., 1992, Thilsing-Hansen et al., 2002;
Enemark et al., 2003). Differing hypotheses have been suggested for these changes, such
as decreased calcium absorption (Enemark et al., 2003) or increased intestinal calcium
absorption secondary to mobilization or resorption of phosphorus (Thilsing-Hansen et al.,
2002). However, some of these studies have used small cow numbers (Enemark et al.,
2003), reported flawed or ill-timed measurements (Thilsing-Hansen et al., 2002; Enemark
et al., 2003) or supplemented SZA for unequal lengths of time (Thilsing-Hansen et al.,
2002). Clarification of the effects of SZA on calcium metabolism, as well as phosphorus
metabolism, is needed.

Few studies have looked into the effects of SZA on milk mineral concentrations.
Calcium was increased in the milk of cows supplemented with 1% DMI SZA (Roussel et
al., 1992). However, calcium was the only mineral analyzed in the milk. As a sodium
aluminosilicate, SZA is broken down in the digestive system into aluminum and
monosilicic acid, an absorbable form of silicon. As a component of SZA, aluminum

must be taken into consideration as it can have neurotoxic effects in humans (N ayak,

43

2002). Increases in milk intended for human consumption should be a concern, and this
possibility should be explored.

A secondary purpose of the current study was to investigate the effect of SZA on
bone health of lactating dairy cows. Lameness in dairy cows is an animal health issue
and an economic issue. Lameness decreases milk yield in cows by approximately 360
kg/cow per 305 d lactation (Green et al., 2002). In the same study, milk yields began to
decrease up to 4 mo before the lameness was diagnosed, and continued to be depressed
up to 5 mo after treatment. In another study, milk yields were found to decrease between
1.5 kg/d and 2.8 kg/d during the first two weeks after diagnosis (Rajala—Schultz et al.,
1999). Dairy producers suffer economic losses not only from decreased production but
also from the cost of treatment. Lameness treatment amounted to 27% of total health
care costs, second behind mastitis, in a British study (Kossaibati and Esslemont, 1997).
As clinical lameness affects milk yield and cow health, measures should be taken to
reduce lameness.

Sodium zeolite A has been associated with reduced skeletal injuries in race
horses, but the mechanism by which this occurred is unknown (Nielsen et al., 1993).
Sodium zeolite A may have an effect on bone turnover. Sodium zeolite A tended to
increase osteocalcin (marker of bone formation) concentrations in broodmares (Lang et
al., 2001b), but had no effect on osteocalcin in yearlings (Lang et al., 2001c).
Additionally, one marker of bone resorption was decreased in supplemented yearlings,
while a second marker of bone resorption was unchanged in the same yearlings (Lang et

al., 2001c). The effects of SZA on bone metabolism should be further explored in cows.

44

The hypothesis was that SZA would alter plasma and milk mineral concentrations
as well as biochemical markers of bone metabolism in lactating Holsteins. The
objectives of this study were to: 1) determine the effects of SZA on plasma and milk
mineral concentrations in the lactating dairy cow, and 2) measure bone markers of
formation and resorption to determine if SZA affects bone turnover in lactating dairy

COWS.

MATERIALS AND METHODS

Cows

The research was conducted at the Michigan State University Dairy Cattle
Teaching and Research Center (DCTRC). The protocol was approved by the Institution’s
Animal Care and Use Committee. Twenty Holstein dairy cows aged 1,150 :1: 50 d and
with an average of 165 i 9 d in lactation at the start of the study were used. Ten cows
were in their first parity and 10 were in their second parity. The cows were pair-matched
by parity and milk production and placed into one of two groups - SZA supplemented
(SS) or control (CO). Weights were taken on the cows via weight tape at the start of the
study (d 0). Weights also were taken at the end of the study (d 30). All cows were
housed in tie-stalls and followed standard protocol of the DCTRC. Individual cow milk

production and feed intake were recorded daily.

Diets

All cows received their normal total mixed ration (TMR). The SS-cows were

supplemented with SZA at a dosage level of 2% of their diet, on a dry matter basis. The

45

SZA, in powdered form, was added to the TMR at the same time as all the other
ingredients on a daily basis. This assured proper mixing within the ration. The control
cows received a separately mixed TMR that contained the same ingredients as the treated
cows, except for the SZA (Table 3). Silicon concentration of the CO diet was 341 i 65
mg/kg, and of the SS diet was 1,222 i 73 mg/kg. All cows were fed their ration once

daily, and had free access to it throughout the day.

Table 3. Total mixed ration (TMR) formulation and
mineral analyses for control (CO) and supplemented (SS)

 

 

diets.
Item CO TMR SS TMR
Formulation“ % %
Corn silage 45.2 44.3
Haylage 11.8 11.6
Corn, ground 9.7 9.5
Soybean meal, 48% 7.6 7.4
Cottonseed with lint 4.7 4.6
Vitamin-mineral mix 4.5 4.4
High moisture corn, bagged 3.9 3.8
Alfalfa hay 3.8 3.7
High moisture corn, silo 2.9 2.9
QLF TMR 20 2.7 2.7
16M7 UltEXT 24/36 2.4 2.3
Energy booster 0.8 0.8
Sodium zeolite A _ 2.0
Mineral analysesb % %
Calcium 0.62 0.61
Phosphorus 0.38 0.36
Silicon 0.03 0.1
Aluminum 0.007 0.3

 

'As-fed basis
bDry matter basis

46

Plasticware

As silicon is a component of glass (Carlisle, 1972), leaching of silicon from glass
may affect silicon analysis so plastic must be used in all aspects of sample collection
(Lang et al., 2001a). Dust contamination also must be minimized. All plasticware,
including pipette tips, collection tubes, scintillation vials, pipettes, and bags, were placed
in 30% nitric acid bath overnight. Plasticware was then rinsed three times with dd H20,
placed into ovens to dry, put into acid-washed bags, and stored in a drawer to avoid dust
contamination.

Sixty-six pl of 15% KZEDTA was placed into 7-ml acid-washed plastic collection
tubes. The tubes were capped with acid-washed conventional rubber stoppers off serum
separator blood collection tubes (Vacutainer, Becton Dickinson, Franklin, NJ), and a
vacuum pump with a hose and needle inserted into the cap created a vacuum in the tubes.

These were used for collection of blood intended for plasma analysis.

Sample Collection

Milk was collected during both the morning and evening milkings on d 0, 15 and
30. A composite sample from all four teats flowed into a sample collection device during
the cows’ normal milking. This sample was then transferred to a 15-ml acid-washed
scintillation vial and immediately fiozen for mineral analysis at ~20 °C. Prior to analysis,
the milk samples were thawed and the morning and evening milking samples were
combined to form one composite sample per cow per day.

Blood was collected from all cows at 0, 15, and 30 (1 prior to daily feeding. Ten

ml of blood was collected into a 10 ml sample tube with no coagulant (Vacutainer, Fisher

47

Health Care, Chicago, IL) for serum collection, and 7 ml was collected into the specially-
made acid-washed tubes containing EDTA plasma collection. The blood was spun at
2,000 x g for 20 min, and then serum was pipetted into microcentrifuge tubes and stored
at -20° C for future analysis of osteocalcin (OC) and deoxypyridinoline (DPD). Plasma
was pipetted into acid-washed plastic microcentrifuge tubes, within a plastic-lined box to
minimize dust contamination, and stored at -20° C pending mineral analysis.

Feed samples from both treatment groups were collected at 0, 15, and 30 d by
removing portions of feed fiom each cow’s feed bin and placing the samples into their

respective labeled acid-washed plastic bags for mineral analysis.

Sample analyses

Milk samples were digested via microwave digestion for calcium and phosphorus
mineral analysis using a modified procedure by Shaw et a1. (2002). Two grams of milk
were placed into acid-washed Teflon-lined digestion vessels. Twenty ml of 40% nitric
acid (70% trace-metal grade; Fisher Scientific, Pittsburgh, PA, USA) were then added to
the milk sample and allowed to pre-digest overnight at room temperature. The vessels
were sealed and placed in the microwave digestor (MARS-5, CEM, Matthews, NC, USA)
and digested on a program consisting of a 10 min ramp up to 190 °C and 200 psi. This
was held for 20 min and was followed by a 10 min cool down. The vessels were vented,
and 2 ml of hydrogen peroxide (30%, H-1009, Sigma-Aldrich, St. Louis, MO, USA) was
added to the digested samples and allowed to cool further. Upon cooling, the samples
were transferred to acid-washed 25 ml volumetric flasks and brought up to volume with

double-deionized water. The digests were diluted 50x with lanthanum chloride and

48

analyzed on an Unicam 989 atomic absorption spectrophotometer (Thermo Elemental
Corp., Franklin, MA, USA) for calcium content. The digests were diluted 10x with
double-deionized water and analyzed with a spectrophotometer (Beckman DU 7400,
Beckman-Coulter, Palo Alto, CA, USA) for phosphorus content.

Plasma samples were diluted 25x and analyzed for calcium on the atomic
absorption spectrophotometer. To determine plasma phosphorus, the plasma samples
were precipitated with 12.5% trichloroacetic acid to remove interfering proteins. The
supernatant was then analyzed on the spectrophotometer.

All samples were run in duplicate. Bovine liver standard (1577b; NIST,
Gaithersburg, MD) was used to establish instrument accuracy. All glassware was soaked
in 30% nitric acid overnight and rinsed five times with double-deionized water (Rincker
etaL,2004)

Milk, plasma and feed samples were sent to an off-campus laboratory (London
Laboratory Services Group, Trace Elements Laboratory, London, Ontario) to be analyzed
for silicon and aluminum concentrations on a Finnigan MAT ELEMENT high—resolution
inductively coupled mass spectrometer (ICP-MS). Approximately 0.3 g of sample was
placed into a plastic tube. Five ml of 70% nitric acid were added to the tubes, which
were then placed into a water bath at 103 °C for 20 min. The digest was diluted with
water (3.5 ml) and 1 m1 of 50% potassium hydroxide solution was added. The digests
were then read on the ICP-MS.

Osteocalcin was determined by an ELISA kit (N ovoCalcin®, Quidel Corporation,
San Diego, CA, USA) following manufacturer’s instructions. Serum dilutions were

determined based on test runs of d 0 samples. Due to the variability between cows, each

49

sample was tested and a dilution was determined for each individual sample. Serum was
diluted between 1:2 and 1:13 for individual samples. Dilutions for d 15 and d 30 samples
were based off the individual d 0 values. Samples were tested in duplicate. The plates
were read at 405 nm optical density on a Spectra Max 340 plate reader (Molecular
Devices Corp., Sunnyvale, CA).

Total serum DPD was determined using a commercial competitive immunoassay
kit (Metra DPD EIA) from Quidel Corporation (San Diego, CA). Samples were not
diluted, and were run in duplicate. The plates were read at 405 nm optical density on a

Spectra 340 plate reader (Molecular Devices Corp, Sunnyvale, CA).

Statistical analysis

There were no effects of lactation and treatrnent*lactation so the data were
pooled. Data were analyzed for repeated measures using the mixed procedure in SAS
8.2. The main effects were treatment, time and treatrnent*time with cow(treatment) as
the error term. When differences were detected on d 0, d 0 was used as a covariate. Data
are displayed as least square means d: standard error of the mean. Differences were

explored at P < 0.05, and trends were explored at P < 0.1.

RESULTS
Weight
There was a treatment*day effect on weight (P = 0.03). Control cows’ weight
remained unchanged from d 0 to d 30 (618 :1: 3 vs 624 i 3 kg; P = 0.22). Treated cows

lost weight from d 0 to d 30 (577 i 3 vs 567 i 3 kg; P = 0.03).

50

Feed intake and milk production
There was an effect of treatrnent*time (P < 0.05) on both feed intake and milk
production (Table 4). The overall treatment effects on feed intake and milk production

are displayed in Table 5.

Table 4. Treatment*time effects (P < 0.05) on feed intake and milk
production in control (CO) and SZA-supplemented (SS) cows

Treatment Day0 Day 15 Day 30 SEM

 

Feed intake, kg DM co 36.9“ 41 .7b 41.4b 1.09
33 36.6“ 339'”2 29.3“2 1.09
Milk production, kg CO 40.1 40.3 38.5 0.72
ss 39.7a 33.3““2 27.992 0.72

 

"I“ Means within rows with different superscripts differ (P < 0.05)
zMeans within columns within variable differ (P < 0.05)

Table 5. Overall effect of treatment between control (CO) and SZA-
supplemented (SS) cows

CO SS SEM P-value

 

Feed intake, kg DM 41.5 31.7 1.4 0.0002
Milk production, kg 39.2 30.8 0.8 <0.0001
Minerals

Significant treatment“time effects (P < 0.1) on silicon and aluminum intake,
plasma silicon and phosphorus concentrations, and milk phosphorus concentration are
displayed in Table 6. Non-significant treatrnent*time effects are not shown. The overall
effect of treatment on mineral intake, plasma concentration, and milk concentration is

displayed in Table 7.

51

Table 6. Treatrnent*time effects (P < 0.1) on mineral intake and plasma and
milk mineral concentrations in control (CO) and SZA-supplemented (SS)

 

 

COWS
Treatment Day 0 Day 15 Day 30 SEM
Si intake, g co 6.98 7.89 7.85 0.60
88 6.93’ 230*"2 19.982 0.60
A1 intake, g co 1.33 1.50 1.50 1.64
88 1.32a 57.4",2 49.7“2 1.64
Plasma Si conc., ug/L co 518a 496’ 445b 17
88 5013 556”" 437c 17
Plasma P conc., mg/dl CO 4.5 5.2 4.8 0.37
88 5.03 0.62'” 0.99“2 0.37
Milk P conc., mg/dl co 68.8" 63.2” 65.7" 1.3
88 71.73 59.7",2 54.3“2 1.3

 

a,b,c Means within rows with different superscripts differ (P < 0.1)
2 Means within columns within variable differ (P < 0.1)

Table 7. Overall effect of treatment on mineral intake and
plasma and milk mineral concentrations in control (CO) and
SZA~supplemented (SS) cows.

 

 

CO SS SEM P-value
Si intake, g 7.9 21.4 0.8 <0.0001
Plasma Si conc,, ug/L 470 496 20 0.37
Al intake, g 1.5 53.6 1.9 <0.0001
Plasma Al conc., ug/L 129 284 30 0.002
Milk Al conc., ug/L 9.9 15.5 1.6 0.02
Ca intake, g 138 114 4 0.0004
Plasma Ca conc., mg/dl 11.9 12.9 0.2 0.0002
Milk Ca conc., mg/dl 130 128 3 0.69
P intake, g 88.8 73.9 2.8 0.0004
Plasma P conc., mg/dl 5.0 0.8 0.2 <0.0001
Milk P conc., mg/dl 65.6 55.3 1.7 0.0006

 

52

Bone markers
There was no effect of treatment*day on osteocalcin (P = 0.24), DPD (P = 0.81),
or CC to DPD ratio (P = 0.51). The effect of treatment on 0C, DPD, and the CC to DPD

ratio is displayed in Table 8.

Table 8. Overall effect of treatment on markers of bone metabolism in control
(CO) and SZA-supplemented (SS) cows.

CO SS SEM P-value

 

 

Osteocalcin 50.8 50.9 4.1 0.99

Deoxypyridinoline 7.07 7.48 0.54 0.6

OCzDPD ratio 7.38 7.93 0.71 0.59
DISCUSSION

The high intake of Al by the SS cows most likely caused the decrease in feed
intake and subsequent decreased milk production by impairing phosphorus absorption.
The absorption of phosphorus was greatly reduced, thus leading to low blood phosphorus.
The SS cows were hypophosphatemic with an average plasma phosphorus of 0.8 mg/dl,
much lower than the normal range given for dairy cows (4-7 mg/dl) (Georgievskii, 1982).
A decrease in blood inorganic phosphorus has been noted with high aluminum intakes for
decades. This decrease has been seen in humans with chronic renal failure (Clarkson et
al., 1972), chicks (Hussein et al., 1990), and lambs (Rosa et al., 1982). Karn (2001)
suggested that ruminants will exhibit clinical signs of phosphorus deficiency when blood
P levels fall below 2.0 mg/dl, as was the case in the current study. These include
decreases in both feed intake and milk production (Call etal., 1987), and unchanged

concentrations of minerals in the milk (Underwood, 1981). High aluminum intakes

53

decrease feed intake via lowering phosphorus absorption. Feed intake decreased when
lambs were fed a diet containing greater than 2 g Al/kg feed (V aldivia et al., 1982). The
SS-diet in the current study contained approximately 3 g Ang feed. Previous studies
found conflicting results on the effect of SZA supplementation on feed intake, with it
either decreasing (Johnson etal., 1988; Thilsing—Hansen et al., 2002) or increasing
(Roussel et al., 1992). Comparisons of the studies revealed that the two studies in which
feed intake decreases were found, as well as the current study, supplemented the cows
with SZA at a rate approximate to 2% of the diet. The study finding an increase in feed
intake fed SZA at levels less than 1.5% of the diet. There appears to be a threshold for
aluminum intake suppressing feed intake, as steers supplemented with 1.2 g Al/kg feed
suffered no ill effects (V aldivia et al., 1978) yet sheep consuming 2 g Al/kg feed had
depressed appetites (Valdivia et al., 1982). The 1.5% of diet may have been below the
threshold, while the 2% exceeded it.

The effects of high aluminum intakes, such as decreased plasma phosphorus, can
be alleviated by additional dietary phosphorus. Chicks fed high levels of aluminum had
decreased growth performance and bone breaking strength (Hussein et al., 1990).
However, the effects were negated when dietary available phosphorus was increased.
Increases in previously depressed plasma phosphorus occurred when animals were
supplemented with phosphorus (Lipstein and Hurwitz, 1982; Hussein et al., 1990).
Decreases in feed intake were also reversed with additional phosphorus (Valdivia et al.,
1977; Rosa et al., 1982). Feeding additional phosphorus to lactating dairy cows in order
to supplement SZA may be a viable option, but current environmental regulations on

phosphorus render this option unfeasible for producers (Stallings and Knowlton, 2005).

54

The decrease in feed intake also could possibly have been due to a change in pH.
A study supplementing SZA to cows found the treated cows to have a higher ruminal pH,
although the authors concluded that the change in pH (0.2 units) may not be of
physiological significance (Johnson etal., 1988). Two cows, one on the SZA supplement
and one not on study, were culled and euthanized for reasons not related to this project.
During the necropsies, visual inspections of the rumen, small intestine, and other parts of
the digestive tract were performed by a licensed veterinarian to determine physiological
differences between the supplemented and nonsupplemented cow. There was no
accumulation of SZA in the supplemented cow. Rumen fluid was collected from both
cows and the pH was determined using litmus paper. As there were only two cows,
statistical analysis could not be performed on the rumen pH. The pH of the cow that
never received SZA was 6.5, and the pH of the treated cow was 8.5. Thus, SZA may
have decreased ruminal pH.

The SS cows had higher plasma silicon concentrations on d 15, compared to the
CO cows. Plasma silicon concentrations increased in the SS cows from d 0 to d 15, but
then decreased to below baseline on d 30. This is most likely due to the decreased feed
intake in the SS cows, and therefore decreased silicon intake.

Plasma calcium concentration was higher in the SS cows (12.9 mg/kg) than the
CO cows (11.9 mg/kg), and was slightly higher than the normal range previously given
for dairy cows of 90-120 mg/kg (Georgievskii, 1982). An effective method to prevent
milk fever is to feed a diet low in calcium prior to parturition, and then feed a high
calcium diet during lactation. The low calcium diet then initiates homeostatic

mechanisms prior to the sudden draw on calcium reserves at parturition. However,

55

formulating diets to be low in calcium is difficult as standard ingredients are often high in
calcium. Sodium zeolite A has been an effective supplement to reduce milk fever
(Thilsing-Hansen and Jorgensen, 2001) by increasing plasma calcium concentrations.
Perhaps the binding of the aluminum in the SZA to phosphorus caused mobilization or
resorption of phosphorus, resulting in a secondary mobilization or resorption of calcium,
as suggested by Thilsing—Hansen et a1. (2002). Plasma calcium concentration increases in
lambs fed low phosphorus diets (Temouth and Sevilla, 1990). The increase in plasma
calcium may result from resorption of bone to acquire phosphorus. Additionally, an
increase in plasma calcium may arise from the blocking of proper formation of
hydroxyapatite by aluminum binding to phosphorus instead of calcium to phosphorus
(Boyce et al., 1982). This would block calcium from bone deposition, thus increasing the
plasma levels. This would explain the lack of an increase in DPD, which would be
expected if bone resorption was occurring. This could also account for the lack of
parathyroid hormone and l, 25 dihydroxy vitamin D responses found in lactating cows
given SZA (Enemark et al., 2003).

The treated cows were able to maintain stable milk calcium concentrations, which
is a common occurrence in animals with phosphorus deficiency. Lactating animals
respond to a deficiency in phosphorus by decreasing milk yield but maintaining mineral
concentrations in the milk (Underwood, 1981). The milk calcium concentrations were
within normal limits for cows (Georgievskii, 1982). Milk phosphorus concentrations
were lower in the SS cows than the control cows. Because of the decreased phosphorus
absorption, the homeostatic mechanisms were probably exhausted and the cows did not

have enough available phosphorus to supply the milk. However, both groups of cows

56

had phosphorus milk values higher than that previously reported in cows (Georgievskii,
1982). Factors such as diet, stage of lactation, and milk production could create the
discrepancy.

Osteocalcin, a vitamin-K dependent protein synthesized in osteoblasts, is a marker
of bone formation (Allen et al., 1998). Sodium zeolite A tended to increase osteocalcin
concentrations in supplemented Arabian mares, as compared to control mares (Lang et
al., 2001b). There were no effects of SZA on osteocalcin concentration in the current
study. Osteocalcin was also not affected in yearling horses on SZA (Lang et al., 2001c)
Deoxypryidinoline, a marker of bone resorption, was also not affected by SZA
supplementation in the cows. Similar to osteocalcin, effects of SZA on bone resorption
markers have brought about mixed results. Two bone resorption markers were measured
in yearlings (Lang eta1., 2001c) and broodmares (Lang et al., 2001b), and each study
showed a decrease (or trend) in one marker, and no change in the other. Due to the
conflicting results, it is likely that SZA has little to no effect on bone metabolism.
However, further research is needed to explore the effects of sodium zeolite A on calcium
and phosphorus metabolism, as bone resorption appears to be unchanged.

In conclusion, sodium zeolite A decreased feed intake and milk production in
lactating cows. Plasma phosphorus concentrations of the SS cows were below the level
at which phosphorus deficiencies occur, most likely a result of the high aluminum
intakes. Feeding additional phosphorus may alleviate the depression in feed intake and
milk production, but is impractical with regards to environmental regulations.
Additionally, the increase in milk aluminum concentration should be of major concern, as

aluminum is linked to many neurotoxic conditions in humans, including Alzheimer’s

57

disease. Sodium zeolite A had no effect on bone markers in the cows, thus its possible
use as a lameness supplement is questionable. Supplementation of 2% SZA to lactating
dairy cows is discouraged as it adversely affects feed intake and milk production and

increases aluminum concentration in milk intended for human consumption.

58

CHAPTER 5
SODIUM ZEOLITE A SUPPLEMENTATION TO NEWBORN CALVES
SUMMARY

The objective of the study was to determine the effects of sodium zeolite A (SZA) on
digestibility, glycosaminoglycan concentration, bone characteristics, mineral metabolism
and tissue mineral composition in bull calves. Twenty calves were placed on study at
three days of age, and were placed according to birth order into one of two groups: SS,
which received 0.05% BW SZA added to their milk replacer and CO, which received
only milk replacer. Blood samples were taken on d 0, 30, and 60 for osteocalcin (OC) and
deoxypyridinoline (DPD) analysis. A total collection was done on d 30 for mineral
metabolism, and on d 60 of the trial, the calves were euthanized and samples were taken
from numerous organs for glycosaminoglycan, mineral, and bone quality analyses. There
were no effects of SZA on digestibility. There were no differences in OC due to
treatment (P=0.12), and CO calves had lower DPD concentrations than SS calves
(P=0.01) but the DC to DPD ratio was not different between treatments (P=0.98).
Glycosaminoglycan concentrations were not different in synovial fluid or cartilage. There
was a trend for SS metacarpi to be longer than the CO metacarpi (P=0.09). No difference
in bone architecture or mechanical properties was detected. Plasma Cu concentration
was increased in SS calves, but P and Mg concentrations were decreased (P<0.05).
Silicon concentrations were increased in the aorta, spleen, lung, muscle, and kidney of
the SS calves, and Al was increased in all SS tissues but bone (P<0.05). Supplementation
of SZA had very little effect on bone metabolism. However, it altered plasma and tissue

mineral composition, specifically tissue Al.

59

INTRODUCTION

Sodium zeolite A (SZA) is an aluminosilicate that is hydrolyzed at low pH into
silicic acid, amorphous aluminum silicates and aluminum (Thilsing-Hansen et al., 2002).
Thus, by supplying SZA to the diet, an animal’s gastrointestinal tract can break down the
zeolite into orthosilicic acid. Silicic acid is absorbed, providing the body with
supplemental silicon. The extent of absorption of Al released from the zeolite is
unknown.

Although several studies have reported on the effects of SZA on blood mineral
concentrations (Ward et al., 1991; Lang et al., 2001b,c; Thilsing-Hansen and Jorgensen,
2001), few, if any, have investigated long term (> 14 d) supplementation on mineral
retention and absorption. Additionally, most studies have focused primarily on blood
concentrations of calcium, phosphorus and silicon. The effects of SZA supplementation
on other plasma minerals and mineral retention and absorption should be investigated.

Silicon has been suggested to be an essential mineral needed for proper bone
growth. Among numerous roles, silicon may be involved in bone mineralization,
endochondral bone formation, and glycosaminoglycan synthesis and composition
(Carlisle, 1974; 1982). In horses, a study conducted on young racing Quarter Horses
showed that supplemental SZA was associated with decreased skeletal-related injury
rates and increased distances horses were able to train before being injured (Nielsen et al.,
1993). A mechanism by which this occurred is unknown. Further research showed a trend

for increased bone formation in broodmares (Lang et al. 2001b).

6O

Studies have investigated the supplementation of SZA to horses (Frey et al., 1992;
Nielsen et al., 1993, Lang et al., 2001b,c) and reported an increase in silicon
concentration in plasma and milk. However, the level of tissue accumulation of silicon
has not been investigated. Rat kidney, liver, bone, skin, spleen and lung were found to
accumulate the greatest amounts of 31—labeled silicic acid one hour after intracardiac
injection of the labeled element (Adler et al., 1986). This appears to be the only study
that investigated the effects of supplemental silicon on tissue uptake. Silicon
supplemented through dietary means, and not via injection into circulatory system, might
have different effects. Also, a livestock species such as the horse or cow, might have
dissimilar tissue responses to supplemental silicon as the tissues might have different
concentrations than the rat due to the livestock diet of grains, forage, and oftentimes,
sand. The effects of long term supplementation of silicon are unknown as well.

Several studies have investigated the effects of high aluminum on tissue mineral
composition in livestock. Rosa et a1. (1982) used' 1,450 mg/kg Al in lambs while
Valdivia et a1. (1982) used a slightly higher level of 2,000 mg/kg, also in lambs. Cattle
fed 1,200 mg/kg Al were also studied (Valdivia et al., 1978). Very few studies have
looked at aluminum accumulation in the tissues. Valdivia et a1. (1982) found increases in
the liver, kidney and muscle. Additionally, rats chronically supplemented with aluminum
(15 wk) showed increases in kidney, spleen, and liver aluminum (Garbossa et al., 1998).
However, no studies supplementing SZA have looked into subsequent tissue mineral
composition. Sodium zeolite A, as an aluminosilicate, contains a substantial amount of
aluminum, and tissue accumulation of aluminum should also be investigated along with

silicon.

61

We hypothesized that supplemental SZA would increase bone turnover, bone
strength, and glycosaminoglycan content, as well as alter mineral metabolism and tissue
mineral composition. The purposes of this study were to determine the effects of SZA
on: 1.) digestibility and mineral metabolism 2.) variables linked to soundness including
markers of bone metabolism, glycosaminoglycan concentration, and bone mechanical
properties and 3.) tissue mineral composition. To determine bone mechanical properties
and tissue mineral composition, a terminal study was necessary. Therefore, we used the

bovine neonate.

MATERIALS AND METHODS

Calves

The research was conducted at Michigan State University’s Dairy Cattle Teaching
and Research Center (DCTRC). The protocol was approved by the Institution’s Animal
Care and Use Committee. Twenty Holstein bull calves were placed on study at three
days of age. The calves were individually housed in stalls bedded with straw. The calves
were assigned to a treatment group based on birth order, with the first calf born placed
into the control group (CO) and the second calf was placed into the group treated with
supplemental sodium zeolite A (SS), and so on until 10 calves were in CO and 10 were in

SS.

62

Treatment

At birth, calves received 2 doses of colostrum without SZA. From birth to 21
days of age calves received 2.4 l of milk replacer (Instant Cow’s Match®, Land O’Lakes
Animal Milk Products Co., Shoreview, MN, USA; Appendix C) twice daily. From 21
days of age until (1 60 calves received 3.3 l of milk replacer twice daily. The CO group
received their normal milk replacer only. The SS group received SZA at a dose of 0.05%
BW. The SZA, in powder form, was added to the SS calves’ milk replacer twice daily.
Calves were weighed weekly so that silicon dosage could be adjusted. Mineral
concentration of the milk replacer and SZA are in Table 9. The calves were allowed free

access to water.

Table 9. Mineral concentrations (ug/ g) of milk replacer and

 

 

SZA fed to calves.

Milk replacer SZA
Al 1 17 130,325
Ca 7,429 93.4
Cu 7.58 0.49
Fe 54.7 74.1
Mg 1,282 ND
P 7,698 ND
Si 479 1,632
Zn 25 18

 

Analyzed by Grand Forks Human Nutrition Research Center
ND = Not detectable
Blood samples
Blood samples were taken on d 0, 30, and 60 of the study. Ten ml of blood from
the jugular vein were collected into a Vacutainer with no coagulant to obtain serum. The

serum was analyzed for bone markers osteocalcin (OC) and deoxypyridinoline (DPD),

63

and plasma was analyzed for mineral content. Since plasma was to be analyzed for
silicon concentration, special care was taken to reduce contamination from glass. Seven-
ml plastic tubes were acid-washed, then 66 pl of 15% KZEDTA was added to the tube.
The tubes were then capped with acid-washed rubber tops, and vacuumed. Blood was
then collected into three of the 7-ml plastic tubes. Both plasma and serum tubes sat for
one hour, and then were centrifuged for 20 min at 1,340 x g. Plasma and serum were
removed and pipetted into microcentrifuge tubes and frozen at -20° C. To avoid
contamination from silicon, plastic acid-washed pipettes and microcentrifuge tubes were

used.

Total collection

On approximately (1 30, calves were placed in stainless steel metabolism crates for
three days for total collection of urine and feces. Urine and feces were collected every 6
h for the 3-d collection period. Urine flowed through wire mesh flooring, landed in a
stainless steel tray, and then flowed through a plastic acid-washed tube which ended in a
plastic acid-washed pitcher with a lid. Every 6 h, urine was poured from the pitcher into
plastic acid-washed graduated cylinders, and the volume was recorded. Ten percent of
the volume was then poured into a plastic acid-washed bottle and placed into a cooler and
then frozen until future mineral analysis. An additional 10% of the volume was placed
into a second plastic acid-washed bottle, 1.9 ml of 12 M HCl per ml urine was added, and
then the urine was frozen until nitrogen analysis. For each individual calf, the urine
collected over the three-day collection period for mineral analysis was thawed, pooled

together, poured through three layers of cheesecloth, and then 250 ml was poured into a

64

bottle and frozen. This was repeated for the acid-treated urine samples intended for
nitrogen analysis. To avoid contamination with urine, feces were collected in a canvas
tail bag lined with a removable plastic bag. Three sides (sides and bottom) of the canvas
bags were fitted to the calves’ buttocks with VelcroTM and livestock glue, with the top
attaching to the top of the calves’ tails with arms. Every 6 h the canvas tail bag was
removed from the calf and the inner plastic bag containing the feces was removed,
weighed, and placed in a cooler. Feces collected from the three-day collection period
were pooled together to form one sample per calf. Feces were thoroughly mixed
together, then approximately 300 g of the pooled feces were frozen, freeze dried, and
analyzed for nitrogen, ADF, NDF and mineral content. Approximately 4 g of feces were
placed in tin containers in the oven to determine dry matter content. Samples were left in
the oven at 105° C overnight until a constant weight was achieved. Dry matters were run

in duplicates, and a CV < 2.5% was obtained.

Tissue collection

On d 60 of the study, calves were sedated with 0.5 ml xylazine intramuscularly
and then euthanized via overdose of sodium pentobarbital into the jugular vein at a
dosage of 0.22 kag BW. Calves were determined to be dead by a lack of corneal
response. The jugular vein and carotid artery were then cut and blood was drained from
the calf. Synovial fluid was immediately collected from the hock and knee joints by a 3—
ml syringe and then placed into two 7-ml plastic acid-washed scintillation vials. The
vials were placed into a cooler on ice. Both front legs were disarticulated at the knee and

placed into a cooler on ice. The carcass was then cut open and the liver, spleen, kidneys,

65

adrenal glands, heart, and pancreas were removed entirely. Total organ weights were
recorded, and then approximately 25 g samples were taken from the organs and placed in
the cooler, with the exception of the adrenal glands, which were kept intact as they
weighed less than 25 g. The organs were then discarded. Twenty-five gram samples
were also taken from the caudal portion of the right lung, mid-section of the trachea,
aorta proximal to the bifurcation at the external iliac artery, and Iongissimus dorsi and placed
into the cooler. The same individual removed the 25 g samples to ensure that the samples
were taken from the same location on each animal. Upon return to the laboratory, the
samples were taken out of the cooler and placed in a freezer at -20° C. The front legs
were taken from the cooler and samples were taken of the mid-section of both the right
and left superficial digital flexor tendon. Articular cartilage was collected from the
weight-bearing regions of the carpi. Both left and right fused metacarpal bones (MC 111
and IV) were stripped down to the periosteum, and all samples were placed in a freezer at

-20° C until further analysis.

Bone marker analyses

Serum OC concentration was determined using a commercial competitive
immunoassay (Metra Osteocalcin EIA) from Quidel Corporation (San Diego, CA) (Hiney
et al., 2004b). Serum samples were diluted with double-deionized water in a 1:30 ratio
for d 0 and 30, and in a 1:20 ratio for d 60. Total serum DPD was determined using a
commercial competitive immunoassay kit (Metra DPD EIA) from Quidel Corporation

(San Diego, CA) (Hiney eta1., 2004b). All samples were run in duplicate.

66

Digestibility analyses

Urine, ground dried fecal samples, and powdered milk replacer were analyzed for
nitrogen by a LECO-FP-2000 (St. Joseph, MO) using AOAC method 990.03. Acid
detergent fiber and NDF were determined using an approved protocol (Goering and Van
Soest, 1970). Bomb calorimetery was used to determine energy content of the milk
replacer and fecal samples. Nitrogen and energy retention was determined by
subtracting output from intake, and ADF, NDF, nitrogen and energy digestibilities were
determined by subtracting fecal output from intake, and then dividing the difference by

intake.

Bone mechanical properties
The right fused third and fourth metacarpal bones were imaged with computed

tomography (Marconi PO 6000) at the level of the mid-diaphysis (120 Kv, 125 mA, 2
sec, slice thickness 10 mm). The transverse images were used to measure cross-sectional
area of total (cortex + medullary canal) and cortical bone, cortical thickness at the dorsal,
palmar, medial and lateral margins and total bone diameter. Bone length was determined
using a ruler with 1 mm precision. Measurements were made in triplicate to ensure
accuracy.

Three-point bending to failure was performed on the right fused third and fourth
metacarpal bones using a universal testing instrument (Model 4202, Instron Corp.,
Canton, MA) according to ASAE Standards (2000) to determine peak fracture force,

ultimate bending strength, and modulus of elasticity. A cross-head speed of 10 mm/min

67

was used with supports set at 9 cm apart and force data were logged every 0.25 mm.
Ultimate bending strength (stress) was calculated by the following equation:

0 = FLC/4I

where

o = ultimate bending stress, Pa

C = distance from neutral axis to outer fiber (m)

I = moment of inertia, m4

L = distance between supports, m

F = applied fracture force, N

C for a hollow ellipse:

C=D/2

The moment of inertia for a hollow ellipse:

1 = O.O49[(B-D3) - (b'd3)]

where

B = lateromedial (LM) bone diameter (outside major diameter), m
b = LM medullary diameter (inside major diameter), m

D = dorsopalmar (DP) bone diameter (outside minor diameter), m

d = DP medullary diameter (inside minor diameter), m

Apparent modulus of elasticity was calculated by using the data from the three-

point bending test. The portion of the force/deformation curve between 10 and 90 % of

68

the peak fracture force was used to calculate the slope of the straight line (F /8) via linear
regression of those data. This portion was chosen because it represented the linear portion
of the curve. Modulus of elasticity (E) was then calculated by the following equation:

E = FL3/4815

where

5 = deformation, m

Bone ash

The left fused third and fourth metacarpal bones were cut at the midshaft point by
a bandsaw and a slice 1 cm wide was obtained. A second slice, 0.5 cm wide, was cut
proximal to the carpus from the previous slice. The 0.5 cm slice was used to determine
mineral content of the bone while the 1 cm slice was used to determine dry matter and
percent ash. Fat was removed from the 1 cm slices via other extraction. The slices were
then dried at 150° C overnight and weighed. Following that the slices were placed in a
muffle firmace at 600° C for 12 h and reweighed. The percent ash was then expressed on

a dry fat-free basis.

Glycosaminoglycan concentration

Articular camge GAG concentration. Three cartilage samples per calf were sent
to the Massachusetts Institute of Technology (MIT) to be analyzed for GAG content.
Recently a method was developed that utilizes magnetic resonance imaging (MRI) to
determine cartilage GAG content using the charged contrast agent gadolinium diethylene

triamine pentaacetic acid (Gd(DTPA)2') (Burstein and Gray, 2003). This method,

69

delayed gadolinium-enhanced magnetic resonance imaging of cartilage (dGEMRIC), is a
sensitive and specific measure of GAG content that can be used for both in vivo and in
vitro analyses. The researchers at MIT were blinded to treatment. The cartilage samples
were initially equilibrated in Hank’s balanced salt solution (Invitrogen) (HBSS) to
measure T10 (Tl without contrast agent). Each sample was imaged individually with
HBSS in tubes. An 8.45T Bruker microimaging system was used. There was one T1
weighted scan per sample. Imaging parameters are stated below:

Sequence: Saturation recovery (msme_vtr)

TR: 100, 150, 300, 400, 600, 900, 1400, 2000, 3200, 5000 msec.

TE: 15 msec

Averages: 1

Slice thickness: 0.5

Slice Number: 3 or 4 depending on sample size

Matrix Size: 128 x128

FOV: 128 x128

Coil size: 10 mm

For dGEMRIC imaging, the samples were re-equilibrated in HBSS containing
lmM GdTPA'2 (Magnivist, Berlex Imaging, NDC 50419-188-02). Each sample was
again imaged individually with the lmM GdTPA'2 solution in tubes. The imaging
parameters were the same as above except the TRs were: 100,125, 175, 275,375,475,600,
900, 1800, and 2700 msec.

Tissue Gd(DTPA)2' concentration was calculated by the following equation:

[Gcrzj=l[_l _._l_]
R 716.1 710

7O

where T1 Gd = T; measured after equilibration of cartilage in saline and Gd(DTPA)2',
T10 = T1 measured after equilibration of cartilage in saline only,

and R = relaxivity of Gd(DTPA)2', which is approximately equal to the relaxivity in

saline (Bashir et al., 1999).

Fixed charge density (P CD) was then determined using the following equation:

FCD =—2[Na+] fill—-____V[Gd2]m (
1 bath ’[qu 1M”, [Gd ’2 11m“

Glycosaminoglycan concentration was calculated using the FCD and assuming -2
moles of charge per mole of disaccharide and a molecular weight of 502.5 g/mole (Bashir
et al., 1999). The equation is as follows:

502.5g/mol)* 10_3

[GA G] = FCD,(

Synovial fluid hyaluronicacid concengation. Hyaluronic acid concentration was
determined in the synovial fluid using a commercial competitive ELISA kit (Echelon,
Salt Lake City, UT). Synovial fluid samples were diluted in a 1:2,500 ratio with double-

deionized water. Samples were run in duplicate.

Mineral analyses

Prior to mineral analysis, tissue samples (excluding bone and cartilage) were
freeze-dried and then ground. Plasma, urine, feces, milk replacer, sodium zeolite A,
water from the dairy farm, and the tissue samples were analyzed for Si, Ca, P, Al, Cu,
Mg, Fe, and Zn at the Grand Forks Human Nutrition Research Center (Grand Forks, ND).

For minerals other than Si, a nitric acid digestion was used. Approximately 0.5 g of

71

sample was weighed into Teflon tubes, and 10 ml of ultra-pure H2N03 was added. The
tubes were then capped, and placed into a heating block at 60 °C for two days. The tubes
were then uncapped, and the temperature of the heating block was raised to 135 °C.
Once the acid evaporated out, and a dry pellet at the bottom of the tube was obtained, 10
ml of ultra-pure H2NO3 was again added. This step was repeated until a pale yellow
pellet was obtained, at which point 7 ml of H2NO3 and 3 ml of H202 was added to the
pellet. This was also allowed to evaporate out. The final pellet was brought up to
volume in 5 ml volumetric flasks with double-deionized water and then analyzed on an
inductively-coupled argon plasma atomic emission spectrometer (PerkinElmer Optima
3100 XL). Samples were digested and analyzed in duplicate. A standard reference
material (National Institute of Standards and Technology, Gaithersburg, MD), no. 1577b
bovine liver, was used for quality control purposes.

For silicon analysis, an alkaline medium was used to digest the samples.
Approximately 0.3 g of sample was weighed into 15 mL conical centrifuge tubes and 1
ml of 0.2 % (wt/v) of sodium dodecyl sulfate (SDS) was added to the tubes, followed by
9 ml of 40 % tetramethyl ammonium hydroxide (TMAH). The SDS was added to allow
for better fat digestion. After the addition of SDS and TMAH, the tubes were capped and
vortexed until all samples was suspended into TMAH mixture. The tubes were then
placed on a rocker for 16 hours. All of the sample should be solubilized, but small
amounts of undigested material may remain. The samples were centrifuged prior to
analysis to remove any suspended particles. No additional dilutions were needed prior to
analysis, thus reducing contamination. The samples were then analyzed by an

inductively-coupled plasma atomic emission spectrometer (PerkinElmer 3300). Typical

72

Diet, standard reference material no. 1548a, was used for quality control purposes instead

of bovine liver as the liver does not have a standard value for silicon.

Statistics

The mixed procedure for repeated measures in SAS was used to analyze weight,
OCD, DPD and OC:DPD ratio data. The model tested for day, treatment, and
treatment*day interactions. The remaining data were analyzed using the general linear
model in SAS. Bartlett’s test for homogeneity of variance was used in SAS. vaariances
were determined to be heterogenous, Welch’s test was used for treatment differences, of
which individual standard error of the means (SEM) were reported. If variances were
homogeneous, then pooled SEMs were reported. Differences were explored at P < 0.05,

and trends were explored at P < 0.10.

RESULTS

There was no effect of treatrnent*day on calf weights (P = 0.19) or average daily

gains (P = 0.89) (Table 10).

Table 10. Body weights and average daily gain
(ADG) of control (CO) and SZA-supplemented

 

 

(SS) calves.
CO SS SEM
Body weight (kg)
Initial 46.9 49.9 0.94
Final 85.5 85.0 0.94
ADG 0.65 0.59 0.03

 

Digestibility variables

73

The average daily volume of urine excreted from the SS calves was greater than
the CO calves (5,338 3: 317 ml vs. 4,348 :1: 317 ml; P = 0.04). After accounting for
differences in BW, there was a trend for SS calves to have a greater daily average volume
of urine excretion on a per kg BW basis (81.0 i 5.3 ml/kg vs. 66.8 :t 5.3 ml/kg; P = 0.07).
There was no difference in average daily fecal excretion (SS: 0.62 i 0.05 kg; CO: 0.58 i
0.05 kg; P = 0.61) or on 3 kg feces/kg BW basis (SS: 0.93 i 0.08 kg/kg BW; CO: 0.88 :1:
0.08 kg/kg BW; P = 0.68). There was a greater fecal dry matter percent in SS calves
compared to control calves (18.5 at 0.6 % vs. 16.4 i 0.6 %; P = 0.03).

There was a trend for ADF and NDF digestibility to be greater in the SS calves
compared to the CO calves (P = 0.08; Table 11). There were no differences in the other

digestibility variables (P > 0.4; Table 11).

Table 11. Digestibility variables in control (CO) and SZA-supplemented (SS)

 

 

calves.
CO SS SEM P-value

ADF intake (g) 5.71 5.71 - -
ADF fecal output (g) 11.2 6.52 1.76 0.08
ADF digestibility (%) -96.2 -14.2 30.8 0.08
NDF intake (g) 12.1 12.1 - -
NDF fecal output (g) 17.9 10.8 2.70 0.08
NDF digestibility (%) -47.7 10.6 22.3 0.08
Nitrogen intake (g) 52.6 52.6 - -
Nitrogen fecal output (g) 21.6 22.6 0.09 0.4
Nitrogen retention (g) 31.1 30 0.9 0.4
Nitrogen digestibility (%) 88.8 89.0 0.8 0.9
Energy intake (Mcal) 5.57 5.57 - -
Energy fecal output (Mcal) 0.42 0.38 0.04 0.5
Energy retention (Mcal) 5.15 5.19 0.04 0.5
Energy digestibility (%) 92.5 93.1 0.6 0.5

 

74

Bone turnover, dimensions, and mechanical properties

There were no treatment*day effects on DC, DPD and OC:DPD ratios. Overall,
CO calves had OC concentrations of 244 i 13 ng/ml, and SS calves had concentrations of
275 i 13 ng/ml but they were not significantly different (P = 0.12). Osteocalcin
concentrations decreased from d 30 to d 60 (284 :t 12 vs 235 i 12 ng/ml; P=0.003).
Overall, SS calves had higher DPD concentrations than CO calves (12.4 i 0.6 ng/ml vs.
10.2 3: 0.6; P=0.01) and there was no day effect (P=0.9). The DC to DPD ratio was not
different between treatments (P=0.98). There was a trend for SS metacarpals to be longer
than the CO metacarpals, and there were no differences in bone dimensions as measured
on the CT images (Table 12). There were no treatment differences in the measured or
calculated values of peak force, ultimate bending stress, moment of inertia, or modulus of
elasticity of the metacarpals (Table 13).

Table 12. Length, diameters and cortical widths of MC III&IV in
control (CO) and SZA-supplemented (SS) calves.

 

CO SS SEM P-value
Bone length (cm) 18.5 18.9 0.18 0.09
Diameters (cm)
Dorsopalmar 1.98 1.98 0.03 0.93
Dorsopalmar medullary 1.26 1.27 0.04 0.99
Lateromedial 2.69 2.70 0.04 0.96

Lateromedial medullary 1.80 l .80 0.05 0.94
Cortical widths (cm)

Dorsal 0.36 0.37 0.01 0.51
Palmar 0.34 0.35 0.01 0.72
Lateral 0.45 0.45 0.01 0.91
Medial 0.46 0.47 0.01 0.50

 

75

Table 13. Bone mechanical properties of MC III&IV from

control (CO) and SZA supplemented (SSLcalves.

 

Peak Ultimate
Fracture Moment Bending Modulus
Force of Inertia Strength of Elasticity
(N) (mm4) (MPa) (MPa)

CO 4243 8.59E-09 l 13 0.45
SS 4340 8.64E-09 l 14 0.47
SEM 123 4.53E-10 6 0.04
P-value 0.58 0.93 0.85 0.68

 

Glycosaminoglycan concentration

Hyaluronic acid concentrations in the synovial fluid were not different between

treatments with the mean of CO being 1,200 i 78 jig/ml and the SS mean being 1,177 i

78 11ng (P = 0.84). The addition of sodium zeolite A to the diet did not increase GAG

concentration in the articular cartilage as determined by dGEMRIC (P = 0.85).

Concentration in the SS calves was 57.9 i 4.6 jig/mg wet wt and it was 56.5 i 4.6 jig/mg

wet wt in the CO calves.

Mineral analysis

Concentrations of the feces and urine from the calves are displayed in Table 14.

The SS calves had a greater fecal concentration of Al, and tended to have a lower Cu

fecal concentration (P < 0.10). There was no difference between treatments in urine

mineral concentrations.

76

Table 14. Fecal and urine mineral concentrations in control
(CO) and SZA-supplemented (SS) calves.

 

 

CO SS SEM P-value
Fecal (mg/g)
Al 2.50 34.0 1.43 < 0.0001
Ca 37.5 36.6 2.55 0.80
Cu 0.02 0.15 0.002 0.07
Fe 0.82 0.72 0.05 0.11
Mg 8.21 7.40 0.51 0.28
P 20.1 20.0 1.30 0.93
Zn 0.29 0.23 0.04 0.34
Urine (pg/ml)
Al 0.35 0.33 0.03 0.63
Ca 7.15 5.81 0.96 0.34
Cu 0.05 0.04 0.004 0.46
Fe 0.08 0.06 0.01 0.41
Mg 1.02 0.74 0.22 0.39
P 785 550 149 0.28
Zn 0.09 0.07 0.03 0.78

 

Data on mineral retention and absorption are shown in Table 15. The SS calves
retained larger amounts of aluminum than the CO calves. Phosphorus apparent
digestibility tended to be decreased in the SS calves.

Plasma mineral concentrations are displayed in Table 16. There was a day effect
for calcium, iron, and phosphorus (P<0.05). Calcium concentration was highest on d 0
compared to d 30 and d 60 (P<0.02). Iron and phosphorus concentrations were lower on
d 30 compared to d 0 and 60 (P < 0.04). There were treatment effects for copper,
magnesium, and phosphorus concentrations with SS calves having higher plasma copper
concentrations, and lower magnesium and phosphorus concentrations (P < 0.05). Values
are within normal ranges previously reported (Georgievskii, 1982), except for phosphorus

and zinc concentrations, which are higher than reported.

77

Table 15. Metabolism of minerals in control (CO) and SZA-supplemented (SS)

 

 

calves.
Intake Urine Fecal Retained Digested
(mg) (mg) (mg) (mg) (%)
Aluminum
CO 130 1.50 223 93.7 -70.8
SS 4,445 1.66 3,752 692 15.7
SEM 70 0.11 169 143 38.2
P-value < 0.0001 0.30 < 0.001 0.001 0.13
Calcium
CO 8,510 30.5 3,393 5,087 60.1
SS 8,513 29.9 4,102 4,381 51.8
SEM 0.05 4.53 304 303 3.6
P-value < 0.0001 0.91 0.12 0.12 0.12
Copper
CO 8.44 0.20 2.03 6.21 75.9
SS 8.46 0.21 1.74 6.51 79.3
SEM 0.0003 0.02 0.29 0.30 3.5
P-value < 0.0001 0.68 0.49 0.49 0.49
Iron
CO 61.5 0.34 75.9 ~14.8 -23.6
SS 63.9 0.31 78.7 —15.1 -23.1
SEM 0.04 0.06 5.13 5.12 8.2
P-value < 0.0001 0.77 0.71 0.97 0.97
Magnesium
CO 1,657 4.78 751 901 54.7
SS 1,657 3.87 814 839 50.9
SEM . 1.22 52.7 52.7 3.2
P-value 0.61 0.41 0.42 0.42
Phosphorus
CO 8,576 3,241 1,824 3,511 78.7
SS 8,576 2,850 2,231 3,494 73.9
SEM 524 163 578 1.9
P-value 0.61 0.09 0.98 0.09
Silicon
CO 533 ND ND - -
SS 587 ND ND - -
SEM 0.88 - - - -
P-value < 0.0001 - - - -
Zinc
CO 28.1 0.36 27.8 -0.007 1.3
SS 28.7 0.34 24.5 3.89 14.7
SEM 0.01 0.11 4.71 4.79 16.9
P-value < 0.0001 0.89 0.62 0.57 0.58

 

78

Table 16. Plasma mineral concentrations (pg/ml) in control (CO) and SZA-
supplemented (SS) calves

 

Day 0 Day 30 Day 60 SEM Day*Trt P-value

Aluminum
co 0.40“ 0.88"" 0.97b 0.18 0.95
88 0.68 0.83 0.72 0.18

Calcium
C0 143“ 126b 1 15b 6 0.54
88 131“ 117“b 115b 6

Copper
co 0.66 0.69 0.67 0.05 0.82
88 0.74 0.812 0.802 0.05

Iron
co 1.82“ 0.93b 1.77“ 0.28 0.61
88 1.57“ 0.93b 2.04“ 0.28

Magnesium
co 22.1 22.1 20.4 1.02 0.79
88 20.1 19.32 18.8 1.02

Phosphorus
C0 89“ 104b 85“ 5 0.25
88 78 87“ 85 5

Silicon
co ND ND ND - -
88 ND ND ND - —

Zinc
co 3.92 3.92 3.94 0.5 0.99
88 3.74 3.91 3.85 0.5

 

"b'c Means within rows with different superscripts differ (P < 0.1)
zMeans within columns within variable differ (P < 0.1)

79

Table 17. Percentage of body weight (%) of organs from
control (cg and SZAfliplemented (SS) calves.

CO SS SEM P-value

 

Adrenals (both) 0.009 0.009 0.001 0.85
Heart 0.64 0.65 0.02 0.72
Kidney (both) 0.57 0.59 0.02 0.45
Liver 1.96 1.82 0.09 0.24
Pancreas 0.07 0.09 0.01 0.15
Spleen 0.65 0.58 0.05 0.36

 

Organ weights, expressed as a percentage of body weight, are presented in Table
17. There were no differences between the two groups of calves (P < 0.05). Dry matter
percentages of the organs and feces collected from the CO and SS calves are presented in
Table 18. Dry matters were increased in the bone, heart and feces (P < 0.03), and tended

to be increased in the muscle of the SS calves (P = 0.08).

Table 18. Organ and feces dry matter (%) from
control (CO) and SZA-supplemented (SS) calves.

CO SS SEM P-value

 

Adrenal 21.0 21.8 0.4 0.20
Aorta 29.7 28.9 0.6 0.33
Cortical bone 87.9 88.1 0.05 0.01
Heart 21.4 23.2 0.4 0.008
Kidney 18.8 19.3 0.4 0.42
Liver 25.7 26.6 0.5 0.18
Lung 20.5 20.2 0.3 0.49
Muscle 23.1 24.2 0.4 0.08
Pancreas 24.4 24.9 1.0 0.72
Spleen 25.5 26.4 0.7 0.37
Tendon 29.5 28.8 0.4 0.28
Trachea 27.4 26.5 0.8 0.38
Feces 16.4 18.5 0.6 0.03

 

80

There was no difference (P = 0.44) in bone ash percentage between the treated
group (66.5 i 0.4 %) and control group (66.1 i 0.4%). Mineral concentrations of the
collected organs are presented in Table 19. Aluminum was increased in all organs of the
SS calves (P < 0.05) except for bone. Sodium zeolite A increased adrenal Zn, aorta Ca,
Mg, P and Zn, heart Mg and Zn, liver Ca and Mg, muscle Ca, pancreas Mg, and tendon
Ca (P £0.05). Additionally, SZA decreased kidney and liver Fe (P _<_0.05). Silicon was
increased in the kidney, lung, muscle and spleen of the SS calves (P < 0.05).

Table 19. Mineral concentrations (pg/g dry weight) of various
tissues from control (CO) and SZA-supplemented (SS) calves.

 

CO SS SEM P-value
Adrenal
A1 8.4 30.7 2.1 < 0.0001
Ca 411 445 16 0.17
Cu 9.59 9.42 0.7 0.86
Fe 189 174 12 0.4
Mg 564 594 16 0.2
P 12,492 13,123 435 0.32
Si BDL BDL - -
Zn 65.3 74.1 2.5 0.02
Aorta
Al 1.68 3.15 0.32 0.005
Ca 333 372 8 0.003
Cu 0.98 1.19 0.10 0.14
Fe 49.8 39.9 6.9 0.32
Mg 174 186 4 0.05
P 1,954 2,169 52 0.009
Si 2.89 4.73 0.3 0.001
Zn 34.4 38.4 1.2 0.03

 

BDL = Below detectable limits

81

Table 19 (cont’d).

 

 

CO SS SEM P-value
Cortical
bone
A1 8.8 17.8 5.1 0.22
Ca 125,493 129,378 5,784 0.64
Cu 0.27 0.34 0.04 0.21
Fe 8.1 7.7 0.8 0.74
Mg 1,771 1,833 114 0.70
P 55,656 56,667 2,590 0.79
Si 3.82 4.46 0.26 0.11
Zn 40.5 44.0 2.8 0.39
Cartilage
Al 2.15 4.30 0.73 0.05
Ca 1,134 1,171 209 0.9
Cu 0.50 0.27 0.12 0.18
Fe 4.26 4.09 0.47 0.81
Mg 96.7 88.4 5.0 0.26
P 690 632 101 0.69
Si 1.51 1.38 0.31 0.77
Zn 3.90 3.75 0.35 0.76
Heart
A1 2.9 15.7 1.4 < 0.0001
Ca 166 180 5 0.1
Cu 14.9 14.6 0.2 0.36
Fe 171 161 6 0.25
Mg 1,045 1,223 5 8 0.04
P 11,425 11,325 218 0.75
Si 3.84 5.09 0.62 0.17
Zn 73.3 79.0 2.0 0.05
Kidney
A1 7.4 55.4 4.7 < 0.0001
Ca 590 542 54 0.54
Cu 13.9 13.7 0.8 0.83
Fe 139 106 10 0.04
Mg 729 818 31 0.05
P 12,564 10,816 939 0.2
Si 8.] 44.2 8.3 0.007
Zn 81.9 86.6 3.6 0.37

 

82

Table 19 (cont’d).

 

 

CO SS SEM P-value
Liver
A1 3.5 65.0 6.2 < 0.0001
Ca 147 185 7 0.0007
Cu 573 603 38 0.57
Fe 224 149 26 0.05
Mg 538 612 15 0.003
P 10,148 10,976 372 0.13
Si 1.43 2.08 0.49 0.36
Zn 141 151 15 0.64
Lung
A1 9.4 28.2 2.7 0.0002
Ca 376 390 15 0.48
Cu 5.6 5.3 0.3 0.39
Fe 319 253 24 0.08
Mg 571 564 6 0.46
P 10,690 10,225 184 0.09
Si 5.36 7.58 0.48 0.005
Zn 81.3 80.1 1.2 0.5
Muscle
A1 4.3 9.0 0.9 0.002
Ca 177 205 7 0.01
Cu 3.0 3.4 0.2 0.31
Fe 44.2 39.9 5.3 0.57
Mg 1,078 1,191 46 0.1
P 9,977 10,425 379 0.41
Si 4.85 7.63 0.95 0.05
Zn 86 92.8 4.74 0.33
Pancreas
A1 4.0 12.6 1.0 < 0.0001
Ca 453 498 25 0.22
Cu 3.86 4.18 0.29 0.44
Fe 112 90 l 1 0.16
Mg 728 851 40 0.04
P 13,014 13,283 660 0.78
Si 7.80 7.34 1.14 0.78
Zn 195 295 37 0.07

 

83

Table 19 (cont’d).

 

 

 

CO SS SEM P-value
Spleen
A1 3.4 22.6 3.4 0.001
Ca 127 115 20 0.66
Cu 2.68 3.11 0.29 0.3
Fe 3,406 6,600 1,938 0.26
Mg 370 388 35 0.73
P 6,601 6,728 396 0.82
Si 2.00 3.60 0.24 0.0002
Zn 61.1 65.5 4.2 0.48
Tendon
A1 7.5 14.6 1.9 0.02
Ca 374 41 l 1 1 0.05
Cu 1.5 1.7 0.2 0.52
Fe 23.5 18.8 2.7 0.22
Mg 221 189 32 0.47
P 2,382 2,141 302 0.56
Si 15.9 22.1 4.3 0.29
Zn 47.1 49.3 5.8 0.77
Trachea
A1 2.16 4.73 0.65 0.03
Ca 1,007 1,052 33 0.36
Cu 0.93 1.13 0.11 0.22
Fe 33.3 31.6 3.55 0.74
Mg 257 255 6 0.75
P 1,753 1,694 63 0.51
Si 5.28 6.84 0.89 0.23
Zn 18.2 18.0 1.1 0.94
DISCUSSION

Previously, studies using SZA have focused on the effects of silicon alone;
however, with the high concentrations of not only Si but Al as well, both minerals should

be considered when evaluating the effects of SZA.

84

Sodium zeolite A did not affect the weight of the calves, nor did it affect average
daily gain. Horses fed varying levels of SZA also had similar weight gains or average
daily gains compared to horses receiving no supplemental SZA (Frey et al., 1992).
Similar results were seen in chickens (Ballard and Edwards, 1988), and pigs (Ward et al.,
1991)

To determine the effect of the SZA on digestibility variables, total collection of
feces and urine was attained. The SS calves had a greater average daily output of urine
over the 3-d total collection period than the CO calves. Water intake was not recorded,
but it may be that the SS calves drank more water than the CO calves. Fecal dry matter
was also increased in the SS calves. It is also interesting that fewer SS calves were
afflicted with scours (8 CO, 4 SS), as noted by the MSU Dairy Farm workers who
worked with the calves on a daily basis and kept a journal of such occurrences. The
addition of zeolites to the diets of ruminants, poultry, and swine can decrease the
incidence of scours as well as other intestinal disorders (Mumpton, 1984). The exact role
the zeolite plays in reducing scours is unknown. One suggestion has been that the effects
are due to the zeolite’s alkalinity and buffering capacity in the gastrointestinal tract
(Mumpton, 1984). Or perhaps the effect is through silicon, as silicon impacts the
immune system (Moseley et al., 1988; Seabom et al., 2002) and thus might reduce scours
and other intestinal problems through this route.

Sodium zeolite A did not affect nitrogen retention or digestibility, nor was the
amount of energy retained or digested affected. However, there was a difference in ADF
and NDF digestibility, with the SS calves tending to have greater fiber digestibilities

although most of the values are negative. Previous literature has shown that natural

85

zeolite supplementation increases ADF digestibility (Sweeney etal., 1984) while
synthetic zeolite supplementation did not affect ADF digestibility (Johnson eta1., 198 8).
The different zeolite sources may have been the cause for the differing results, with the
synthetic zeolite probably being most like the SZA used in the current study. While it
appears that SZA increased fiber digestibility, more than likely an outside factor or a side
effect of SZA supplementation affected the results. Although intake of milk replacer was
identical in the two groups of calves (they consumed all that was given them), one factor
that was unaccounted for was the intake of straw bedding. As significant straw
consumption was unexpected, values for intake were not collected. The ADF and NDF
values in the milk replacer were relatively low, while values in the fecal samples were
much higher. However, the consumption of straw high in fiber might account for the
higher fecal fiber values. The lower digestibility values in the CO calves could be due to
higher straw intakes compared to the SS calves. It is difficult to determine if the
differences in fiber digestibilities are from a direct effect of SZA in the digestive tract on
fiber digestion, or from an indirect effect on feed intake. The increase in straw
consumption may also contribute to the decrease in fecal dry matter in the CO calves.
The straw would attract water, thus decreasing dry matter.

It is unknown if the desire to seek supplemental sustenance in the CO calves was
increased above normal, or if it was depressed in the SS calves. A decrease in feed intake
has been seen in cows supplemented with zeolites (Johnson et al., 1988; Thilsing-Hansen
et al., 2002; Turner, Chapter 4, Dissertation). However, studies have also shown an
increase in feed intake with zeolite supplementation (Roussel et al., 1992). The

differences in results could be due to differences in amounts of SZA supplemented, as the

86

studies in which decreases were found fed SZA at levels approximately equal to 2% of
the diet, while the increase in feed intake came after SZA supplementation at 1.5% of the
diet. The SS calves on the current project were receiving approximately 3% of their diet
in SZA on a dry matter basis, and it varied with their body weight. This might have been
the cause of the potential decrease in foraging behavior (thus feed intake) in the SS
calves.

Deoxypyridinoline, a marker of bone resorption, was higher in SS calves than CO
calves. Despite the increase in DPD in SS, there were no treatment differences in the ratio
of CC to DPD, indicating that as DPD increased in the SS calves, so did 0C. This
suggests that there was neither a net loss nor gain of bone, which is also reflected in the
lack of differences seen in measured mechanical properties of the bone. Because DPD
was higher in the SS calves but there was no difference in the ratio, this may signify the
SS calves were experiencing more rapid bone turnover than the CO calves.

The SS calves tended to have longer metacarpi than the CO calves. This finding is
consistent with studies in other species of animals. Carlisle (1980) found that chicks fed a
diet supplemented with 250 mg Si/kg feed had greater tibial lengths compared to control
chicks fed a basal diet containing 1 mg Si/kg feed. Furthermore, greater femoral lengths
have been reported in growing rats supplemented with 35 mg/kg of Si versus control rats
receiving only a basal level of 2 mg Si/kg feed (Nielsen and Poellot, 2004). Although
longitudinal growth of the calf metacarpi was affected by silicon, radial growth was not.
Length is obtained through endochondral ossification, while thickness is obtained

through the apposition of new bone on both the endosteal and periosteal surfaces.

87

Endochondral ossification has been shown to be increased in animals on a silicon
adequate diet, as compared to animals on a silicon deficient diet Carlisle, 1980).

Supplemental silicon did not affect metacarpi mechanical properties such as peak
fracture force, ultimate bending stress and modulus of elasticity. This finding is
consistent with a previous study in which supplemental silicon did not affect the same
variables in the femurs of growing rats (Nielsen and Poellot, 2004). Therefore, the lack
of an effect on bone breaking properties may indicate that silicon does not have a major
effect on bone crystal formation or function. However, low dietary silicon decreased
plasma osteopontin, and increased both plasma sialic acid concentration and urinary
helical peptide excretion, indicating that silicon can regulate extracellular matrix protein
synthesis (Nielsen and Poellot, 2004).

Bovine metatarso—phalangeal synovial fluid contains approximately 0.5 mg/ml
hyaluronic acid (Coleman et al., 1999). The hyaluronic acid concentration
(approximately 1.2 mg/ml) in the 60-d-old calves in the current study was a little higher
than that of the previous study (Coleman et al., 1999). However, the age of the calves
should be taken into account as younger subjects seem to have higher hyaluronic acid
concentrations than older subjects (Balazs, 1982). Although silicon has been suggested
as a component of hyaluronic acid (Schwarz, 1973) and is potentially involved in the
synthesis of glycosaminoglycans (Carlisle, 1974), supplemental silicon did not increase
hyaluronic acid concentrations in the SS synovial fluid compared to the CO synovial
fluid, nor did it increase glycosaminoglycan concentrations in the articular cartilage.
Increased glycosaminoglycan concentrations were found in chicks compared to chicks

that were on a diet deficient in silicon (Carlisle, 1974). The CO diet, while lower in

88

silicon than the SS diet, was not deficient in the mineral. It would appear that silicon is
not a limiting factor for glycosaminoglycan synthesis, specifically hyaluronic acid, when
diets are not deficient in the mineral.

Total glycosaminoglycan concentration in the articular cartilage from the weight-
bearing portions of the carpus and fused metacarpi also failed to increase with the
addition of bioavailable silicon via sodium zeolite A. Perhaps GAG concentration did
not increase because of an aluminum effect. The cartilage aluminum levels were
increased in the SS calves. Following injection of an aluminum compound into rabbit
knees, aluminum invaded the synovial cell layer and caused an apparent loss of
proteoglycan in superficial zones of tibial and femoral cartilages (Chary-Valckenaere et
al., 1994). Superficial zones of cartilage were the targets of interest in the current, and
with the subsequent rise in cartilage aluminum content, perhaps possible increases in
GAG concentration caused by additional silicon were negated by the additional
aluminum.

Intake of most of the minerals was greatest in the SS calves, with the exception of
Mg and P which had identical intakes between the groups. Magnesium and phosphorus
were not detected in the SZA, and the calves ate identical amounts of milk replacer which
would account for the identical intakes of those minerals. Except for aluminum and
silicon, the difference between intakes of the other minerals is negligible and should not
be taken into account.

Feeding SZA or aluminum to livestock has resulted in alterations to calcium
and/or phosphorus metabolism, either by increasing calcium plasma concentrations

(Allen et al., 1986; Turner, Chapter 4, Dissertation) and retention or digestibility

89

(Robinson et al., 1984), or by decreasing plasma phosphorus concentration (Thilsing-
Hansen et al., 2003; Turner, Chapter 4, Dissertation) and retention or digestibility

(V aldivia et al., 1982). In the present study calcium retention and digestion were not
affected by additional SZA, but phosphorus digestion tended to be depressed.
Additionally, a trend for greater amounts of phosphorus excreted in the feces was seen in
the SS calves, indicating less phosphorus absorption. The phosphorus can bind to the
aluminum, creating non-absorbable complexes (Krueger et a., 1985).

The SS calves had lower plasma phosphorus concentrations, averaged across all
days, than the CO calves. This has been seen previously in cows fed SZA (Thilsing-
Hansen et al., 2003; Turner, Chapter 4, Dissertation). Unlike the cows though, the
plasma concentration did not drop below the normal range and the calves were not in a
hypophosphatemic state (Georgievskii, 1982). The increase in Al intake decreased the
phosphorus absorption, thus plasma phosphorus concentration was decreased. However,
the accompanying increase in plasma calcium concentration seen previously in the
lactating cows (Turner, Chapter 4, Dissertation), was not found in the current study,
perhaps because the decrease in phosphorus was much smaller in the calves (about 10%)
compared to the cows (about 84%). The larger decrease in the lactating cows resulted in
subsequent drastic homeostatic mechanisms to maintain phosphorus levels. The calves
did not have to go to such drastic measures, such as resorption of bone, which is
supported by the similarity of calcium and phosphorus cortical bone content between the
SS and CO calves. Therefore, the resulting increase in plasma calcium following bone

resorption did not occur in the SS calves.

90

Previous literature has suggested that silicon is found in “unusually large
quantities” in the connective tissues such as aorta, trachea, bone and tendon (Carlisle,
1972). The high silicon content of connective tissue is hypothesized to be mainly from
its presence as an integral component of the glycosaminoglycans and their protein
complexes that contribute to the structural framework of the tissues. In the current study,
“unusually large quantities” were not noted in the connective tissues, and the connective
tissues did not contain most of the silicon. Tendon did contain the highest amount of
silicon, but the trachea was in the middle of the other tissues, and the aorta was towards
the bottom of the group. The suggestion that the high silicon content in the connective
tissues is due to its composition in glycosaminoglycans should be reevaluated, as the
connective tissues in the present study did not contain the highest silicon concentrations.

Sodium zeolite A supplementation increased aorta, kidney, lung, muscle, and
spleen silicon concentrations. The large increase in kidney silicon concentration (445%)
could be due to high silicon in the urine. Alternatively, the high silicon concentrations
could be indicative of silicon toxicity in the urinary tract. Silicon toxicity can cause
calculi (Carlisle, 1984). Calculi were created in rats and dogs fed high levels of a silicon
supplement, not silica (Carlisle, 1984). However, the formation of calculi is a relatively
unclear process. Besides urinary problems, silicon toxicity also affects the lungs.
Silicosis has been suggested to occur following inhalation of dust and silicate, which
stimulates a severe reaction in the lungs (Carlisle, 1984). Based on the silicon
accumulation in the lungs in the SS calves, it may be plausible that silicon deposited in

the lungs from the blood may also contribute to silicosis, not just inhaled silicon.

91

The increase in silicon in the aorta most likely caused an increase in the zinc
concentration. Tohno et a1. (2002) found a significant correlation between silicon and
zinc in the jugular vein. Additionally, aluminum, calcium, phosphorus and magnesium
are directly correlated in the aorta (Minami et al., 2001). In the current study, the
aluminum most likely increased first, thus increasing the concentrations of the three other
minerals. The aorta was the tissue with the greatest number of changes in mineral
concentrations.

There was no effect of SZA on plasma aluminum in the SS calves. After feeding
high levels of aluminum, Hussein et a1. (1990) also found that plasma aluminum
remained unchanged. Previously, the lack of plasma response was thought to indicate
that dietary aluminum is poorly absorbed and has little effect on animal life (McCollum
et al., 1982). King et al. (1997) found that apparent gastrointestinal uptake of Al is
between 0.1-0.3%, which was probably slightly underestimated. The SS calves in the
current study retained approximately 15% aluminum.

Plasma aluminum concentrations likely do not change because the aluminum is
rapidly transported to the tissues. Valdivia et a1. (1982) found that at least a portion of
dietary aluminum is absorbed and transported as there were increases in the liver, kidney
and muscle of lambs following supplementation of 2,000 mg Al/kg feed. Additionally,
rats chronically supplemented with aluminum (15 wk) showed increases in kidney,
spleen, and liver aluminum (Garbossa etal., 1998). Although there was not an increase
in plasma aluminum in SS calves, there were dramatic increases in aluminum in all SS
tissues analyzed except cortical bone. The smallest increase was in the aorta, which

increased 87.5%, and the largest was in the liver, with a 1,757% increase in aluminum.

92

The average increase was 384%. This reiterates the fact that the level of aluminum
cannot be overlooked in sodium zeolite A, and must be considered when evaluating data.

While the extent of tissue damage by the aluminum accumulation is not known in
the calves, several studies have indicated that the accumulation could be harmful.
Decreased alkaline phosphatase activity in aluminum supplemented rats was concluded to
be an indication of aluminum toxicity, as there was substantial hepatic aluminum
accumulation (Guo et al., 2004). Hepatocytes can be destroyed by large deposits of
aluminum (N ayak, 2002). The supplemented rats also had lower values of anemia-
related variables such as hematocrit, hemoglobin, and plasma iron, most likely fi'om
aluminum interference with iron. Due to the high concentration of aluminum, future
studies should determine the effects of sodium zeolite A supplementation on
neurofunction of the brain. In humans, aluminum toxicity has been associated with
memory loss, tremor, jerking movements, impaired coordination, sluggish motor
movement, loss of curiosity, and ataxia (N ayak, 2002). Rats fed 0.3% aluminum in their
drinking water displayed specific cognitive impairment (J ope and Johnson, 1992)
Concern should be taken when feeding SZA to animals, particularly to animals producing
a food product intended for human consumption, as there was a significant increase in the
Iongissimus dorsi muscle of the SS calves. Aluminum has been linked to several disease
syndromes in humans, not the least of which is Alzheimers Disease.

A decrease in plasma magnesium concentration occurred following
supplementation with SZA. Although decreased, the plasma magnesium values were still
within normal limits (17-25 jig/m1) for calves (Georgievskii, 1982). The decrease was so

small that the effects are probably of little physiological value. Steers fed high aluminum

93

also had decreased magnesium blood levels (Allen and Robinson, 1980; Allen et al.,
1984; Allen etal., 1986), as did chicks (Hussein et al., 1990) and lambs (Rosa et al..,
1982; Valdivia et al., 1982). The mechanisms by which aluminum interacts with
magnesium are not as clearly defined as with phosphorus (Allen, 1984). Unlike with
phosphorus, the effect of aluminum on magnesium does not appear to be from the
formation of unabsorbable complexes (Allen and Fontenot, 1984). Although serum
magnesium levels are depressed, there are no changes in magnesium absorption (Valdivia
et al., 1982; Allen and Fontenot, 1984). In the current study, there was also not a change
in magnesium retention or absorption. However, there were increases in heart, kidney,
liver, muscle and pancreas magnesium concentrations. It appears that the magnesium is
cleared from the plasma and deposited in the tissues of the SS calves.

Plasma copper concentrations were higher in the SS calves than the CO calves.
However, there were no differences in retention or absorption. Moreover, copper tissue
concentrations were not changed by SZA supplementation. The cause of the increase in
plasma copper concentration in the SS calves is unclear, although similar results occurred
in aluminum supplemented rats (Guo et al., 2004).

Concentrations of zinc, copper, and magnesium in the liver, kidney, heart,
pancreas, spleen, and muscle were similar to values reported in Holstein bull calves
(N eathery et al., 1990b). However, the iron concentrations of the tissues in both the SS
and CO calves in the current study were approximately half that of values reported by
Neathery et al. (1990b). Additionally, the calves were in a negative iron balance. Both
groups of calves appeared to be releasing iron from their tissues and excreting it out. It is

also interesting to note that on d 30, when the mineral balance collection was begun,

94

plasma iron was at its lowest point in both groups. The plasma then rebounded on d 60
and returned to baseline (d 0) levels, yet tissue iron was still below normal values.
Aluminum decreases iron absorption and tissue levels (Han et al., 2000); however, the
control calves had negative iron balances as well as the supplemented calves. As water
comprised the majority of the calves’ diets (powdered milk replacer was mixed into it),
the water from the faucet used to distribute the calves’ water was tested for mineral
concentration, and was taken into account in the mineral balance. Thus, the occlusion of
water minerals was not a factor. The low tissue iron may arise from low dietary levels.
The iron requirement for calves on milk replacer is 100 mg Fe/kg feed; however, the
“requirement” for veal calves is less than 50 mg/kg (NRC, 2001). Veal calves are
purposefully fed well below their iron requirement so that maximal myoglobin formation
and hemoglobin formation are decreased (Underwood and Suttle, 1999). The calves in
the current study were being fed as veal calves, and the iron concentration of the milk
replacer was approximately 55 mg Fe/kg feed, well below the true requirement (100
mg/kg) for calves. Thus, their tissue iron would be decreased because they would not
have excess iron to store. The liver iron concentration of young animals depends on the
content of the feed and may serve as an indicator of adequate supply to the animal
(Georgievskii, 1982). The one tissue however that was not low in iron was the spleen,
which contained the highest amount of iron. Iron is known to accumulate in the spleen
due to the spleen’s red blood cell (containing hemoglobin) storage function
(Georgievskii, 1982).

With the exception of lung phosphorus, the only mineral to decrease in the tissues

of the SS calves was iron. Decreases in tissue iron, particularly liver iron, have been

95

reported previously (Han et al., 2000). The decreases in iron may be a result of an
interaction with aluminum. The mechanisms by which aluminum affects iron
metabolism are debatable. Aluminum can compete with iron for absorption (Cannata et
al., 1991; van der Voet, 1992; Han et al., 2000); however, retention and absorption of
iron in the current study appeared to be unaffected by SZA supplementation. It may be
that there is a higher affinity of aluminum for transferrin (Buys and Kushner, 1989; Guo
etal., 2004), which might explain the increases in aluminum and the decreases in iron.
However, transferring may be only 30% saturated with iron, which leaves ample open
binding sites for aluminum (Buys and Kushner, 1989). The reduction of F6” to Fe”,
important for translocation of iron and for the release of iron from ferritin, is mediated by
ceruloplasmin, which may be inhibited by aluminum (Jeffrey et al., 1996). Tissue ferritin
levels were decreased in chicks supplemented with aluminum (Han et al., 2000). The
cause of the decreased tissue ferritin was unknown, but the iron regulatory protein may
have failed to accurately sense the iron status of the cell due to the presence of aluminum.
Or perhaps the aluminum was competing with the iron for binding sites on the iron

regulatory protein.

CONCLUSIONS
The addition of sodium zeolite A had little effect on digestibility variables, with
the trend for increased ADF and NDF digestibility more than likely a result of
unaccounted for straw intake. Previous studies have shown no effect on ADF

digestibility by synthetic zeolite supplementation.

96

Since a trend for increased length was the only difference in the bovine
metacarpal bones between the control and SZA supplemented animal, the means by
which SZA supplementation was associated with a decrease in lameness in racing
Quarter Horses (Nielsen et al., 1993) is still unknown. Based on the DPD and DC to DPD
ratio data possibly revealing more active bone metabolism in the SS calves, our current
hypothesis is that the mechanism in the racehorses was rapid repair of subclinical
injuries, rather than prevention of damage. More research needs to be conducted to
further elucidate the mechanism by which SZA decreases musculoskeletal injury rates.

Plasma mineral concentration responses to SZA, such as the decreases in
magnesium and phosphorus, are in agreement with previous literature. The decrease in
phosphorus apparent digestibility more than likely caused the decrease in plasma
phosphorus; however, the cause of the plasma magnesium decrease is unknown. It
appears though to be a result of clearance of the magnesium from the blood and
deposition in the tissues. Sodium zeolite A increased silicon concentrations in the lungs,
spleen, kidney, and aorta of the SS calves. There was an increase in aluminum retention,
but plasma aluminum was unaffected. The excess aluminum was likely rapidly cleared
from the blood and deposited into various tissues. Substantial increases in aluminum
were found in all tissues except bone. Decreases in tissue iron could be linked to
aluminum interactions, and the increases in zinc could be linked to the decreases in iron.
Due to interactions of aluminum and silicon, to obtain a better understanding of tissue

response to supplemental bioavailable silicon, a source lacking aluminum is necessary.

97

CHAPTER 6

CONCLUSIONS AND IMPLICATIONS

Supplemental sodium zeolite A bad no effect on existing osteochondrotic lesion
size as determined by radiographs. However, a better approach might be researching the
effects of additional dietary silicon on prevention of osteochondrotic lesions, not
treatment. One of the difficulties in a study such as the one investigating SZA on
osteochondrotic lesion resolution is analysis sensitivity. Perhaps the SZA could have
affected the lesions, but the effects were not visible yet because there were no differences
in lesion size as determined using radiographs. Radiographs are useful in determining the
actual location and size of the lesions, but offer little to no information on the metabolic
state of the lesion or surrounding cartilage. The use of a technique such as delayed
gadolinium-enhanced magnetic resonance imaging of cartilage (dGEMRIC) could be
helpful in such a study by supplying information such as glycosaminoglycan
concentration and location in the cartilage. This technique would be a more sensitive
measure than radiographs, and could detect changes in the cartilage before radiographs
could. It could reveal the current state of cartilage metabolism.

Feeding SZA to the lactating cows and young calves produced numerous effects
on mineral metabolism, particularly that of calcium and phosphorus. The results were
somewhat consistent between the two groups of animals, but the effects were not as
dramatic in the calves as they were in the cows. The addition of 2% DMI SZA to the
lactating cows drastically decreased feed intake, milk production, and plasma phosphorus

concentration while increasing calcium plasma concentration. The decrease in plasma

98

phosphorus was substantial, and the concentrations were below levels determined to be
hypophosphatemic. The calves, fed 0.05% BW SZA, had decreased plasma phosphorus
and tended to have decreased phosphorus absorption. The decrease in plasma
phosphorus was slight, and levels were still with normal ranges. Although the calves
were only fed equal amounts of milk replacer, which was consumed entirely, fiber
digestion data would indicate that the control calves consumed more straw bedding than
the supplemented calves, although this was not recorded. Decreases in feed intake in
adult cows have previously been reported (Johnson et al., 1988; Thilsing-Hansen et al.,
2002). The calves had very little rumen development, determined by visual inspection
during tissue harvesting, and were basically monogastrics. Perhaps this is the reason for
the slightly decreased phosphorus absorption and plasma phosphorus concentration,
while the cows had drastically decreased plasma concentrations. Additionally, the level
of phosphorus in the diet may have played a role, as increasing dietary phosphorus when
feeding high levels of aluminum alleviates the negative effects of the aluminum (Lipstein
and Hurwitz, 1982; Rosa et al., 1982; Hussein et al., 1990). The cows were fed 0.36 %
phosphorus, which is right at their requirement (NRC, 2001), while the calves’ diet
consisted of 0.76% phosphorus, which was approximately 8% above their requirement of
0.70% (NRC, 2001). There was no excess phosphorus in the cow diet, so the high
aluminum had drastic effects on the phosphorus metabolism of the cows. The slight
excess in the calf diet was marginally sufficient in preventing the decrease in phosphorus
absorption as it was only a trend (P = 0.09).

Sodium zeolite A has relatively little effects on bone metabolism. There was no

effect of SZA on osteocalcin in either the lactating cows or young calves. While the

99

cows displayed no changes in DPD concentration, the calves had increased DPD values.
Several studies have measured bone markers of metabolism in SZA supplemented
animals and found mixed results. Yearlings on SZA had no decrease in PYD
concentrations, but ICTP concentrations were decreased (Lang et al., 2001c).
Alternatively, SZA supplemented broodmares had no decrease in ICTP, but a trend for
decreased PYD on d 30 existed; however, there was no difference on d 45 (Lang et al.,
2001b). In the current study, there was no difference between OC and DPD ratios
between the calf groups. As DPD was increased in the SS calves, it might signify that the
calves had an increased rate of bone turnover, but no net loss or gain of bone compared to
controls. Besides measurement of bone markers, the bones were mechanically tested and
architecture was examined. There were no differences between the two groups of calves,
validating the bone marker analysis that sodium zeolite A may have only marginally
affected bone metabolism. The measurement of bone markers in a nutritional study
utilizing SZA should be reconsidered as strong results are not obtained, and it seems to be
unlikely that SZA has a large effect on bone metabolism.

There was a trend for a difference in length of the fused metacarpi bones between
the SS and CO calves. The bones tended to be longer in the SS calves, as were femoral
bones in rats supplemented with silicon (Nielsen and Poellot, 2004). This may be
explained by the potential association of silicon with adequate endochondral ossification
(Carlisle, 1980), the process for longitudinal growth. Lengths of other bones were not
investigated, and measures such as wither and hip heights were not obtained. It is
unknown if other long bones such as the femur were affected by the SZA

supplementation. It would be interesting to further investigate the effects of silicon on

100

long bone growth. Specifically, would the overall height of the animal be increased, or
was the process accelerated so that eventually the control animals would reach the same
height. If the bones remained longer than the control animals, it could create positive
effects on animals such as race horses. The longer bones could create longer stride
lengths, and thus possibly faster times. Alternatively, if the silicon simply accelerated the
process of growth, problems could be created. Many problems can arise when horses are
on a higher plane of growth.

Marked changes in tissue mineral composition occurred following SZA
supplementation. The tissue with the most changes was the aorta, with increases in
aluminum, calcium, magnesium, phosphorus, silicon, and zinc. Calcium, phosphorus,
magnesium and aluminum concentrations are correlated in the aorta (Minami et al.,
2001), as are silicon and zinc (Tohno et al., 2002). The increase in silicon could prove to
be an interesting find, in that the aorta may be strengthened, which could aid in reducing
exercise-induced pulmonary hemorrhage in race horses, suggested to be caused by failure
of the pulmonary capillaries (West et al., 1991, 1993; Manohar and Hutchens, 1993).
Alternatively, the increase in calcium in the muscle may also be an interesting find.
Increased calcium in muscle makes the meat more tender (Montgomery et al., 2004), and
thus more palatable to humans. However, the tenderness arises from proteolysis
(Koohmaraie et al., 1988), which might be harmful in an athletic animal such as the
horse.

The accumulation of aluminum by most of the tissues is alarming, and should be
of concern. The lack of an increase in plasma aluminum does not warrant a conclusion

that the animal is not affected by the aluminum intake. In a performance animal such as

101

the horse, SZA supplementation may be detrimental to athletic performance as it is
possible that aluminum may reduce hemoglobin and hematocrit (Guo et al., 2004).

The use of 2% sodium zeolite A as a dietary supplement to animals should be
reconsidered. The substantial accumulation of tissue aluminum and therefore potential
adverse consequences would preclude any benefits, regardless of whether or not the
animal is intended for human consumption. The possibility of excess aluminum in
human diets is a situation which should be avoided.

Determining the effects of sodium zeolite A on variables becomes slightly
complicated as the effects of aluminum, silicon, and aluminum-silicon interactions must
be taken into account. In this particular project, the high levels of aluminum appear to
override the effects of bioavailable silicon. Perhaps glycosaminoglycan concentrations in
cartilage would have increased if silicon acted alone, and did not have to overcome
potential counter effects by aluminum. In addition, the distribution of supplemental
silicon in tissues might have been different if not for the aluminum interactions.
Therefore, when designing studies to evaluate the effects of bioavailable silicon, it would

be best to use supplements without aluminum.

102

APPENDICES

103

APPENDIX A

MINERAL BALANCE IN HORSES FED SODIUM ZEOLITE A

SUMMARY

The purpose of the study was to investigate the effects of sodium zeolite A, a
supplement high in aluminum and silicon, on mineral metabolism of adult horses. Eight
mature Arabian geldings were randomly placed into one of two groups: control (CO) or
supplemental silicon (SS). All horses received 1.5 kg of a commercial sweet feed and
had ad libitum access to hay daily and the SS horses received an additional 200 g of
sodium zeolite A per day. Following a 10-d adaptation period, the horses were brought
into stalls and fitted with total collection devices. Feces and urine were collected for 3
consecutive days. Hay intake was recorded. Blood samples were taken on d 13 for
mineral analysis. The addition of sodium zeolite A to the horses’ diets did not affect their
weight (P = 0.95). There was no difference (P = 0.30) in hay intake between SS horses
and CO horses. Intakes of Al and Si were greater in the SS horses (P < 0.03). Sodium
zeolite A increased plasma Si concentration and increased both fecal and urine Si
excretion (P < 0.05). Although plasma Al concentrations were not changed, Al retention
tended to be greater in SS horses (P = 0.08). Calcium retention and apparent absorption
were increased by the SZA supplementation (P < 0.05), and there were no effects on
phosphorus absorption. Supplementation of sodium zeolite did not affect weight and
feed intake of the horses. Calcium retention and absorption were increased, and Al
retention tended be increased, but P was unaffected. Caution should be used when
supplementing horses with SZA as a lack of a plasma response does not indicate tissue

uptake, and the combination of high A] intake and retention could lead to adverse effects.

104

INTRODUCTION

Silicon, as the second most abundant element found in the earth’s crust, is found
in many substances such as dust, soil, and glass. Although silicon is found throughout
the environment, it is not readily absorbable in the gastrointestinal tract. A bioavailable
form of silicon, sodium zeolite A (SZA) has been used in numerous studies to determine
the effects of supplemental silicon. As a sodium aluminosilicate, SZA is broken down in
the digestive system into aluminum and monosilicic acid, an absorbable form of silicon
(Thilsing-Hansen et al., 2002).

Sodium zeolite A has been used in both ruminant (Johnson et al., 1988; Roussel et
al., 1992; Thilsing-Hansen et al., 2002; Turner, Chapter 4, Dissertation) and equine
studies (Frey et al., 1992; Nielsen et al., 1993; Lang et al., 2001a,b). However, the results
of the studies have yielded different results between the species. Several, if not most,
studies in which cows were supplemented with SZA described decreases in feed intake
(Johnson etal., 1988; Thilsing-Hansen et al., 2002; Turner, Chapter 4, Dissertation).
However, no adverse effects have been reported in the equine studies. The decreases in
feed intake seen in the ruminants are most likely a result of a phosphorus deficiency
caused by the high intake of aluminum fi“om SZA. Significant changes in both
phosphorus and calcium metabolism have been reported in cows supplemented with SZA
(Enemark et al., 2003; Thilsing-Hansen and Jorgensen, 2001; Thilsing-Hansen et al.,
2003). While mineral metabolism studies have been conducted in cows supplemented
with SZA, no such studies have been done in the horse. By investigating the effects of
SZA supplementation on mineral metabolism, specifically calcium, phosphorus,

aluminum and silicon, perhaps the cause for the difference between equine and ruminant

105

reactions to SZA may be elucidated. Our hypothesis was that calcium and phosphorus
metabolism would not be affected by supplemental SZA in horses. Our objective was to
determine the effects of sodium zeolite A supplementation on mineral metabolism of

horses.

MATERIALS AND METHODS
Horses
Eight mature Arabian geldings with an average body weight of 468 d: 7 kg were
used for the study. The horses were housed on dry-lots during a lO-d diet adaptation

period. The protocol was approved by the Institution’s Animal Care and Use Committee.

Treatment

The geldings were pair-matched by weight and randomly placed into one of two
groups: control (CO) or supplemental silicon (SS). All horses received 1.5 kg of a
commercial sweet feed and had ad libitum access to hay daily. The SS horses received
an additional 200 g of SZA per day, top-dressed on their sweet feed. Mineral
concentrations of the sweet feed, hay, and SZA are in Table 1A. The total basal diet
(CO) contained approximately 75 mg/kg silicon and 115 mg/kg aluminum. With the
addition of SZA the total SS diet had a silicon concentration of approximately 87 mg/kg
and an aluminum concentration of approximately 2837 mg/kg. These numbers are based
on the average hay intake of the two groups, and are therefore approximations. During
the diet adaptation period the horses were individually fed their rations. Horses were

weighed on d 0 and again on d 13.

106

Table 1A. Mineral concentrations (mg/kg dry wt) of

 

 

the diet

Al Ca P Si
Sweet feed 97.7 5,850 5,080 68.7
Hay 113 9,640 3,330 72.6
SZA 133,000 132 38.7 646

 

Total collection

Following the 10-d adaptation period, the horses were brought into stalls and
fitted with total collection devices, or nappies (Equisan, Melbourne Victoria, Aust.).
Feces and urine were collected for 3 consecutive days. The nappies were emptied every
8 h, and total feces and urine were weighed, and the weight was recorded. Then 10% of
the feces and urine was saved and placed into a cooler. The remaining portions were
discarded. At the end of the total collection period the saved urine and feces were pooled
so that each horse had one urine and one fecal sample for the 3-day period. The
composite urine and feces were stored in the freezer until mineral analysis.

During the collection period the hay intake of the horses was recorded. Samples

of the hay, sweet feed, and sodium zeolite A were taken for mineral analysis.

Blood collection

Blood samples were taken on d 13 of the study, the last day of the total
collections. Since plasma was to be analyzed for silicon concentration, special care was
taken to reduce contamination from glass. Seven-ml plastic tubes were acid washed, then
66 pl of 15% K2EDTA were added to the tube. The tubes were then capped with acid-
washed rubber tops, and vacuumed. Blood was then collected into three of the 7-ml

plastic tubes. The tubes were centrifuged for 20 min at 1,340 x g. Plasma was removed

107

and pipetted into microcentrifuge tubes and frozen at -20° C. To avoid contamination

from silicon, plastic acid-washed pipettes and microcentrifuge tubes were used.

Mineral analysis

The fecal samples were freeze-dried, and then ground in a mill. The samples of
sweet feed and hay were ground as well. Subsamples of the feces, hay, sweet feed, and
SZA were reserved and placed in a drying oven to obtain dry matter data. Approximately
0.5 g of feed, SZA, hay, and feces were placed into plastic tubes. Approximately 1 m1 of
plasma and 2 ml of urine were placed into plastic tubes as well. Five m1 of 70% nitric
acid were added to the tubes containing feed, SZA, hay and feces, and 2 ml of 70% nitric
acid were added to the tubes containing urine and plasma. The tubes were then placed
into ovens at 90 °C and allowed to sit overnight. Then 10,000 jig/ml cesium (ionization
synchronizer) and 100 ppm yttrium (Y) were added to all tubes so that the final
concentrations were 500 ppm cesium and 1 ppm Y. Samples were then transferred to
volumetric flasks (25 ml for hay, feed, SZA and feces, 10 ml for plasma and urine),
brought up to volume with double-deionized water, and poured into plastic conicles. The
diluted samples were then analyzed with an inductively-coupled plasma mass
spectrometer at the Diagnostic Center for Population and Animal Health at Michigan

State University (East Lansing, MI, USA). Samples were run in triplicate.

Statistical analysis

Data were analyzed with the general linear model in SAS 8.2. Bartlett’s test for

homogeneity of variance was used in SAS. If variances were determined to be

108

heterogenous, Welch’s test was used for treatment differences, of which individual
standard error of the means (SEM) were reported. If variances were homogeneous, then
pooled SEMs were reported. Differences were explored at P < 0.05, and trends were

explored at P < 0.10.

RESULTS

The addition of SZA to the horses’ diets did not have an effect on their weight (P
= 0.95). There was no difference (P = 0.30) in hay intake between SS horses (8.57 :1: 0.81
kg DM) and CO horses (7.27 i 0.81 kg DM). Fecal dry matter was higher in SS horses
than CO horses (21.7 i 0.8 % vs. 17.7 i 0.8 %; P = 0.01).

Plasma, fecal and urinary mineral concentrations are displayed in Table 2A.
Sodium zeolite A increased plasma Si, fecal Al, and urinary Si concentrations, decreased
fecal Ca and tended to decrease fecal P concentration.

Table 2A. Plasma, fecal, and urine mineral
concentrations in control (CO) and SZA-supplemented

 

 

(SS) horses.
CO SS SEM P-value
Plasma (pg/ml)
Al 0.24 0.23 0.08 0.95
Ca 119 119 2 0.94
P 89.3 86.6 1.8 0.34
Si 0.53 1.02 0.12 0.03
Fecal (mg/kg)
A1 655 5,955 767 0.003
Ca 9,590 8,071 446 0.05
P 6,124 5,690 273 0.3
Si 72.2 99.1 10.4 0.12
Urine (pg/ml)
Al 0.19 0.26 0.05 0.46
Ca 2,377 1,829 452 0.42
P 7.17 4.99 0.7 0.07
Si 28.6 46.6 3.3 0.008

 

109

Information such as intake, absolute mineral amounts in feces and urine,
retention, and absorption is included in Table 3A. Intakes of Al and Si were greater in
the SS horses (P < 0.03). There was a greater fecal Al and Si output and a greater urinary
Si output from the SS horses (P < 0.02). Calcium retention and apparent digestion and

aluminum retention were affected by the SZA supplementation (P < 0.10).

Table 3A. Metabolism of minerals in control (CO) and SZA-supplemented
(SSLhorses.

Intake (g) Urine (g) Feces (g) Retained (g) Digestegf/o)

 

Aluminum
CO 0.95 0.001 2.49 -1.55 -172
SS 26.6 0.002 23.5 3.16 11.8
SEM 0.09 0.0006 1.55 1.59 71.6
P-value < 0.0001 0.28 < 0.001 0.08 0.12
Calcium
CO 77.6 16.7 36.5 24.4 52.9
SS 90.2 16.2 33 41.0 62.9
SEM 7.86 4.50 2.87 4.81 1.92
P-value 0.30 0.93 0.42 0.05 0.01
Phosphorus
CO 31.2 0.05 23.1 7.6 24.0
SS 35.6 0.04 23.4 11.7 33.1
SEM 2.7 0.008 1.6 1.8 3.78
P-value 0.30 0.61 0.92 0.15 0.14
Silicon
CO 0.62 0.19 0.27 0.15 54.7
SS 0.84 0.40 0.40 0.03 51.3
SEM 0.05 0.04 0.03 0.06 6.4
P-value 0.04 0.006 0.02 0.22 0.73

 

DISCUSSION
Sodium zeolite A is an aluminosilicate that is hydrolyzed at low pH into silicic

acid, amorphous aluminum silicates and aluminum (Thilsing-Hansen etal., 2002). The

110

high amounts of aluminum in the SZA as compared to silicon necessitates evaluating
SZA as a source of aluminum as well as bioavailable silicon. The SZA significantly
increased the intake of silicon and aluminum in the horses.

Previous studies found conflicting results on the effect of SZA supplementation
on feed intake in cows. Cows supplemented with 2% SZA decreased their feed intake
(Johnson et al., 1988; Thilsing-Hansen et al., 2002; Turner, Chapter 4, Dissertation) while
cows supplemented with less than 1.5% SZA increased their intake (Roussel et al., 1992).
Additionally, steers supplemented with 1,200 mg Al/kg feed suffered no ill effects on
intake (Valdivia et al., 1978) yet sheep consuming 2,000 mg Al/kg feed had depressed
appetites (Valdivia etal., 1982). There appears to be a threshold at which aluminum or
SZA begin to adversely affect animals.

The decreases in feed intake may be a result of the binding of aluminum to
phosphorus in the digestive system, essentially reducing dietary phosphorus to the
animal. High dietary aluminum interferes with phosphorus metabolism in animals by
forming unabsorbable complexes in the digestive tract (Deobald and Elvehjem, 1935). A
significant decrease in plasma phosphorus in response to supplementation of aluminum
has been seen in lambs (Rosa et al., 1982), beef cows (Allen et al., 1986) and chicks
(Hussein et al., 1990). The decrease usually occurs approximately 1 wk after
supplementation. The animals then begin to exhibit classic signs of phosphorus
deficiency, the most common deficiency in ruminants. The most common sign of
phosphorus deficiency is a reduction in feed intake (Call et al., 1987).

Alternatively, decreases in feed intake in previous equine studies on SZA

supplementation have not been reported (Nielsen et al., 1993; Lang et al., 2001b,c).

lll

Subsequently, SZA did not affect feed intake in the current study. Phosphorus deficiency
is rare in horses, and there is not much literature on the subject. However, no mention of
a decrease in feed intake has been indicated in association with phosphorus deficiency in
horses. Additionally, in contrast with ruminant literature, phosphorus absorption in the
horses was not affected by the SZA. Moreover, a decrease in plasma phosphorus
concentrations was also not seen in the horses, and the values were slightly higher than
the range previously given for horses of 30-70 mg/kg (Georgievskii, 1982).

The discrepancies between equine and bovine studies may come about because of
digestive tract differences. Besides a P requirement for the whole animal, there is an
additional P requirement for rumen microbes in ruminants which must be met for optimal
microbial activity (Bryant et al., 1957). As horses lack rumens, this would be a non-issue
in equine studies. However, it has been suggested that while a decrease in microbial
digestion occurs due to P-deficiency, the decrease in feed intake may actually be due to a
non-digestive tract disturbance such as a disturbance of intracellular metabolism due to
low blood P levels (Milton and Ternouth, 1985). If this were true, although horses do not
have rumens and would therefore not have ruminal microbial activities to maintain, they
might still have disturbed intracellular metabolism due to low P levels in the blood.

Perhaps the different responses of horses and ruminants to dietary SZA are due to
the level of phosphorus in the diet. The negative effects of high aluminum intakes can be
alleviated by supplementing dietary phosphorus. Increases in previously decreased
plasma inorganic phosphorus has been observed when phosphorus was supplied to the
animals (Lipstein and Hurwitz, 1982; Hussein et al., 1990). Decreases in gain and feed

intake in aluminum-supplemented lambs were reversed when the animals were fed

112

additional phosphorus (V aldivia et al., 1977; Rosa et al., 1982). It has been suggested by
Lipstein and Hurwitz (1982) that 0.76 g of supplemental phosphorus is required to offset
the effect of 1 g of aluminum. Animals that are borderline phosphorus deficient would
seem to be most at risk of the negative Al effects on P metabolism. The phosphorus
requirement for the light working horses in the study is 18 g (NRC, 1989). They were
receiving 27-39 g on a dry matter basis, well above their requirement. However, the
cows in the previous study (Turner, Chapter 4, Dissertation) had a requirement of 0.36%
P (NRC, 2001), and were receiving approximately 0.36% P in their diet. That may have
caused them to be more susceptible to high aluminum intakes because they had less “to
spare” then the horses. Therefore, the A1 would have bound up the cows’ phosphorus
and caused a deficiency while the horses were able to overcome the binding.

Few studies have focused on the effects of aluminum in horses. One study found
that there were no ill effects of feeding 1,370 mg Al/kg feed, but 4,500 mg/kg decreased
plasma phosphorus, and increased plasma and urinary calcium concentration in ponies
(Schryver et al., 1984). Roose et a1. (2000) found that feeding mature Thoroughbreds
either 160 mg Al/kg feed or 930 mg Al/kg feed produced no adverse effects on
phosphorus or calcium metabolism. Additionally, there were no changes in nutrient
digestibilities such as dry matter, crude protein, fiber and mineral digestibilities.

The SZA increased plasma concentrations of silicon in the SS horses which is a
common finding (Nielsen et al., 1993; Lang et al., 2001b,c; Mazella et al., 2005). The SS
horses had greater amounts of total Si excreted in both the urine and feces. The major
route of silicon excretion was through the urine, as is the case in humans (Reffitt et al.,

1999). However, there was no difference in either retention or absorption between the

113

CO and SS horses. The specific reason for the lack of an increase in Si absorption is
unknown. Perhaps there was not a need for the additional silicon, and therefore it was
not absorbed. Or perhaps some of the silicon bound to Al in the digestive tract, forming a
nonabsorbable compound, and thus the Si was excreted in the feces with the A1. A
relationship exists between silicon and aluminum such that the presence of silicon will
reduce aluminum absorption and toxicity. Orthosilicic acid reduces aluminum absorption
in humans (Edwardson et al., 1993).

The aluminum intake was quite high in the SS horses, but plasma aluminum
concentration in the SS horses was unaffected by the increase. The SS horses also tended
to retain more aluminum than the CO horses, although there was no difference in
digestibility. However, it cannot be assumed, since plasma aluminum levels and apparent
digestibility did not increase, that the ingested aluminum from SZA does not adversely
affect the horse. Plasma aluminum concentrations are not indicative of tissue storage
(Krueger et al., 1985), and increases in plasma aluminum are not always seen following
aluminum ingestion (Hussein etal., 1990). Calves supplemented with SZA did not have
an increase in plasma aluminum nor was there a difference in absorption, yet 12 of 13
tissues investigated had a 384% average increase in aluminum (Turner, Chapter 5,
Dissertation). Further research, such as tissue mineral metabolism effects, should be
conducted in the horse before concluding that aluminum from SZA does not adversely
affect horses.

Calcium metabolism can also be affected by high dietary intakes of aluminum. In
the current study, the addition of SZA increased retention and digestibility of calcium.

This effect has been seen in ruminants. Robinson et a1. (1984) found an increase in Ca

114

retention and availability when non-lactating cows were fed an additional 1000 mg/kg Al
in their feed for 20 days. Increases in serum Ca were also observed in lactating beef
cows receiving 1,730 mg Al/kg feed (Allen et al., 1986). Unlike in the previous lactating
cow study (Turner, Chapter 4, Dissertation), there was not an increase in plasma calcium
in the horses. The increase in blood Ca may be as a result of homeostatic measures in
response to the decreased P. Decreases in Ca concentration in the bone have been
reported after high Al supplementation (Allen, 1984). However, when additional
phosphorus was added to the diets, the increase in plasma Ca was less dramatic (Hussein
et al., 1990). Therefore, in the current study, unlike in the cow study, there was an
abundance of dietary phosphorus, thus blood calcium failed to increase. The plasma
calcium values were within normal ranges previously reported for the horse
(Georgievskii, 1982).

Fecal dry matter was greater in SS horses than CO horses. This has been seen in
calves as well (Turner, Chapter 5, Dissertation). There are numerous reports on the
addition of zeolites to the diets of ruminants, poultry, and swine decreasing incidence of
scours as well as other intestinal disorders (Mumpton, 1984). The exact role the zeolite
plays in reducing scours is unknown. One suggestion has been that the effects are due to
the zeolite’s alkalinity and buffering capacity in the gastrointestinal tract (Mumpton,
1984). Or perhaps the effect is through silicon, as silicon impacts the immune system
(Moseley et al., 1988; Seaborn et al., 2002) and thus might reduce scours and other
intestinal problems through this route.

In conclusion, the supplementation of 200 g of SZA to the horses did not

significantly affect phosphorus metabolism under the current study conditions.

115

Specifically, the drastic changes seen in ruminants, particularly that of phosphorus blood
concentration, did not occur in the horses. However, there was an increase in calcium
retention and digestion. Consideration should be given to phosphorus levels in the diet,
as the overabundance of P in the horses’ diets could have overridden the potential ill
effects of the aluminum. Conclusions about negative effects of aluminum on horses
cannot be made as plasma aluminum concentration and absorption are not indicative of

tissue mineral composition.

116

APPENDIX B

Although the calf nutritional study (Chapter 5) failed to produce differences in
bone mechanical properties, an interesting effect of exercise was found. The calves
received no formal exercise, and left their stalls only to be weighed once per week.
During the walk to the scale, which was approximately 75 m, the calves were allowed to
go at their own pace. Some calves ran while others walked the entire time. There was
one calf in particular that was much more active than the other calves, as noted by the
researchers. Although most calves were not named, the high activity level of this calf
earned him the name “Spitfire” within three days of his birth. It was interesting to note
that Spitfire had the greatest peak fracture force, ultimate bending stress and modulus of
elasticity of the MC 111 & IV. The calf with the second highest peak fracture force and
ultimate bending stress, and one of the highest modulus of elasticity was a calf that had
escaped fiom his stall for the day and had free access to the dairy farm. He was found at
the far end of the farm about 450 m away. The bone dimensions and mechanical
properties of the two calves with the highest activity levels were compared to the other
calves to identify other possible differences (Table 1B).

Spitfire and Runaway had shorter lateromedial and lateromedial medullary
diameters, but their lateral and medial cortical widths were slightly larger, if not the
same, as the other calves. Additionally, the medullary areas of Spitfire and Runaway’s
bones were substantially less than that of the other calves, even though the cortical area
was not that different (Runaway’s is practically identical). The total area also appears to

be less in the two active calves compared to the others. The percent cortical area of total

117

area was highest in Spitfire and Runaway. Therefore, it appears that exercise gives the
calves a more circular than oval medullary cavity and smaller total area with a greater
percentage of cortical area to total area. Thus, this yields bones with greater peak
fracture force, ultimate bending stress, and modulus of elasticity. Similar data were seen
in horses and calves. Hiney et al. (2004a) found that young horses sprinted 82 m/d for 5
d/wk for 56 d had decreased dorsopalmar medullary cavities and increased dorsal and
medial cortical widths compared to group-housed and stalled horses. In addition, young
calves exercised for a short duration had smaller medullary cavities and a larger
percentage of cortical bone area than calves confined to stalls (Hiney et al., 2004b). The
changes in bone led to a trend for increased fracture force of the bones. Therefore the
changes in bone exhibited by the two very active calves in the present study more than
likely caused the higher fracture forces.

The osteocalcin and deoxypyridinoline comparisons were also interesting Table
2B). Spitfire had less of a decrease in osteocalcin compared to the others, particularly
during the first 30 (1. Therefore, be possibly had greater bone formation than the others.
It is also interesting to note that he had a larger decrease in OC during (1 30-60 than the
rest of the calves. It may be due to the fact that Spitfire was getting larger, and therefore
had less space to jump around in his stall and his activity level had to decrease. The two
active calves also had substantially greater decreases in DPD than the other less active
calves, indicating less bone turnover. The comparisons made between the two active
calves and the other calves reinforce the idea that exercise is extremely important to bone

development, particularly early in life. It is also intriguing that 60 d of nutritional

118

supplementation failed to produce results, while bouts of sporadic exercise seemed to

affect the bone.

Table 18. Differences in bone dimensions and mechanical properties between two
highly active calves and the other calves.

 

Spitfire Runaway Others (n = 18)
Diameters (cm)
Dorsopalmar 1.86 1.91 2.00 (1.77-2.15)
Dorsopalmar medullary 1.10 1.16 1.28 (1.06-1.45)
Lateromedial 2.45 2.56 2.72 (2.45-2.90)
Lateromedial medullary 1.49 1.60 1.83 (1.49-2.07)
Cortical widths (cm)
Dorsal 0.40 0.39 0.37 (0.32-0.43)
Palmar 0.35 0.39 0.34 (0.29-0.39)
Lateral 0.47 0.49 0.45 (0.37-0.53)
Medial 0.49 0.53 0.46 (0.40-0.54)
Areas (mm2)
Cortical area 279 307 306 (269-338)
Medullary area 128 140 175 (128-232)
Total area 407 447 481 (404-538)
% cortical of total area 68.5 68.7 63.8 (56.8-68.7)
Mechanical properties
Peak fracture force (N) 5085 4715 4224 (3630-5085)
Ultimate bending stress
(MPa) 145 145 110 (87-145)
Modulus of elasticity (MPa) 0.64 0.55 0.44 (0.30-0.64)

 

119

Table ZB. Comparisons of osteocalcin and deoxypyridinoline between two highly
active calves and the other calves.

 

Bone markers (ng/ml) Spitfire Runaway Others (11 = 18)
OC (1 0 337 527 407 (100-738)
OC (1 30 294 412 284 (128-412)
OC (1 60 202 312 233 (152-324)
DPD d 0 18.2 20.4 16.1 (11.4-24.6)
DPD d 30 11.0 14.0 11.3 (6.4-16.2)
DPD d 60 10.4 11.9 11.3 (7.4-17.0)

Net changes
OC from d 0 to 30 -43 -115 -123 (-440-150)
OC fromd30 to 60 -92 -100 -51(-113-117)
OC from d 0 to 60 -l35 -215 -l74 (427-144)
DPD from d 0 to 30 -7.2 -6.4 -4.8 (-13.5-1.8)
DPD from d 30 to 60 -0.6 -2.1 0.0 (-2. l-2.4)
DPD from d 0 to 60 -7.8 -8.5 -4.8 (-l4.2-2.8)

 

120

APPENDIX C

Milk Replacer Data for Calf Study

Instant Cow’s Match® NT Medicated
Manufactured by: Land O’Lakes Animal Milk Product Co., Shoreview, MN

Active drug ingredients
Oxytetracycline .............................................................................. 100 g/ton
Neomycin base (from neomycin sulfate) ................................................ 200 g/ton

Guaranteed analysis

Crude Protein ........................................................................... MIN 28.00%
Crude Fat ................................................................................. MIN 20.00%
Crude Fiber .............................................................................. MAX 0.15%
Calcium ................................................................................... MIN 0.75%

Calcium .................................................................................. MAX 1.25%
Phosphorus ............................................................................... MIN 0.70%
Vitamin A ......................................................................... MIN 10,000 IU/lb
Vitamin D3 ........................................................................... MIN 2,500 IU/lb
Vitamin E ............................................................................... .MIN 50 IU/lb

Ingredients

Dried whey, dried why protein concentrate, dried whey product, dried skimmed milk,
dried milk protein, animal fat (preserved with ethoxyquin), lecithin, polyethylene glycol
(400) monooleate, dicalcium phosphate, calcium carbonate, methionine supplement, 1-
lysine, vitamin A acetate, vitamin D3 supplement, vitamin E supplement, thiamine
mononitrate, riboflavin, niacin supplement, calcium pentothenate, biotin, ascorbic acid,
pyridoxine hydrochloride, folic acid, vitamin B12 supplement, choline chloride, roughage
products, calcium silicate, manganese sulfate, zinc sulfate, ferrous sulfate, copper sulfate,
cobalt sulfate, copper sulfate, cobalt sulfate, ethylenediamine dihydriodide and sodium

selenite.

121

LITERATURE CITED

Abreo, K., J. Glass, and M. Sella. 1990. Aluminum inhibits hemoglobin synthesis but
enhances iron uptake in Friend erythroleukemia cells. Kidney Int. 37:677-681.

Adler, A.J., Z. Etzion, and GM. Berlyne. 1986. Uptake, distribution, and excretion of
31silicon in normal rats. Am. J. Physiol. 2512E670-E673.

Allen, M.J., W.E. Hoffmann, D.C. Richardson, and G.J. Breur. 1998. Serum markers of
bone metabolism in dogs. Amer. J. Vet. Res. 59:250-254.

Allen, V.G., and D.L. Robinson. 1980. Occurrence of aluminum and manganese in grass
tetany cases and their effects on solubility of calcium and magnesium. Agron. J.
72:957-962.

Allen, V.G. 1984. Influence of dietary aluminum on nutrient utilization in ruminants. J.
Anim. Sci. 59:836-844.

Allen, V.G., and J .P. F ontenot. 1984. Influence of aluminum as sulfate, chloride and
citrate on magnesium and calcium metabolism in sheep. J. Anim. Sci. 59:798-
803.

Allen, V.G., D.L. Robinson, and PG. Hembry. 1984. Effects of ingested aluminum
sulfate on serum magnesium and the possible relationship to hypomagnesemic
tetany. Nutr. Rep. Int. 29: 107-1 15.

Allen, V.G., F .P. Horn, and J .P. Fontenot. 1986. Influence of ingestion of aluminum,
citric acid and soil on mineral metabolism of lactating beef cows. J. Anim. Sci.
62: 1396-1403.

Alvarado, A.F., M. Marcoux, and L. Breton. 1989. The incidence of osteochondrosis in a
Standardbred breeding farm in Quebec. Proc. 35th Ann. Conv. AAEP. 293-307.

Anderson, RR. 1989. Comparison of major minerals and trace elements in cattle, guinea
pig and human milk. J. Dairy Sci. 72:312-313.

Archibald, J .G., and H. Fenner. 1957. Silicon in cow’s milk. J. Dairy Sci. 40:703-706.

Arieff, AL 1990. Aluminum and the pathogenesis of dialysis dementia. Environ.
Geochem. Health 12:89-93.

ASAE. 2000. Shear and three-point bending test of animal bone. ASAE STANDARDS.
ANSI/ASAE Standard S459. Amer. Soc. Agri. Eng. St Joseph, MI.

122

Balazs, EA. 1982. The physical properties of synovial fluid and the special role of
hyaluronic acid. In: Disorders of the knee. A.J. Helfet, ed. Lippincott,
Philadelphia, PA. P. 61-75.

Ballard, R., and HM. Edwards. 1988. Effects of dietary zeolite and vitamin A on tibial
dyschondroplasia in chickens. Poult. Sci. 67:113-119.

Bashir, A., M.L. Gray, and D. Burstein. 1996. Gd-DTPAZ' as a measure of cartilage
degradation. Magn. Reson. Med. 36:665-673.

Bashir, A., M.L. Gray, J. Hartke, and D. Burstein. 1999. Nondestructive imaging of
human cartilage glycosaminoglycan concentration by MRI. Magn. Reson. Med.
41 :857-865.

Bellia, J.P., J .D. Birchall, and NB. Roberts. 1996. The role of silicic acid in the renal
excretion of aluminum. Ann. Clin. Lab. Sci. 26:227-233.

Birchall, J .D. 1991. The toxicity of aluminum and the effect of silicon on its

bioavailability. In: Aluminum in chemistry, biology, and medicine. M. Nicolini,
P.F. Zatta and B. Corain. Eds. pp. 53-69.

Birchall, J .D. 1992. The interrelationship between silicon and aluminum in the biological
effects of aluminum. In: Aluminum in biology and medicine. Ciba Foundation
Symposium. 50-68.

Birchall, J .D., and J .S. Chappell. 1989. Aluminum, water chemistry, and Alzheimer’s
disease. Lancet. i:953.

Birchall. J .D., C. Exley, J .S. Chappell, and M.J. Phillips. 1989. Acute toxicity of
aluminum to fish eliminated in silicon-rich acid waters. Nature. 338: 146-148.

Boyce, B.F., H.Y. Elder, H.L. Elliot, 1. Fogelman, G.S. Fell, B.J. Junor, G. Beastall, and
LT. Boyle. 1982. Hypercalcaemic osteomalacia due to aluminum toxicity. Lancet
2:1009-1012.

Bryant, M.P., I.M. Robinson, and H. Chu. 1957. Observations on the nutrition of
bacteroides succinogenes — a ruminal cellulolytic bacterium. J. Dairy Sci.
42:1831-1847.

Burstein, D., and M. Gray. 2003. Potential of molecular imaging of cartilage. Sports
Med. Arthro. 11:182-191.

Buys, SS, and J .P. Kushner. 1989. Hematologic effects of aluminum toxicity. In:
Aluminum and health. A critical review. H.J. Gitelmaned. pp. 235-256.

123

Call, J .W., J .E. Butcher, J .L. Shupe, R.C. Lamb, R.L. Boman, and AB. Olson. 1987.
Clinical effects of feeding low dietary phosphorus concentrations in feed to
lactating dairy cows. Am. J. Vet. Res. 48:133-136.

Calomme, M.R., and DA. Vanden Berghe. 1997. Supplementation of calves with
stabilized orthosilicic acid. Effect on the silicon, Ca, Mg, and P concentrations in
serum and the collagen concentration in skin and cartilage. Biol. Trace Elem.
Res. 56:153-165.

Cannata, J .B., I. Fernandez-Soto, M.S. F emandez-Menendez, J .L. Femandez-Martin, J .R.
Mosquera, S.J. Macgregor, J .M. Brock, and D. Halls. 1991. Role of iron
metabolism in absorption and cellular uptake of aluminum. Kidney Int. 39:799-
803.

Carlisle, EM. 1970. Silicon: A possible factor in bone calcification. Science 167:279.

Carlisle, EM. 1972. Silicon: An essential element for the chick. Science. 178:619-621.

Carlisle, EM. 1974. Silicon as an essential element. Fed. Proc. 33:1758-1766.

Carlisle, EM. 1976. In vivo requirement for silicon in articular cartilage and connective
tissue formation in the chick. J. Nutr. 106:478-484.

Carlisle, E.M. 1980a. A silicon requirement for normal skull formation. J. Nutr. 10:352-
359.

Carlisle, E.M. 1980b. Biochemical and morphological changes associated with long bone
abnormalities in silicon deficiencies. J. Nutr. 10: 1046-1056.

Carlise, EM. 1981. Silicon: a requirement in bone formation independent of vitamin D1.
Calcif. Tissue. Int. 33:27-34.

Carlisle, EM. 1982. The nutritional essentiality of silicon. Nutr. Rev. 40:193-198.

Carlisle, EM. 1984. Silicon. In: Biochemistry of the essential ultratrace elements. Earl
F rieden (Ed). New York. 257-291.

Carlisle, E.M., and M.J. Curran. 1987. Effect of dietary silicon and aluminum on silicon
and aluminum levels in rat brain. Alzheimer Dis. Assoc. Disord. 1:83-89.

Chary-Valckenaere, I., P. Fener, J .Y. Jouzeau, P. Netter, E. Payan, J. Floquet, D. Bumel,
M. Kessler, J. Pourel, and A. Gaucher. 1994. Experimental articular toxicity of
aluminum compounds in vivo. J. Rheumatol. 21 :1542-1547.

Chazan, J .A., N.P. Libbey, M.R. London, L. Pono, and JG. Abuelo. 1991. The clinical
spectrum of renal osteodystrophy in 57 chronic hemodialysis patients: A

124

correlation between biochemical parameters and bone pathology findings. Clin.
Nephrol. 35:78-85.

Christenson, RH. 1997. Biochemical markers of bone metabolism: an overview. Clin.
Biochem. 30:573-593.

Clarkson, E.M, V.A. Luck, W.V. Hynson, R.R. Bailey, J .B. Eastwood, J.S. Woodhead,
V.R. Clements, J.L.H. O’Riordan, and HE. De Wardener. 1972. The effect of
aluminum hydroxide on calcium, phosphorus and aluminum balances, the serum

parathyroid hormone concentration and the aluminum content of bone in patients
with renal failure. Clin. Sci. 43:519-531.

Coleman, P.J., D. Scott, R.M. Mason, and JR. Levick. 1999. Characterization of the
effect of high molecular weight hyaluronan on trans-synovial flow in rabbit knees.
J. Physiol. 514:265-282.

Crowe, N.A., M.W. Neathery, W.J. Miller, LA. Muse, C.T. Crowe, J .L. Vamadoe, and
D.M. Blackmon. 1990. Influence of high dietary aluminum on performance and
phosphorus bioavailability in dairy calves. J. Dairy Sci. 73:808-818.

Dalin, G., B. Sandgren, and J. Carlsten. 1993. Plantar osteochondral fragments in the
metatarsophalangeal joints in Standardbred trotters; result of osteochondrosis or
trauma? Equine Vet. J. Suppl. 16:62-65.

Deobald, H.J., and CA. Elvehjem. 1935. The effects of feeding high amounts of soluble
iron and aluminum salts. Am. J. Physiol. 111:118-123.

Drape, J .L., P. Thelen, P. Gay-Depassier, O. Silbennann, and R. Benacerraf. 1993.
Intraarticular diffusion of Gd-DOTA after intravenous injection in the knee: MR
imaging evalution. Radiol. 188:227-234.

Edwardson, J .A., P.B. Moore, I.N. Ferrier, J.S. Lilley, G.W.A. Newton, J. Barker, J.
Templar, and J .P. Day. 1993. Effect of silicon on gastrointestinal absorption of
aluminum. Lancet. 342:21 1-212.

Eisinger, J ., and D. Clairet. 1993. Effects of silicon, fluoride, etidronate and magnesium
on bone mineral density: a retrospective study. Magnes. Res. 6:247-249.

Enemark, J .M.D., A.M. Spangaard Frandsen, T. Thilsing-Hansen and R.J. Jorgensen.
2003. Aspects of physiological effects of sodium zeolite A supplementation in
dry, non-pregnant dairy cows fed grass silage. Acta Vet. Scand. Suppl. 97:97-117.

Frey K.S., G.D. Potter, T.W. Odom, D.M. Senor, V.D. Reagan, V.H. Weir, J. Elslander,
S.P Webb, E.L. Morris, W.B. Smith, and KB. Weigand. 1992. Plasma silicon and
radiographic bone density in weanling Quarter Horses fed sodium zeolite A. J.
Equine Vet. Sci. 12:292-296.

125

Garbossa, G., G. Gélvez, G. Pérez, J. Stripeikis, M. Tudino M., and A. Nesse. 1998. Oral
aluminum administration to rats with normal renal function. 2. Body distribution.
Human Experim. Toxicol. 17:318-322.

Georgievskii, VI, 1982. Mineral composition of bodies and tissues of animals. In:
Mineral nutrition of animals. Eds. Georgievskii, Annenkov, and Samokhin.

Goering, H.K., and P]. Van Soest. 1970. Forage and fiber analysis. Agricultural
Handbook no. 379. US. Dept. of Agriculture.

Goodman, W.G. 1986. Experimental aluminum-induced bone disease: Studies in vivo.
Kidney Int. 29:S32-S36.

Green, L. E., V. J. Hedges, Y. H. Schukken, R. W. Blowey, and A. J. Packington. 2002.
The impact of clinical lameness on the milk yield of dairy cows. J. Dairy Sci.
85:2250-2256.

Grondahl, AM. 1992. The incidence of bony fragments and osteochondrosis in the
metacarpo- and metatarsophalangeal joints of Standardbred trotters. Equine Vet.
Sci. 12:81-85.

Guo, C.-H., G.-S. W. Hsu, L.-Y. Lin, Y.-H. Wang, C.-Y. Lin, and M.-S. Yeh. 2004.
Distribution patterns of trace metals and of lipid peroxidation in plama and
erythrocytes of rat exposed to aluminum. Biol. Trace Elem. Res. 101:61-71.

Han, J ., J. Han, M.A. Dunn. 2000. Effect of dietary aluminum on tissue nonheme iron
and ferritin levels in the chick. Toxic. 142:97-109.

Hauptkorn, S., J. Pavel, and H. Seltner. 2001 Determination of silicon in biological
samples by ICP-OES after non-oxidative decomposition under alkaline
conditions. Fres. J. Anal. Chem. 370:246-250.

Hills, BA, and MK. Monds. 1998. Enzymatic identification of the load-bearing
boundary lubricant in the joint. Br. J. Rheu. 37: 137-142.

Hiney, K.M., B.D. Nielsen, and D.S. Rosenstein. 2004a. Short-duration exercise and
confinement alter bone mineral content and shape in weanling horses. J. Anim.
Sci. 82:2313-2320.

Hiney, K.M., B.D. Nielsen, D.S. Rosenstein, M.W. Orth, and HP. Marks. 2004b. High-

intensity exercise of short duration alters bovine density and shape. J. Anim. Sci.
82:1612-1620.

126

Hussein, A.S., A.H. Cantor, T.H. Johnson, and R.A. Yokel. 1990. Relationship of dietary
aluminum, phosphorus, and calcium to phosphorus and calcium metabolism and
growth performance of broiler chicks. Poult. Sci. 69:966-971.

J effcott, LB. 1991. Osteochondrosis in the horse — searching for the key to pathogenesis.
Equine Vet. J. 23(5):331-338.

J effcott, LB, and F.M.D. Henson. 1998. Studies on growth cartilage in the horse and

their application to the aetiopathogenesis of dyschondroplasia (osteochondrosis).
Vet. J. 156:177-192.

Jeffrey, E.H., K. Abreo, E. Burgess, J. Cannata, and J .L. Greger. 1996. Systemic
aluminum toxicity: effects on bone, hematopoietic tissue, and kidney. J. Toxicol.
Environ. Health. 48:649-665.

Johnson, M. A., T. F. Sweeney, and L. D. Muller. 1988. Effects of feeding synthetic
zeolite A and sodium bicarbonate on milk production, nutrient digestion, and rate
of digesta passage in dairy cows. J. Dairy Sci. 71 :946-953.

J ope, RS, and G.V.W Johnson. 1992. Neurotoxic effects of dietary aluminum. In:
Aluminum in biology and medicine. Ciba Foundation Symposium. 254-267.

Jugdaohsingh, R., D.M. Reffitt, C. Oldham, J.F. Day, L.K. Fifield, R.P.H. Thompson,
and J .J . Powell. 2000. Oligomeric but not monomeric silica prevents aluminum
absorption in humans. Am. J. Clin. Nutr. 71:944-949.

Karn, J .F . 2001. Phosphorus nutrition of grazing cattle: a review. Anim. Feed Sci. Tech.
89: 133-153.

King, S.L., J .P. Day, C. Oldham, J .F . Popplewell, P. Ackrill, RE. Moore, G.A. Taylor,
J .A. Edwardson, L.K. Fifield, K. Liu, and R.G. Cresswell. 1997. The influence of
dissolved silicate on the physiological chemistry of aluminum, studied in humans

using tracer 26Al and accelerator mass spectrometry. Nuc. Instr. Meth. Physics
Res. B. 123:254-258.

Koohmaraie, M., A.S. Babiker, A.L. Schroeder, R.A. Merkel, and TR. Dutson. 1988.
Acceleration of post-mortem tenderization in ovine carcasses through activation

of Ca2+-dependent proteases. J. Food Sci. 53:1638-1641.

Kossaibati, M.A., and R.J. Esslemont. 1997. The costs of production diseases in dairy
herds in England. Vet. J. 154(1):4l-51.

Krueger, G.L., T.K. Morris, R.R. Suskind, and EM. Widner. 1985. The health effects of
aluminum compounds in mammals. CRC Crit. Rev. Tox. 13:1-24.

127

Lang, K.J., B.D. Nielsen, C.I. O'Connor and KL. Waite. 20013. Comparison of blood
sample collection methods and their effect on plasma silicon concentrations.
Proc. 17 Equine Nutr. Phys. Symp. p. 375-376.

Lang, K.J., B.D. Nielsen, K.L. Waite, G.M. Hill, and M.W. Orth. 2001b. Supplemental
silicon increases plasma and milk silicon concentrations in horses. J. Anim. Sci.
79:2627-2633.

Lang, K.J., B.D. Nielsen, K.L. Waite, J. Link, G.M. Hill, and M.W. Orth. 2001c.
Increased plasma silicon concentrations and altered bone resorption in response to

sodium zeolite A supplementation in yearling horses. J. Equine Vet. Sci.
21(11):550-555.

Lillich J .D., A.L. Bertone, C.J. Malemud, S.E. Weisbrode, A.J. Ruggles, and SS.
Stevenson. 1998. Biochemical, histochemical and immunochemical

characterization of distal tibial osteochondrosis in horses. Am. J. Vet. Res. 58:89-
98.

Lipstein, B., and S. Hurwitz. 1982. The effect of aluminum on the phosphorus
availability in algae-containing diets. Poult. Sci. 61 :951-954.

Manohar, M., E. Coney, and E. Hutchens. 1993. Pulmonary vascular pressure in near-
maximally exercised horses. Equine Pract. 15:15-21.

Martyn, C.N., D.J.P. Barker, C. Osmond, E.C. Harris, J .A. Edwardson, and RF. Lacey.
1989. Geographical relation between Alzheimer’s disease and aluminum in
drinking water. Lancet. 1:59—62.

Mazzella, G., R. Godbee, W. Schurg, and M. Ams. 2005. Plasma silicon concentrations
in weanling horses fed sodium zeolite A or azomite A. Proc. Equine Sci. Soc.
Tucson, AZ. May 3l-June2, 2005. p. 353.

McCollum, E.V., O.S. Rask, and J .E. Becker. 1982. A study of the possible role of
aluminum compound in animal and plant physiology. J. Biol. Chem. 77:753-764.

Milton, J .T.B., and J .H. Temouth. 1985. Phosphorus metabolism in ruminants. 11. Effects
of inorganic phosphorus concentration upon feed intake and digestibility. Aust. J.
Agric. Res. 36:647-654.

Minami T., S. Tohno, M. Utsumi, Y. Moriwake, M.-O. Yamada, and Y. Tohno. 2001.
Selective accumulations of aluminum in five human arteries. Biol. Trace. Elem.
Res. 79:29-38.

Montgomery J .L., J .R. Blanton, R.L. Horst, K.J. Morrow, V.G. Allen, D.B. Wester, and

M.F. Miller. 2004. Effect of supplemental vitamin D3 concentration on
concentrations of calcium, phosphorus, and magnesium relative to protein in

128

subcellular components of the Iongissimus and the distribution of calcium within
longissimus muscle of beef steers. J. Anim. Sci. 82:2742-2749.

Moseley, P.L., M. Monick, and G.W. Hunninghake. 1988. Divergent effects of silica on
lymphocyte proliferation and immunoglobulin production. J. Appl. Physiol.
65:350-357.

Mumpton, EA. 1984. The role of natural zeolites in agriculture and aquaculture. In: Zeo-
Agriculture: Use of natural zeolites in agriculture and aquaculture. Eds. W.G
Pond and FA. Mumpton. P. 3-27.

Mumpton, F.A., and RH. Fishman. 1977. The application of natural zeolites in animal
sciences. J. Anim. Sci. 45: 1 188-1 194.

Nayak, P. 2002. Aluminum: impacts and disease. Envir. Res. Sec. A 89:101-115.

Neathery, M.W., N.A. Crowe, W.J. Miller, C.T. Crowe, J .L. Vamadoe, and D.M.
Blackmon. 1990a. Effects of dietary aluminum and phosphorus on magnesium
metabolism in dairy calves. J. Anim. Sci. 68:1133-1138.

Neathery, M.W., N.A. Crowe, W.J. Miller, C.T. Crowe, J .L. Vamadoe, and D.M.
Blackmon. 1990b. Influence of dietary aluminum and phosphorus on zinc
metabolism in dairy calves. J. Anim. Sci. 68:4326-4333.

Nielsen, B.D., G.D. Potter, E.L. Morris, T.W. Odom, D.M. Senor, J .A. Reynolds, W.B.
Smith, M.T. Martin, and EH. Bird. 1993. Training distance to failure in young
racing Quarter Horses fed sodium zeolite A. J. Equine Vet. Sci. 13:562-567.

Nielsen, PH. 1991. Nutritional requirements for boron, silicon, vanadium, nickel, and
arsenic: current knowledge and speculation. FASEB J. 5:2661-2667.

Nielsen, F .H. and R. Poellot. 2004. Dietary silicon affects bone turnover differently in
ovariectomized and sham-operated growing rats. J. Trace Elem. Exper. Med.
17: 137-149.

NRC, 1989. Nutrient requirements of horses. 5th ed. National Academies Press,
Washington, DC.

NRC, 2001. Nutrient requirements of dairy cattle. 7"n ed. National Academies Press.
Washington, DC.

Parantainen, J ., E. Tenhunen, R. Kangasniemi, S. Sankari, and F. Atroshi. 1987. Milk and
blood levels of silicon and selenium status in bovine mastitis. Vet. Res. Comm.

1 1:467-477.

Persson, L. 1971. On the synovia in horses. Acta Vet. Scand. Suppl. 35: 1-77.

129

Price, F .M., J .R. Levick, and RM. Mason. 1996. Glycosaminoglycan concentration in
synovium and other tissues of rabbit knee in relation to hydraulic resistance. J.
Physiol. 495:803-820.

Rajala—Schultz, P. J ., Y. T. Grohn, and C. E. McCulloch. 1999. Effects of milk fever,
ketosis, and lameness on milk yield in dairy cows. J. Dairy Sci. 82:288-294.

Reeves, P.G., F .H. Nielsen, and GO Fahey Jr. 1993. AIN-93 purified diets for laboratory
rodents: final report of the American Institute of Nutrition ad hoc writing
committee on the reformulation of the AIN-76A rodent diet. J. Nutr. 123:1939-
1951.

Reffitt, D.M., N. Ogston, R. Jugdaohsingh, H.F.J. Cheung, B.A.J. Evans, R.P.H.
Thompson, J .J . Powell, and G.N. Hampson. 2003. Orthosilicic acid stimulates
collagen type I synthesis and osteoblastic differentiation in human osteoblast-like
cells in vitro. Bone. 32:127-135.

Reffitt, D.M., R. Jugdaohsingh, R.P.H. Thompson, and J .J . Powell. 1999. Silicic acid: its
gastrointestinal uptake and urinary excretion in man and effects on aluminum
excretion. J. Inorg. Biochem. 76:141-147.

Rico H., J .L. Gallego-Lago, E.R. Hernandez, L.F. Villa, A. Sanchez-Atrio, C. Seco, and
J .J . Gervas. 2000. Effect of silicon supplement on osteopenia induced by
ovariectomy in rats. Calc. Tiss. Int. 66:53-55.

Rimal, B., A.K. Greenberg, and W.M. Rom. 2005. Basic pathogenetic mechanisms in
silicosis: current understanding. Curr. Opin. Pulm. 1 1:169-73.

Rincker, M.J., G. M. Hill, J. E. Link, and J. E. Rowntree. 2004. Effects of dietary iron
supplementation on growth performance, hematological status, and whole-body
mineral concentrations of nursery pigs. J. Anim. Sci. 82:3189-3197.

Robinson, D.L., O.J. Hemkes, and A. Hemp. 1984. Relationships among forage

aluminum levels, soil contamination on forages, and availability of elements to
dairy cows. Neth. J. Ag. Sci. 32:73-80.

Roose, K.A., K.E. Hoekstra, J .D. Pagan, and R.J. Geor. 2000. Effect of an aluminum
supplement on nutrient digestibility and mineral metabolism in Thoroughbred
horses. Proc. Equine Nutr. Conf. May 4-5, Lexington, Ky. p. 217-224.

Rosa, I.V., P.R. Henry, and CB. Ammennan. 1982. Interrelationship of dietary

phosphorus, aluminum, and iron on performance and tissue mineral composition
in lambs. J. Anim. Sci. 55:1231-1240.

130

Roussel, J. D., J. K. Thibodeaux, R. W. Adkinson, G. M. Toups, and L. L. Goodeaux.
1992. Effect of feeding various levels of sodium zeolite A on milk yield, milk

composition and blood profiles in thermally stressed Holstein cows. Int. J. Vit.
Nutr. Res. 62:91-98.

Sandgren, B., G. Dalin, and J. Carlsten. 1993. Osteochondrosis in the tarsocrural joint
and osteochondral fragments in the fetlock joints in Standardbred trotters. I.
Epidemiology. Equine Vet. J. Suppl. 16:31-37.

Schryver, H.F., D.L. Millis, L.V. Soderholm, J. Williams, and HF. Hintz. 1986.
Metabolism of some essential minerals in ponies fed high levels of aluminum.
Cornell Vet. 76:354-360.

Schutze, N., M.J. Oursler, J. Nolan, B.L. Riggs, and T.C. Spelsberg. 1995. Zeolite A
inhibits osteoclast-mediated bone resorption in vitro. J. Cell. Biochem. 58:39-46.

Schwarz, K. 1973. A bound form of silicon in glycosaminoglycans and polyuronides.
Proc. Nat. Acad. Sci. 70:1608-1612.

Scott, D., P.J. Coleman, R.M. Mason, and J .R. Levick. 2000. Concentration dependence
of interstitial flow buffering by hyaluronan in synovial joints. Microvasc. Res.
59:345-353.

Seabom, CD, and EH. Nielsen. 1993. Silicon: a nutritional beneficence for bones, brain
and blood vessels? Nutr. Today. July/Aug: 13-18.

Seabom, CD, and F .H. Nielsen. 2002a. Silicon deprivation decreases collagen formation
in wounds and bone, and omithine transaminase enzyme activity in liver. Biol.
Trace Elem. Res. 89:251-261.

Seabom, CD, and F .H. Nielsen. 2002b. Silicon deprivation and arginine and cystine
supplementation affect bone collagen and bone and plasma trace mineral
concentrations in rats. J. Trace Elem. Exper. Med. 15:113-122.

Seabom, C.D., M. Briske-Anderson, and F .H. Nielsen. 2002. An interaction between
dietary silicon and arginine affects immune function indicated by Con-A-induced
DNA synthesis of rat splenic T-lymphocytes. Biol. Trace Elem. Res. 87:133-142.

Shaw, D.T., D. W. Rozeboom, G. M. Hill, A. M. Booren, and J. E. Link. 2002. Impact of
vitamin and mineral supplement withdrawal and wheat middling inclusion on
finishing pig growth performance, fecal mineral concentration, carcass

characteristics, and the nutrient content and oxidative stability of pork. J. Anim
Sci. 80:2920-2930.

131

Spencer, H., L. Kramer, C. Norris, and D. Osis. 1982. Effect of small doses of aluminum-
containing antacids on calcium and phosphorus metabolism. Am. J. Clin. Nutr.
36:32-40.

Stallings, CC, and KP. Knowlton. 1995. Strategies to reduce amounts of nitrogen and
phosphorus in dairy rations. Virginia Cooperative Extension publication 404-130,
p. 1-3.

Sweeney, TR, A. Cervantes, L.S. Bull, and R.W. Hemken. 1984. Effects of dietary
clinotilolite on digestion and rumen fermentation in steers. In: Zeo-agriculture
use of natural zeolites in agriculture and aquaculture. W.G. Pond and F .A.
Mumpton, eds. Westview Press, Boulder, CO. p. 177-187.

Temouth, J .H., and CC. Sevilla. 1990. The effects of low levels of dietary phosphorus
upon the dry matter intake and metabolism of lambs. Aust. J. Agric. Res. 41 : 175-
184.

Thilsing-Hansen, T., R.J. Jorgensen, J .M. Enemark, and T. Larsen. 2002. The effect of
zeolite A supplementation in the dry period on periparturient calcium,
phosphorus, and magnesium homeostasis. J. Dairy Sci. 85(7): 1855-1862.

Thilsing-Hansen, T., R.J. Jorgensen, J .M. Enemark, R. Zelvyte, and A. Sederevicius.
2003. The effect of zeolite A supplementation in the dry period on blood mineral
status around calving. Acta vet. Scand. Suppl. 97:87-95.

Thilsing—Hansen, T., and R.J. Jorgensen. 2001. Hot topic: prevention of parturient paresis
and subclinical hypocalcemia in dairy cows by zeolite A administration in the dry
period. J. Dairy Sci. 84:691-693.

Thompson, A., S.L. Hansard, and M.C. Bell. 1959. The influence of aluminum and zinc
on upon the absorption and retention of calcium and phosphorus in lambs. J.
Anim. Sci. 18:187-194.

Tohno, S., Y. Tohno, T. Minami, Y. Moriwake, C. Azuma, and Y. Ohnishi. 2002.
Relationships among element contents in the internal jugular vein similar to the
arteries. Biol. Trace Elem. Res. 88:223-233.

Underwood, B.J. 1981. Mineral nutrition of livestock. 2nd edition.

Valdivia, R., C.B. Ammerman, C.J. Wilcox, and RR. Henry. 1978. Effect of dietary
aluminum on animal performance and tissue mineral levels in growing steers. J.
Anim. Sci. 47:1351-1356.

Valdivia, R., C.B. Ammerman, P.R. Henry, J .P. Feaster, and OJ. Wilcox. 1982. Effect of

dietary aluminum and phosphorus on performance, phosphorus utilization and
tissue mineral composition in sheep. J. Anim. Sci. 55:402-410.

132

van den Hoogen, B.M., C.H.A. van de Lest, P.R. van Weeren, L.M.G. van Golde, and A.
Bameveld. 1999. Changes in proteoglycan metabolism in osteochondrotic
articular cartilage of growing foals. Equine Vet. J. 31:38-44.

van der Voet, GB. 1992. Intestinal absorption of aluminum. InzAluminum in biology and
medicine. Symp. Alum. Biol. Med., Ciba Foundation, London, Nov .1991. p. 109-
122.

Vuillermoz, B., Y. Wegrowski, J .-L. Contet-Audonneau, L. Danoux, G. Pauly, and F.X.
Maquart. 2005. Influence of aging on glycosaminoglycans and small leucine-rich
proteoglycans production by skin fibroblasts. Molec. Cell. Biochem. 277:63-72.

Ward, T.L., K.L. Watkins, L.L. Southern, P.G. Hoyt, and DD. French. 1991. Interactive
effects of sodium zeolite-A and copper in growing swine: growth, and bone and
mineral tissue concentrations. J. Anim. Sci. 69:726-733.

West, J .B., O. Mathieu-Costello, J .H. Jones, E.K. Birks, R.B. Logemann, J .R. Pascoe,
and W.S. Tyler. 1993. Stress failure of pulmonary capillaries in racehorses with
exercise-induced pulmonary hemorrhage. J. Appl. Physiol. 75: 1097-1 109.

West, J .B., K. Tsukimoto, O. Mathieu-Costello, and R. Prediletto. 1991. Stress failure in
pulmonary capillaries. J. Appl. Physiol. 70: 1731-1742.

Wettstein, A., J. Aeppli, K. Gautschi, and M. Peters. 1991. Failure to find a relationship
between aluminum in drinking water and dementia. Int. Arch. Occup. Environ.
Health 63:97-103.

Winalski, C., P. Aliabadi, R.J. Wright, S. Shortkroff, C.B. Sledge, and B. Weissman.

1993. Enhancement of joint fluid with intravenously administered gadopentetate
dimeglumine: technique, rationale, and implications. Radio]. 187:179-185.

133

VITA

Kari Krick Turner was born on March 27th, 1978 in West Palm Beach, FL to Peter
and Sara Krick. Upon graduation from Wellington High School in 1996, Kari attended
the University of Florida. She obtained her Bachelor of Science degree in Animal
Science and a Minor in Agribusiness Sales and Management in 2000. Kari then received
a John Lee Pratt Fellowship in Animal Nutrition to attend Virginia Polytechnic Institute
and State University for a Master of Science degree with Dr. David Kronfeld. Her
research focused on immunoglobulins in endurance horses as affected by exercise, and a
possible selective IgA deficiency which had previously not been identified in horses.
After completing her M.S. degree in Animal Science in 2002, Kari began her doctoral
work at Michigan State University under the direction of Dr. Brian Nielsen. Her research
focused on sodium zeolite A and its effects on mineral metabolism and bone/joint
physiology in dairy cows, calves and horses. Following completion of her Ph.D. Kari

will begin work at the University of Georgia as an assistant professor of equine science.

134

   

milliljllllgljjllljjljl