.
.t.

x
.i ..
{ﬂ
ﬂ

2*»fo A
#11}:
: v .I

I .

in:
“a 3...:
i...

“i, .

534v!

Ell.- I35

:2}: a
a... '. ﬁ
3'

00
‘. Ii: .
Ill.

 

7
.21.!1‘ .l.$

#1 In... ‘11}. § .1

: lx..h.y...n|. itlr 3f
nlfvctl

QUE-3.3. .1 it...
C .szfrl-” 1 .109—

 

 

 

: ..l'|%vbu.¥lrz trill-{T ‘IIuA (It!
., 7‘ I
.Alpttin‘v. )ilIl'hth I
IIQF Iv).v!i!agf .vi trill-Ill!
it! zest-3.151.: ii?! 193.. .i’tgl. II
‘ . ‘ it: 5 ‘Q. ‘|||\ l5!.'.|ll

E

 

THESIS

{\r\;\
V' ~ L.

 

Illl\IllllHillllllllllllllllllllllllllllllllll
UEHARY 3 1293

Michigan S312;
University

 

 

 

This is to certify that the

thesis entitled

EFFICACY OF EQUI-Si ON INCREASING PLASMA SILICON
CONCENTRATIONS AND ALTERING BONE AND COLLAGEN
METABOLISM IN HORSES '

presented by

Kristine J. Lang

has been accepted towards fulﬁllment
of the requirements for

M.S. Animal Science

 

degree in

 

Major professor

427’ M

Date

 

0-7639 MS U 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

 

 

 

 

 

 

 

 

 

 

 

 

 

 

11/00 chlRC/DaoDmpGS—p.“

 

EFFICACY OF EQUI-Sim ON INCREASING PLASMA SILICON
CONCENTRATIONS AND ALTERING BONE AND COLLAGEN METABOLISM
IN HORSES
By

Kristine Josephine Lang

A THESIS
Submitted to
Michigan State University
In partial fulﬁllment of the requirements for the degree of
MASTER OF SCIENCE

Department of Animal Science

2000

ABSTRACT

EFFICACY OF EQUI-SiTM ON INCREASING PLASMA SILICON

CONCENTRATIONS AND ALTERING BONE AND COLLAGEN METABOLISM IN
HORSES
By
Kristine Josephine Lang
Twelve Arabian mare/foal units (Exp. 1) and 20 Arabian and Quarter Horse

yearlings (Exp. 2) were used to determine if EQUI-SiTM supplementation would increase
plasma and milk silicon (Si) concentrations and alter systemic markers of bone
metabolism. Horses were pair-matched and randomly assigned to two groups, 2% EQUI-
SiTM supplemented (S) or control (C). Samples were taken on d O, 15, 30 and 45. Plasma
and milk Si concentrations and milk calcium (Ca) and phosphorus (P) concentrations
were determined. Serum was analyzed for osteocalcin (OC), carboxy-terrninal
pyridinoline cross-linked telopeptide region of type I collagen (ICTP), and pyridinoline
and deoxypyridinoline (PYD). All S horses had higher plasma Si concentrations than C
(Exp. 1 & 2), and S mares' milk (Exp. 1) had higher Si concentrations on d 45 and higher
Ca and P concentrations on d 15. There were no statistical differences in OC, ICTP or
PYD in mares or foals (Exp. 1). In Exp. 2, S yearlings had lower ICTP concentrations on

d 45. Results Of these two studies suggest that EQUI-SiTM increases plasma and milk Si

concentrations, and may alter bone and collagen biochemical serum markers in horses.

ACKNOWLEDGMENTS

I would like to begin by thanking my advisory committee members, Drs. Brian
Nielsen, Michael Orth, Gretchen Hill and Hal Schott, for their guidance and support
throughout my Master's program. Each of you provided expertise in this research, and
helped me to explore science from a big picture approach. I would especially like to
thank my major professor, Dr. Brian Nielsen, for his endless support, willingness to
always lend a hand, and attention to detail. Brian, you made it impossible for me tO say
no to coming to such a great school, and I am so thankful for the opportunities you have
provided me. In fact, I liked my Master's so much, I'm continuing on at MSU for my
PhD. I very much look forward to my future, and feel certain that I have much more tO
learn from you all.

I am very grateful for the funding of MexAmeriCan Trading Corporation and Mr.
Al Basile. You generous donation of EQUI-SiTM and additional funding for this research
is very appreciated.

I would also like to thank Karen Waite. Karen, you are truly an inspiration. You
are always there to ﬁll so many capacities, be it teaching new lab techniques, helping

collect samples from omery foals, editing my thesis, or acting as a sounding board and

iii

vent source. But above all, you are a wonderful friend. I value our friendship, and only
hope to see it grow.

I would like to thank Tonia Peters, Jennifer Hawkins, Jane Link and Roy Radcliff
for your assistance in lab assays. I very much appreciate the time you gave to help me on
numerous occasions.

Thank you to Paula Hitzler and Shorty Parks for helping with the animal work on
this project. You cooperation and experience are very appreciated. I would also like tO
thank my undergraduate AN S490 students. I know waking up early during summer
vacation is not every college student's dream, but you did, and didn't complain. I thank
you for that. You made it possible to complete the enormous amount of animal care and
collections that were involved with these projects.

There is so much support that my fellow graduate students offer, and our
camaraderie is very important to me. Thank you Renee Bell, Molly Nicodemus, Kris
Hiney and David Mello for making each day fun. I've learned a lot from each of you, and
feel I'm all the better for the time we've worked together. I would also like to thank all Of
you that helped with my project, from mucking stalls and feeding, to sample collections.

Finally, I would like to thank my family and friends. No matter where I end up in
life, remember that I carry a part of you with me always. I thank you for the love,

support and guidance you have given. I am so blessed by the relationship we all share.

iv

TABLE OF CONTENTS

 

 

 

 

LIST OF TABLES ,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, viii

INTRODUCTION ,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, 1

CHAPTER 1

REVIEW OF LITERATURE ,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, 4
Physiology of Bone ..................................................................................................... 4
Bone Growth and Development ,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, 7
Physiology of Collagen ........................................................................ ll
Collagen Synthesis ,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, 12
Biochemical Markers Of Bone and Collagen ,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, 14
Milk .................................................................. 16
Silicon __ ....................................................................................... 18
Rationale for Experiments ,,,,,,, -- ................................................. 26

CHAPTER 2

EXPERIMENTAL PROCEDURE: EXPERIMENT 1 ..............................................
Assignment of Treatments and Management of Animals ......................................
Growth Measurements
Sample Collection ______________________________________________________________________________
Minimizing Silicon Contamination and Special Cleaning of Labware .................
Biochemical Marker Analysis ..................................................................................
Silicon Analysis ........................................................................................................
Calcium and Phosphorus Analysis ,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,
Statistical Analysis. ,_ .................................................

RESULTS AND DISCUSSION: EXPERIMENT l ...................................................

EQUI-SiTM and Plasma Silicon Concentrations Of Mares ......................................
EQUI-SiTM and Milk Silicon, Calcium and Phosphorus Concentrations ,,,,,,,,,,,,,,
EQUI-SiTM and Plasma Silicon Concentrations Of Foals _ _.
EQUI-SiTM and Bone Metabolism Of Mares and Foals __________________________________________
EQUI-SiTM Supplementation and Foal Growth"

 

 

 

 

 

27
27
29
29
30
31
34
34
35
37
37

42
43
52

CHAPTER 3

 

 

 

 

 

 

 

 

 

 

 

EXPERIMENTAL PROCEDURE: EXPERIMENT 2 ,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, 53
Assignment of Treatments and Management of Animals ,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, 53
Growth Measurements ............................. 56
Sample Collection ............. 56
Minimizing Silicon Contamination and Special Cleaning of Labware ,,,,,,,,,,,,,,,,, 58
Biochemical Marker Analysis ............................ 58
Silicon Analysis ,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, 58
Statistical Analysis ,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, 58

RESULTS: EXPERIMENT 2 ............................ 60
EQUI-SiTM Supplementation and Plasma Silicon Concentrations 60
EQUI-SiTM Supplementation and Bone and Collagen Metabolism 61
EQUI-SiTM Supplementation and Yearling Growth _67

CHAPTER 4

CONCLUSIONS AND IMPLICATIONS ,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, 69

APPENDD( A
Non-normalized milk Ca concentrations and ANOVA tables (Project 1) ,,,,,,,,,,,, 74
Non-normalized plasma Si concentrations for yearlings (Project 2) ,,,,,,,,,,,,,,,,,,,,, 74

APPENDIX B
Proc Mixed Tables ,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, 76

APPENDD( C
Proc Corr Tables ,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, 83

APPENDIX D
Hoof Measurement Data Tables: Experiment 2 ,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, 86
Parasite Data Tables: Experiment 2 ,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, 86

APPENDIX E
Comparison Of Blood Sample Collection Methods and Their Effect on
Plasma Si Concentrations 88

 

vi

APPENDIX F
Osteocalcin and Carboxy-terminal Pyridinoline Cross-link Telopeptide Region
Of Type I Collagen Measurements in Arabian Horses for the First Year
Of Life 95

LITERATURE CITED“ 104

 

vii

LIST OF TABLES

 

 

 

 

 

 

 

Table 1. Calculated daily feed composition for mares on an as fed basis ,,,,,,,,,,,,,,,,, 28
Table 2. Calculated total daily nutrients for mares on an as fed basis ______________________ 28
Table 3. Means of plasma Si concentrations (pg/L) in mares ,,,,,,,, .. 37
Table 4. Means of milk Si concentrations (pg/L) 38
Table 5. Means of differences from d 0 for milk Ca concentrations (mg/g) _____________ 40
Table 6. Means of P concentrations (mg/g) in milk _ ,,,,,,,,,,,, 41
Table 7. Means of plasma Si concentrations (ug/L) in foals 42
Table 8. Means of 0C concentrations (ng/mL) in mares ........................................... 44
Table 9. Means of OC concentrations (ng/mL) in foals _45
Table 10. Means of ICTP concentrations (uglmL) in mares 46
Table 11. Means Of ICTP concentrations (ug/mL) in foals _______________________________________ 48
Table 12. Means Of PYD concentrations (ng/mL) in mares, ,,,,,,,,,,, 49
Table 13. Means Of PYD concentrations (ng/mL) in foals ,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, 51
Table 14. Means of growth measurements for foals ,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, 52

Table 15. Calculated daily feed composition table for Arabian yearlings

on an as fed basis ,,,,,,,,,,,,,,,,,,,,,, 54
Table 16. Calculated daily nutrient table for Arabian yearlings

on an as fed basis _ 54

 

viii

Table 17. Calculated daily feed composition table for Quarter Horse yearlings
on an as fed basis

 

Table 18. Calculated daily nutrient table for Quarter Horse yearlings
on an as fed basis

 

Table 19. Means of differences from day 0 for plasma Si concentrations (ug/L)

in yearlings

 

Table 20. Means of OC concentrations (ng/mL) in yearlings ....................................

Table 21. Means of ICTP concentrations (ug/mL) in yearlings ................................

Table 22. Means Of PYD concentrations (ng/mL) in yearlings .................................

Table 23. Means Of growth measurements in yearlings

 

Appendix Table 1A. Non-normalized means Of milk Ca

concentrations (mg/g) in Arabian mares

 

Appendix Table 1B. Proc mixed table for non-normalized milk Ca

concentrations (mg/g) for Arabian mares

ooooooooooooooooooooooooo

Appendix Table 3A. Non-normalized means of plasma Si
concentrations (ug/L) in yearlings ,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,

Appendix Table 4A. T-test for D O for plasma Si concentrations for yearlings ,,,,,,

Appendix Table 1B. Proc mixed table for plasma Si concentrations (pg/L) for

Arabian mares

 

 

 

Appendix Table 2B. Proc mixed table for milk Si concentrations (ug/L) for

Arabian mares

 

Appendix Table 3B. Proc mixed table for milk Ca concentrations (mg/g) for

Arabian mares

 

ix

,__55

55

60

62

63

66

68

74

74

74

76

76

Appendix Table 4B.

Arabian mares

Proc mixed table for milk P concentrations (mg/g) for

 

Appendix Table 5B.

Arabian foals

oooooooooooooooooooooooooooooooooooooooooooo

 

 

Appendix Table 6B.

Arabian mares

 

Appendix Table 7B.

Arabian foals

 

Appendix Table 8B.

Arabian mares

 

Appendix Table 9B.

Arabian foals

 

Appendix Table 10B. Proc mixed table for serum PYD concentrations (ng/mL)

for Arabian mares

 

Appendix Table 11B. Proc mixed table for serum PYD concentrations (ng/mL)

for Arabian foals

 

Appendix Table 12B. Proc mixed table for plasma Si concentrations (pg/L) for

yearlings ____________________

.........................................................

 

Appendix Table 13B. Proc mixed table for normalized plasma Si concentrations

(ng/mL) for yearlings

 

Appendix Table 14B. Proc mixed table for serum OC concentrations (ng/mL)

for yearlings

 

Appendix Table 15B. Proc mixed table for serum ICTP concentrations (til/L)

for yearlings ..............

 

Appendix Table 16B. Proc mixed table for serum PYD concentrations (ng/mL)

for yearlings

 

76

77

77

77

77

77

78

78

78

78

79

Appendix Table 17B.
Appendix Table 18B.
Appendix Table 19B.
Appendix Table 20B.

Appendix Table 21B.

Appendix Table 1C.
Si levels

Proc mixed table for foal wither height (cm), 80

Proc mixed table for foal hip height (cm) ,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, 80
Proc mixed table for yearling wither height (cm) ,,,,,,,,,,,,,,,,,, 80
Proc mixed table for yearling hip height (cm) ,,,,,,,,,,,,,,,,,,,,,,, 8O
Proc mixed table for yearling weight (kg) ,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, 81

Proc corr table for D 0 milk Si concentration and foal plasma

 

 

 

 

 

 

 

 

Appendix Table 2C. Proc corr table for D 15 milk Si concentration and foal plasma
Si levels 83
Appendix Table 3C. Proc corr table for D 30 milk Si concentration and foal plasma
Si levels 83
Appendix Table 4C. Proc corr table for D 45 milk Si concentration and foal plasma
Si levels" 83
Appendix Table 5C. Proc corr table for D 0 milk Si concentration and mare plasma
Si levels ____________________________________________________________________________ 84
Appendix Table 6C. Proc corr table for D 15 milk Si concentration and mare plasma
Si levels 34
Appendix Table 7C. Proc corr table for D 30 milk Si concentration and mare plasma
Si levels ,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, 84
Appendix Table 8C. Proc corr table for D 45 milk Si concentration and mare plasma
Si levels ....... 34
Appendix Table 1D. Mean Toe Length (cm) for Yearlings ....................................... 86
Appendix Table 2D. Mean Right Heel Length (cm) for Yearlings ,,,,,,,,,,,,,,,,,,,,,,,,,,, 86

xi

Appendix Table 3D. Mean Left Heel Length (cm) for Yearlings ............................. 86

Appendix Table 4D. Mean Parasite Egg Number for Arabian Yearlings
on d 45 ....... 86

Appendix Table SD. Mean Parasite Egg Number for Quarter Horse Yearlings
on d 45 86

Appendix Table 1B. Mean Si concentrations (pg/L) for three blood collection
methods ...... 92

Appendix Table 1F. Calculated daily nutrient table on an as fed basis ,,,,,,,,,,,,,,,,,,,,, 97

Appendix Table 2F. Mean OC and ICTP concentrations (i SEM) in horses from
birth through d 150 -_..100

xii

INTRODUCTION

Bone-related injuries and diseases involving bone loss plague the equine industry.
Methods for enhancing bone formation are continually being investigated in hopes Of
increasing skeletal integrity and durability. Increased bone metabolism can occur for a
variety Of reasons: growth/development, bone damage, hormonal stimulation
(endogenous factors), and exogenous factors (diet). Increasing exercise, for example, can
alter bone metabolism and physiology. However, excessive exercise can be detrimental
to cartilage tissue. Therefore, nutritionally supplemental means (exogenous stimulator)
of enhancing bone formation appears to be an alternative way to maximize bone strength.
We invested the role of providing pharmacological concentrations of silicon as a
supplemental method of increasing bone formation.

Silicon (Si) is the second most abundant element of the Earth's crust (Carlisle,
1972), and is present in everything from the air that we breathe to the sand that we walk
on. It is even found in many foods that we eat every day. Not all Si is available for
absorption in the gut. Sodium zeolite A is a hydrated aluminosilicate compound that
breaks down into monosilicic acid in the gut, and serves as an absorbable Si source.
Sodium zeolite A compounds have been supplemented in a variety of animal species for

reasons that range from increasing eggshell thickness (Roland, 1988) to decreasing bone-

related injuries in horses (Nielsen et al., 1993). EQUI-Sim, which is a sodium zeolite A
compound, is the source Of Si chosen for the current study.

Silicon has been reported to have a role in bone physiology and metabolism, by
increasing osteoblastic activity (Brady et al., 1991; Keeting et al., 1992) and decreasing
osteoclastic activity in vitro (Schiitze et al., 1995). However, limited research has been
done studying the effects of supplemental Si on bone metabolism and physiology in viva.
While increased bone metabolism in rats supplemented with a Si source has been
reported (Seaborn and Nielsen, 1994), it is not clear if increased bone metabolism due to
Si has a positive or negative effect on bone histomorphometry. No studies have
investigated the role of supplemental Si in altering bone metabolism in horses. Nielsen et
al. (1993) found that supplemental sodium zeolite A decreased the incidence Of bone-
related injuries in young racing Quarter Horses; however, no determination of bone
turnover was attempted in that study.

Based on the limited information about supplemental Si and bone physiology and
metabolism in vitro and in viva, supplemental Si in the horse warrants investigation.
Horses are continually straining (loading) their skeletal system when performing the
intense activities that comprise the current sport horse industry, and advancements in
increased bone formation may prove to be advantageous in increasing the longevity Of
the horse as well as in decreasing bone-related injuries. If supplemental Si can enhance

bone formation, advancements may be made in increasing the durability Of sport horses.

Therefore, our objectives were to determine if a supplemental Si source (EQUI-Sim)
could: I) increase plasma Si concentrations in mares and yearlings, 2) increase milk Si
concentrations in mares, thereby increasing plasma Si concentrations in foals, and 3)
improve bone and collagen metabolism in horses, as evident by changes in bone and

collagen markers.

CHAPTER 1

REVIEW OF LITERATURE

Physiology of Bone

Bone is a unique and dynamic connective tissue that is characterized by distinct
cells and mineralization of an organic matrix. This rigid tissue provides support for the
body, protection of organs, marrow, the animal with the ability to move, and regulation of
ﬂuids and ions in the body. Bone exists in two main forms: cortical and trabecular bone.
Cortical bone is a solid mass of bone, whereas trabecular is more mesh-like, or woven
like a net. Because trabecular bone has a greater surface area than cortical bone, mineral
is drawn from trabecular bone during short term calcium and phosphorus deﬁciencies in
order to maintain equilibrium within the body. Trabecular, or cancellous bone, uses the
arrangement of the trabeculae to help reduce stress applied to the bone from loading. The
adult human skeleton consists Of approximately 80% cortical bone. Trabecular bone is
located on the ends of long bones, in vertebrae, and envelopes the marrow (As reviewed
by Ice, 1988; Wasserman, 1984; Hays and Swenson, 1984).

Long bones are made up Of the diaphysis, or shaft, and the epiphysis, or end Of the
bone. The diaphysis and epiphysis are connected by the metaphysis. In young animals,
the epiphysis and metaphysis are separated by a thick cartilaginous tissue known as the

growth plate. The long bone elongates via the growth plate, as new bone is formed at the

base Of the plate where blood vessels inﬁltrate the cartilage matrix. In a mature animal,
the growth plate is replaced by trabecular bone, and the bone will cease to grow in length
(As reviewed by Ice, 1988; Wasserman, 1984).

Both cortical and trabecular bone are made up Of larnellae, which consist Of
circular collagen ﬁbers and mineralization. Within this layer of collagen, lacunae, or
cavities, are connected by a network Of channels known as canaliculi. These cavities and
Channels comprise the communication system for bone. One concentric unit of lamellae,
and its associated lacunae and canaliculi, is known as an osteon or Haversian system.
Osteoblasts, or bone forming cells, become entrapped in the lacunae and canaliculi and
secrete intracellular bone matrix, known as osteoid. Osteoblasts will completely
surround themselves with osteoid, mature into osteocytes, and reside in the lacunae until
death (As reviewed by Ice, 1988; Lawrence and Fowler, 1997).

Bone is made up Of an organic matrix and inorganic salts. The organic matrix is
largely composed Of collagen, mucopolysaccharides and glycoprotein. Type I collagen
makes up about 90% of the organic extracellular matrix. Within the matrix, amorphous
ground substance is released to house the collagen ﬁbers in a glue-type fashion, and also
accumulates the inorganic salts, or minerals. The mineral in bone is most commonly
found as crystalline hydroxyapatite, Ca,O(PO,,)6(OH)2. These highly organized crystals
are about 30 to 50 Angstroms wide, can be as long as 600 Angstroms, and make up about

two-thirds of the dry weight of the bone matrix. Apatite provides the skeleton with

rigidity, and is also a calcium (Ca) and phosphorus (P) reservoir. Mature bone (net
weight) is approximately 30% organic material (mostly collagen), 45% mineral, and 25%
water (AS reviewed by Jee, 1988; Lawrence and Fowler, 1997; Pritchard, 1972; Hays and
Swenson, 1984, Wasserman, 1984).

Lawrence et al. (1994) reported that the mineralization process of equine bone
was almost complete (76%) by 12 months of age, and that the bone mineral content of the
equine third metacarpal did not differ by sex. Calcium and P have an important
relationship in bone formation. Dietary Ca and P are absorbed in the small intestine, and
the absorption rate depends on the form of the nutrient, the Ca:P ratio, and dietary
concentrations of these and related nutrients. Additionally, vitamin D, manganese,
magnesium, ﬂuoride, copper, potassium, sodium, chloride and silicon (Si), may have a
role in bone formation and metabolism by altering the structure of the organic and/or
inorganic components of bone (As reviewed by Hays and Swenson, 1984; Wasserman,
1984).

Bone metabolism is regulated by a wide variety of hormones, cytokines and other
factors. Metabolism is also regulated by the activity Of five different types Of bone cells,
including: osteoprogenitor, bone lining cells, osteoblasts, osteocytes, and osteoclasts.
Osteoprogenitor cells consist of preosteoblasts, which will undergo mitosis and
differentiation to become osteoblasts. Bone lining cells are ﬂat, inactive, and/or resting

Cells that cover the surface of bone. These bone lining cells also serve as ion barriers

between the canaliculi and interstitial ﬂuids. The osteoblast, or bone forming cell,
originates from osteogenic (undifferentiated mesenchymal stem cells) lineage and
secretes minerals, amorphous ground substance, osteocalcin, osteonectin, water and type
I collagen into the matrix. Osteoblasts have rough endoplasmic reticulum and golgi, and
these two structures are largely responsible for producing type I collagen proteins.
Osteocytes are mature bone cells that have many long cytoplasmic branches projecting
from the main body. Finally, osteoclasts are the multinucleated, highly motile cells that
are responsible for resorption, and are derived from preosteoclasts in the bone marrow.
Active osteoclast cells are usually found in resorption pits, or Howship’s lacunae. The
junction of osteoclast cell and bone forms a rufﬂed border, which forms an impermeable
microenvironment, and enhances resorption. Hormones, growth factors, ion
concentrations, and nutrition regulate all ﬁve bone cells. Mesenchymal stem cells can
also differentiate into bone marrow cells, which can then differentiate into blood-formin g
cells (As reviewed by Jee, 1988; Nijweide et al., 1986; Marks and Popoff, 1988;
Pritchard, 1972; Wasserman, 1984; Lawrence and Fowler, 1997).
Bone Growth and Development

There are two forms Of bone growth -— endochondral and intramembranous
ossiﬁcation. The compound (bone or cartilage) that provides the starting material for
bone formation characterizes each of these forms of osteogenesis. Endochondral

ossiﬁcation indicates growth occurring in the cartilage, which is made up Of

chondrocytes, and occurs in long bones at the growth plate. The growth plate is
composed of four zones. The ﬁrst zone, known as the resting zone, is near the epiphysis
of the bone, and is made up of hyaline cartilage formed by chondrocytes. Although the
exact function of this zone is not clear, chondrocytes are likely recruited into the
maturation process during growth. Under this layer is the proliferative zone, which
consists of cells that multiply and synthesize extracellular matrix. The hypertrophic zone
begins when chondrocytes differentiate and hypertrophy. Finally, in the zone Of
calciﬁcation, terminally differentiated chondrocytes, near the metaphysis, begin to lay
down mineral into the matrix (As reviewed by Jee, 1988). The chondrocytes in this
region ultimately undergo apoptosis (Hatori et al., 1995). Endothelial and phagocytic
cells initiate vascularization in the empty lacunae (Jacobson et al., 1997). At the
chondro-osseous junction, where calciﬁcation and chondrocyte cell death have begun, the
marrow begins to form the primary spongiosa. In this region, osteoblasts in the
surrounding matrix of the woven bone continue to release osteoid. Additionally, the
marrow carries osteoclasts and resorbs the bone. This synergistic effect of formation and
resorption helps to shape the long bone (As reviewed by Vaughan, 1980).
Intramembranous growth occurs on bone surfaces or ‘in membrane’, such as
under the periosteum of long and ﬂat bones. Mesenchymal stem cells congregate in the
area of new bone formation and differentiate into osteoblasts. The osteoblasts can then

secrete matrix, which will become calcified. This increases the amount of bone in the

area. Some osteoblasts become surrounded by osteoid and will become osteocytes.
Coupled with the bone formation process is osteoclastic activity on the innermost surface
Of the cortical bone. In this manner, the marrow cavity grows larger at a proportional rate
to the outer portion of the cortical bone. This results in an increased diameter of the
diaphysis, without altering the width of the cortical bone (As reviewed by Wasserman,
1984).

After initial formation of bone, the tissue remains very dynamic. During long
bone formation, certain areas need to be shaped as the bone continues to grow.
Therefore, in modeling, formation and resorption are separate, and may occur at different
sites. If the two processes are not coupled, a drift of the bone can occur, resulting in an
eccentric shaft. During growth, modeling occurs continuously, and there is an ultimate
net gain of bone via this process (As reviewed by Jee, 1988; Lawrence and Fowler, 1997;
Wasserman, 1984). Modeling and remodeling increases bone strength and stiffness by
adding bone to areas where deformation will be the greatest, the outside of the bone,
known as the periosteum (Kimmel, 1993).

Bone formed by modeling during the growth phase tends to be structurally
inferior to the bone that is formed during remodeling. Additionally, adult bone tends to
change over time, become damaged by microfractures as a result of loading, or reduce in
quality as it ages. For these reasons, remodeling is an important mechanism utilized by

the body to replace and repair bone. Remodeling is a different process than modeling,

whereby formation and resorption events are coupled and isolated to the same location.
Remodeling occurs to replace the Haversian systems upon their demise and to respond to
altered stresses and strains placed on the bone (As reviewed by Jee, 1988; Lawrence and
Fowler, 1997; Wasserman, 1984; Lanyon, 1994). There are three types of strain: tensile,
compressive and shear. The cells that make up bone undergo these strains with normal
use (El Haj and Thomas, 1994). Therefore, loaded bones must adapt by continually
altering its shape and architecture (Lanyon, 1994). Remodeling may strengthen bone by
replacing damaged areas caused by strain, without altering bone mass (Kimmel, 1993).
Remodeling can also occur during times of disuse, and acts to decrease the total amount
of bone present, thereby decreasing the strength of bone.

Remodeling stages include: quiescence, activation, resorption, reversal,
formation, and return to quiescence. The bone surface is normally in the quiescent stage,
and the cells are inactive. Upon activation, osteoclasts are recruited from the bone
marrow, and migrate to the site of activation, where they remove mineral and matrix in
the resorption period. The reversal phase is characterized by the repression of
osteoclastic activity, and deposition of cement into the resorption cavity. Finally, new
bone formation begins with matrix formation and sequential mineralization. Once this
cycle has completed, the surface of the bone will return to quiescence (As reviewed by
Parﬁtt, 1984). The site of resorption and formation, where the remodeling occurs, is

referred to as the bone remodeling unit. Furthermore, this turnover cycle usually is

10

completed in about four months. Trabecular bone remodeling occurs on the surface,
while cortical bone remodeling occurs by osteoclasts tunneling through bone, with new
osteoblastic apposition occuring during formation from the cement line inwards, until a
new Haversian canal is established (As reviewed by Jee, 1988).

Physiology of Collagen

Collagens are a family of proteins found in most tissues, including bone and
cartilage. Collagen provides connective tissue with Strength, and makes up a major
portion of bone and a minor portion of cartilage composition. This supramolecular
aggregate is associated with the extracellular matrix, and is stabilized by the triple helix
that is characteristic of collagen (As reviewed by Wasserman, 1984; Lawrence and
Fowler, 1997; Marks and Popoff, 1988; van der Rest and Garrone, 1991).

Collagen molecules have a repeating pattern of glycine-proline-hydroxyproline-
glycine-x-y amino acids, where x and y denote any other amino acid. Hydroxyproline
and hydroxylysine are two amino acids found predominately in connective tissue, and are
prevalent in collagen. For steric reasons, glycine is the only amino acid that can be in the
center of the superhelix, or collagen molecule.

There are many types of collagen, which are found in various connective tissues
throughout the body. Type I collagen is the most abundant form of the molecule, and is
found in bone, tendon and skin and many other tissues. This molecule has an [Ct-1(1)]2 or-

2(I) conformation. Type II collagen is largely found in cartilage and between vertebrae,

11

and has an [(1-1 (11)]3 structure. Type I and II collagens are examples of ﬁbril-forming
collagens. Other collagen types, such as basement membrane, microﬁbrillar, anchoring
ﬁbril, hexagonal network-forming, ﬁbril-associated with interrupted triple helices,
transmembrane, and multiplexin collagens, form different structures. These collagens
have various (1 subunits , that help differentiate each type of collagen (As reviewed by
Lawrence and Fowler, 1997; van der Rest and Garrone, 1991; von der Mark, 1999).

The interﬁbrillar space between type I collagen ﬁbrils is the site of bone
mineralization. Minerals associate in a rigid, crystalline structure known as
hydroxyapatite. Type I collagen is the most prevalent type (90%) of collagen in bone.
Additionally, collagen comprises about a third of the dry weight of bone organic matrix
(As reviewed by Pritchard, 1972; Wasserman, 1984; Lawrence and Fowler, 1997).
Collagen Synthesis

Collagen is synthesized intracellularly as preprocollagen, followed by
hydroxylation of proline and glycosylation of hydroxylysine, to yield procollagen. The
ends are cleaved, and procollagen is expelled from the cell (As reviewed by Wasserman,
1984). Extracellularly, the procollagen molecules undergo hydrolysis to form
tropocollagen, which is the basis for matrix collagen ﬁbrils (As reviewed by Lawrence

and Fowler, 1997; Wasserman, 1984). The polypeptide chains Of tropocollagen wind in a

12

left-handed fashion, and have a 30° rise, formed when every third amino acid residue is
turned into the center (van der Rest and Garrone, 1991).

Fibril forming types of collagen molecules, like those in bone, are arranged
parallel to each other, with spaces strategically placed between ﬁbers, once the molecules
have been cleaved at the N - and C-telopeptides to form collagen ﬁbrils extracellularly.
The ends of the ﬁbers can then become crosslinked. Crosslinkage between collagen
molecules at hydroxylysine and lysine sites provide collagen with its stabilization, and
form the striations in the collagen ﬁbrils. Collagen, in conjunction with its crosslinking,
yield a scaffold for minerals to bind in the matrix (AS reviewed by Wasserman, 1984; van
der Rest and Garrone, 1991, Lawrence and Fowler, 1997, Marks and Popoff, 1988).

Pyridinoline (PYD), is a 3-hydroxypyridinium derivative consisting of three
amino acids and three carboxyl groups (Fujimoto and Moriguchi, 1978). AS crosslinks
originally develop from lysine and hydroxylysine residues, lysyl oxidase converts these
amino acid residues to an aldehyde, through oxidative dearnination, and the residues may
then mature into trifunctional, non-reducible cross-links (AS reviewed by Last et al.,
1990). An analog crosslink, deoxypyridinoline (DPD), is formed in the similar manner as
PYD, but involves reaction with a lysine residue from the helix rather than hydroxylysine
(Ogawa et al., 1982; reviewed by Robins, 1999). Eyre et al. (1988) found that
hydroxypyridinium residues persist in human bone through adulthood, thereby

establishing them as ﬁnal crosslinks (Eyre et al., 1988).

13

Biochemical Markers of Bone and Collagen

Equine research has developed a need for sensitive, non-invasive techniques to
measure bone metabolism. Fluid samples can be taken and analyzed for proteins or
enzymes that indicate activity or bone turnover. Serum and urinary assay kits are
commercially available to measure speciﬁc molecules of this type. Osteocalcin (OC) is
one such marker. The function of 0C is unclear (Price, 1982; Lian and Gundberg, 1988),
however several characteristics are known about the protein. Osteocalcin is a non-
collagenous protein synthesized largely by the osteoblast, contains three amino acids Of
gamma carboxyglutamic acid, and is therefore known as bone Gla protein (Fraher, 1993).
Osteocalcin is vitamin K-dependent (Price, 1982), binds Ca, and may be involved in the
control of mineralization (Fraher, 1993; Lepage et al., 1991). The Gla residues speciﬁc
to 0C are formed after translation of the protein occurs by a carboxylase enzyme system.
In the presence of Ca, these Gla residues undergo conformational changes that allow CC
to bind to hydroxyapatite, and then accumulate in bone matrix (Lian and Gundberg,
1988). Furthermore, OC has a high afﬁnity for hydroxyapatite (Price, 1982). The
protein exists in a 1:1 ratio with collagen in bone, and is proportional to hydroxyapatite
(Lian and Gundberg, 1988). It is believed that OC may regulate the size of the
hydroxyapatite crystal (As reviewed by Gundberg and Nishimoto, 1999). Osteocalcin is
cleaved through proteolysis, and is cleared from circulation by the kidneys (Gomez et al.,

1994). Serum concentrations of OC is probably a product of diffusion into the

14

circulatory system prior to binding to the hydroxyapatite (Price, 1982). Osteocalcin is
mostly formed by osteoblast activity, and thus is highly correlated with bone formation
(Lian and Gundberg, 1988; Kannus et al., 1996). Therefore, serum OC may be used as a
marker of bone synthesis.

When animals are exposed to natural light, OC undergoes a circadian rhythm
(Lepage et al., 1991). Osteocalcin concentrations in normal female Standardbred horses
were found to be consistent during the day from 0700 to 1900. These concentrations then
fell in the early evening hours (1900 to 2000), and increased at night until a peak was
reported at 0500 the next morning (Lepage et al., 1991). Lepage et al. (1991) therefore
concluded that blood samples should be taken during the daylight hours, since this period
experienced the least ﬂuctuation in OC concentrations. However, in a study where
horses were exposed to only ﬂorescent lighting, a circadian rhythm was not observed
(Hope et al., 1993).

Carboxy-terminal pyridinoline crosslinked telopeptides of type I collagen (ICTP)
are a 9kD portion Of type I collagen that is released during resorption of bone (Hassager
et al., 1994), and can be quantitated in serum and urine (Risteli et al., 1993). Once type I
collagen is broken down into ICI‘P, it cannot be used again for synthesis Of new collagen
in bone (Hassager et al., 1994), and is cleared from the body’s circulation through the
kidneys (Tahtela and Tholix, 1996; Risteli et al., 1993). Concentrations of ICT P decrease

With age of the horse, up until about four years of age, after which there were no

15

signiﬁcant differences (Price et al., 1995). A diurnal cycle has been reported with ICTP
concentrations (Risteli et al., 1993).

Finally, PYD and DPD are crosslinking amino acids residues of collagen released
from bone matrix during turnover. An enzyme linked immunoassay (ELISA) exists for
PYD and DPD crosslinks in serum and urine. Since PYD is found in cartilage and bone,
PYD and DPD can be used as biochemical markers to analyze metabolic bone or
cartilage diseases. Increased urinary PYD was seen in patients with osteoarthritis and
rheumatoid arthritis (Robins et al., 1986).

Milk

Milk contains carbohydrate, lipid, minerals, protein, and water. Casein is the
largest portion of proteins in milk. Irnmunoglobins in colostrum are synthesized from
amino acids in the blood. Lactose is the primary carbohydrate of milk, and is formed
from blood glucose. Triglycerides make up most of the lipids in milk (As reviewed by
Jacobson and McGilliard, 1984).

Peak lactation in the mare is found at 90 d postpartum (Ashcraft and Tyznik,
1976). Mares' milk has been reported to contain 1.6% fat, 2.4% protein, 6.1% lactose,
and .5% ash (As reviewed by Jacobson and McGilliard, 1984). Gibbs et al. (1982)
studied 14 mature Quarter Horse mares to determine milk yield and formulate a lactation
curve; they reported that average milk yield was 10.9 kg/d (2.1% Of body weight). Parity

Signiﬁcantly inﬂuenced milk yield, as multiparous mares averaged 11.7 kg/d. Daily milk

16

yield signiﬁcantly declined with time, with percent solids being highest 10 d after
parturition and declining thereafter (Gibbs et al., 1982). From a second study with five
mares in rnidlactation, milk composition was reported to be 10.5% total solids, being
1.29% fat, 1.93% protein, 6.91% sugar, and .42% ash on a whole-milk basis (Oftedal et
al., 1983). Additionally, this study measured milk intake of foals using D20 kinetics, and
reported that daily milk intake averaged 16 kg at 11 d, 15 kg at 25 d and 17.6 kg at 39 d.

The mineral content of mare’s milk has been the topic of few studies. Since
minerals are so vital in the development of the young animal for various physiological
functions, including bone mineralization, further studies are required to completely
characterize the mineral content Of mare milk. Utilizing ﬁve Thoroughbred mares,
Schryver et al. (1986) analyzed milk for Ca, P, Mg, K, Na, Cu and Zn. They reported
that the ash content of the milk was highest in the first week and decreased 26% during
the first four-wk period, 16% the next four-wk period, 16% the next eight-wk period, and
ﬁnally 50% in the next l6—wk period (Schryver et al., 1986).

Foals born of Ca deﬁcient mares had signiﬁcantly smaller cannon diameters
(Glade, 1993). Additionally, Glade (1993) found that it took 16 wks longer for the Ca
deficient mares to ﬁnally return to baseline in sound velocity and bone breaking strength
of both the right and left cannon bones when compared to the Ca sufﬁcient mares,
indicating a decrease in mineral content of the long bone for a longer period of time in Ca

deﬁcient mares. The study concluded that pregnancy and lactation affects the Skeletal

l7

system and mineral homeostasis, and that adequate Ca intake is necessary during
gestation and lactation to prevent skeletal mineral loss (Glade, 1993).

Trace minerals have been deﬁned in human milk. In a study that followed six
women through the ﬁrst ﬁve mon of lactation, the following trace elements were found in
large quantities in order Of highest to lowest concentrations: zinc, Si, copper, iron,
barium, and strontium. Over time, zinc, iron, and copper concentrations decreased, 1"
however strontium, barium and Si did not change as lactation advanced (Anderson,
1993). However, more research needs to be done to further understand the role of these
trace minerals in milk.
Silicon

Silicon is the second most abundant element in the earth’s crust. Silicon is a
member Of Group IV on the periodic table, and is an alkali and alkaline earth cation.
Because Si can not form double bonds due to a large radius, all silicates are rigid
structures (Van Soest et al., 1983). Several decades ago, most research involving Si
centered around toxicity of silica and ﬁbrous silicates, especially in their role with
silicosis. However, in the 1970’s, elemental Si was found to be important for normal
growth and development in chicks, when specialized highly puriﬁed diets were fed

(Carlisle, 1972). This element has been studied in relation to its role in the skeletal

system in animals (Carlisle, 1972, 1978).

18

Herbivores consume large amounts of Si through the grazing process, which
passes through mostly unabsorbed. However, a portion of the dietary Si is absorbed and
utilized in the body or excreted. Silicon is normally excreted without side effects,
although uroliths may form and block urinary passages, which may cause death (Hays
and Swenson, 1984). As a result of its presence in many foodstuffs and unknown
requirement status, Si deﬁciencies are unknown under normal dietary conditions (Van
Soest et al., 1983). Additionally, in the case of systemic hypersilicemia, a decrease in the
activity of rat hepatic and renal antioxidant enzymes superoxide dismutase, catalase and
glutathione peroxidase was reported with .1% Si and .2% Si aqueous solutions (Najda et
al., 1994). These data may be important when conducting studies using excessive dietary
Si concentrations.

Silicon is naturally found in the environment as silica, SiOz. Grains and grain
fractions are very high in Si content, including whole-grain barley and oats, rice bran and
ﬂour (Pennington, 1991). When silica becomes hydrated, orthosilicic acid, Si(OH),, is
formed. Once Si derivatives are converted to silicic acid in the body, Si is absorbed
through the intestine and excreted via urine (Benke and Osborn, 1979; Carlisle, 1982;
Van Soest et al., 1983). The concentrations of Si in blood are increased when silicates
were fed to rats (Carlisle, 1982) and horses (Frey et al., 1992; Nielsen et al., 1993). In

animal tissue, Si is found in the highest concentration in skin and connective tissue.

19

Silicon is also found in measurable quantities in milk. In the horse, Si composes .0244%
Of milk ash, which equates to 65 ug Si/ml of milk (Schwarz, 1978).

Because Si is so ubiquitous in the environment, and laboratory contamination may
occur, special laboratory precaution must be taken to minimize contamination (Carlisle,
1972). Graphite furnace atomic absorption spectrometry is utilized to measure Si in
biological samples of plasma and urine. Glass pipettes, syringes or vacuum tubes should
be used, since glass contains Si. Plastic labware should be used instead (Gitelman and
Alderman, 1990).

Sodium zeolite A (EZA) is a sodium aluminosilicate clay substance. Zeolites are
hydrated, crystalline alurninosilicates of alkali and alkaline earth cations (Mumpton and
Fishman, 1977), and have a three-dimensional framework of interconnected cavities 11.4
A in diameter, with pore openings that are 4.2 A in diameter (Roland et al., 1985).
Furthermore, zeolites can reversibly gain or lose water, and change the bound cation
easily, without altering the structure signiﬁcantly. The framework consists of
$0,," tetrahedra, and maintains an overall neutral charge. When trivalent aluminum
replaces the Si in a tetrahedral, an overall negative charge is yielded. Finally, to
counteract this negative charge, a cation is bound to the structure to return neutrality to
the compound. These cations commonly are Na“, CaH and K“ (Mumpton and Fishman,
1977; Van Soest et al., 1983). This neutralizing cation provides the silicate with a high

ion exchange capacity, especially for Ca“. Sodium aluminosilicate’s Caz“ exchange

20

capacity is more than 7.0 meq/g and may be beneﬁcial in capacities ranging from bone
metabolism (Seaborn and Nielsen, 1994) and mineralization (Carlisle, 1974) to egg shell
quality (Roland et al., 1985). “Ion exchange capacity is basically a function of the degree
Of substitution of aluminum for silicon in the framework structures; the greater the
substitution, the greater the charge deﬁciency, and the greater the number of alkali or
alkaline earth cations required for electrical neutrality” (Mumpton and Fishman, 1977).

Sodium zeolite A breaks down to monosilicic acid in the gut, and when fed to
horses caused a dose-related increase in plasma Si concentrations (Frey et al., 1992).
Nielsen et al. (1993) found that nine h post-feeding, a high Si diet (2.8% EZA of total
diet) yielded signiﬁcantly higher plasma Si concentrations in horses for 30, 90, 120 and
180 d of treatment. In hens intubated with EZA, serum Si and Al concentrations
increased signiﬁcantly over the control. However, the aluminum concentration was only
1/ 100 of the Si concentration (Rabon, Jr. et al., 1995). Additionally, Rabon, Jr. et al.
(1995) reported that .75% dietary EZA and 200 IU/kg cholecalciferol resulted in a two-
fold increase in serum Si concentrations compared tolOO IU/kg dose of cholecalciferol
and .75% Si.

In vitro, silicic acid increased Transforming Growth Factor — Beta (TGF—B),
which is produced in bone and platelet cells (Keeting et al., 1992). This may be

important in fracture and wound healing, as an increase in TGF-B causes an increase in

21

cell proliferation (Lind etal., 1995). If Si supplementation can activate TGEB in vivo, an
enhanced repair mechanism may result.

Sodium zeolite A increases egg speciﬁc gravity as the amount of EZA is
increased in the diet (Miles et al., 1986). When synthetic EZA was compared to naturally
occurring zeolites in the diet of 360 Hyline W36 hens, 3% natural zeolite addition did not
inﬂuence egg speciﬁc gravity, whereas EZA improved egg speciﬁc gravity (Roland,
1988). Improved speciﬁc gravity will result in reduced breakage during handling and
shipping. Roland et al. (1985) determined that zeolite A improved Dekalb X1 and Hyline
W36 hens’ ability to use Ca in eggshell formation and improved egg speciﬁc gravity.

In a study on White Leghorn cockerels that were fed a Si supplement Of sodium
metasilicate, the chicks grew more rapidly and obtained a higher weight after 23 d as
compared to the control chicks on a specially formulated silicon deﬁcient diet (Carlisle,
1972). Carlisle (1972) also found that Si deficient chicks had absent wattles and much
smaller combs, and their leg bones were shorter, of smaller circumference, and had a
thinner cortex.

Silicon may play a role in the calciﬁcation process. Because Si is found in
metaphyseal blood vessels and in Si-rich sites of bone, it is felt that as the blood vessels
invade the metaphysis, the calciﬁcation process is activated, and Si may be involved in
these sequences. Furthermore, Si may be coupled to Ca and aid in mineralization

(Carlisle, 1974). When the Ca content of the diet is low, and high amounts of Si are

22

supplemented, a higher percentage Of ash is recovered in a shorter period of time from
chick tibia bones. Therefore, Si may increase the rate of bone mineralization when Ca
levels are low (Carlisle, 1974).

Sodium zeolite A may have a role in cartilage metabolism. For example, Si
supplementation in chicks increased glycosaminoglycan levels in cartilage (Carlisle,
1976). This may be important in maintaining cartilage health, as glycosaminoglycans
draw water into cartilage allowing for its stiff and compressive characteristics (Nixon,
1991). Furthermore, Si deﬁciencies caused a decrease in articular cartilage of chicks,
whereas Si supplementation increased total connective tissue (Carlisle, 1976).

Silicon’s role in connective tissue seems to largely be related to collagen. Chicks
fed a Si-deﬁcient diet had thinner legs and smaller combs when compared to their size.
These tissues are normally composed of a high collagen content. Also there was less
hexosamine content in articular cartilage (Carlisle, 1976). Furthermore, a study done on
embryonic skull bones grown in an organ culture (Carlisle and Alpenfels, 1978), showed
Si supplemented systems had a marked increase in collagen content of the bones over the
non-supplemented groups after 12 d.

Silicon has been shown to inﬂuence bone in numerous studies. For example, in a
study where broilers had one wing immobilized and were fed sodium ﬂouride or sodium
silicate, a decrease in humeri strength was seen as a result of the immobilization in the

sodium ﬂouride group. However, the broilers fed sodium silicate showed no effects of

23

immobilization the last two wks Of the experiment. However, sodium ﬂuoride produced
an increase in bone breaking strength over the control, where those treated with sodium
silicate did not. The authors concluded that the results in bone breaking strength and
bone loss in this study still leave questions to be answered (Merkley and Miller, 1983).
In vitro and in viva results show that Si might improve in bone formation.
Carlisle and Alpenfels (1978) found that Si supplemented bones in culture increased 23%
(dry weight) over non-supplemented bones. This weight increase was due to an increase
in collagen content. Silicon supplemented bones showed an 88% increase in collagen
content over 12 (1 compared to the non-supplemented bones, which only increased 45%.
Additionally, Si supplemented bone cultures had an increase in Ca content of 14%, while
non-supplemented cultures remained at base level (Carlisle and Alpenfels, 1978).
Furthermore, zeolite A (assessed by 3H-thyrnidine uptake) increased proliferation of
human osteoblast-like cells in vitro when administered at doses of .1 tO 10 ug/mL. A
269% increase was seen with the 10 ug/mL dose over the control. Similarly, 100 jig/ml
Of zeolite A increased cellular alkaline phosphatase activity 310% (Brady et al., 1991).
The authors suggest that zeolite A may have an anabolic effect on osteoblast-like cells,
and may be beneﬁcial in diseases involving bone loss, such as osteoporosis (Brady et al.,
1991). It is important, however, to realize that zeolite A will break down into silicic acid
in the gut, and therefore, in vitro studies on the whole zeolite A compound may not have

a direct application to in vivo systems.

24

Bone metabolism was increased when weanling Sprague-Dawley rats were fed
normal physiological amounts of Si (35 ug Si as sodium metasilicate/g diet) as evident by
increased alkaline and acid phosphatase activity in homogenized femurs. In the Si
deﬁcient rats (.6 ug Si/g diet), the enzyme activities were lower (Seaborn and Nielsen,
1994). Additionally, when these rats were supplemented with a bone implant that was
harvested from either Si deﬁcient or Si supplemented rats, those that had the Si
supplemented implant showed higher enzymatic activity than those rats with the deﬁcient
implant. The researchers concluded that while the mechanism for the effect of
physiological amount of Si on bone turnover is not fully understood, they feel it might
involve a mediator that is transported throughout the body, as evidenced by increased
enzymatic activity due to the supplemented implant (Seaborn and Nielsen, 1994).

In the horse, EZA produced a dose related increase in plasma Si levels in both
weanling and yearling Quarter Horses (Frey et al., 1992; Nielsen et al., 1993). Frey et al.
(1992) found that horses supplemented with 2% EZA experienced a trend to gain Optical
bone density initially, but there was no difference between treatments at the end Of the
168 d study in any of the EZA supplemented groups. Therefore, if EZA does affect
Optical bone density, it is likely that it occurs early on, and is short lived. Furthermore,
Nielsen et al. (1993) found that Quarter Horses in race training that were fed EZA ran
farther and completed more strides before experiencing bone-related injuries when

Compared to non-supplemented horses. A positive correlation (R2 = .54) between plasma

25

Si concentrations and distance traveled before the horses experienced a bone-related
injury was also observed (Nielsen et al., 1993). However, no attempt at determining bone
metabolism as indicated by biochemical markers was made in this study, which should be
examined in horses.
Rationale for Experiments

High performance sport horses may experience bone-related injuries at any point,
and bone failure will halt a horse's athletic career. These injuries impact the welfare Of
the horse, as well as the economics of the industry, because Of increased expenses to treat
the injury, lose of use, and extensive therapy. Research has been conducted to examine
ways of alleviating the risk of bone-related injuries, or treating them once they have
occurred. Horsemen employ a variety of techniques to increase bone density and
strength, in hopes Of decreasing injury rates. Management, exercise and nutrition have
served as means for enhancing a horse's skeletal durability. Based on the literature cited
involving the relationship between supplemental Si and bone, I hypothesize that an
absorbable form of Si would increase plasma Si concentrations and alter bone turnover in
horses. We designed two experiments to test this hypothesis in broodmares, their foals,
and yearling horses. Our overall objectives were to determine if EQUI-SiTM, a sodium
zeolite A compound, could: I) increase plasma Si concentrations in the above horses, 2)
increase milk Si concentrations in mares, thereby increasing plasma Si concentrations in

foals, and 3) alter systemic bone and collagen metabolism in the above horses.

26

CHAPTER 2

EXPERIIVIENT AL PROCEDURE: EXPERIMENT 1

Assignment of Treatments and Management of Animals

Twelve Arabian broodmares were pair-matched by expected foaling date and
assigned to one of two treatments groups of six horses each. All mares belonged to the
Michigan State University Horse Teaching and Research Center. Sixteen mares were
originally required for this study, however four mares were removed prior to being
assigned to a treatment group due to dystocia or other foaling problems that would
jeopardize the health of the foal or mare. These animals were not replaced due to a lack
Of pregnant mares. Prior to the study, the mares were housed on pasture, and were not
brought into stalls until about a week before foaling. Mares were monitored at night, and
assisted in foaling by students and staff at the Horse Teaching and Research Center.

Once foaling occurred, the horses were pair-matched by actual foaling date and
placed into a randomly assigned predetermined treatment group. The treated group (S)
received approximately 2% of the diet of EQUI-SiTM as fed (.22 kg/day). The control
group (C) received 0% EQUI-SiTM in their diet. Mares were fed to maintain body

condition, with a 56% roughage (by weight) and 44% total concentrate diet, which met

27

 

Table 1. Calculated daily feed cormmsition for mares on an as fed basis

 

Nutrient Alfalfa Grass Hay Corn Oats Protein Molasses
Hay and Forage Pellet
DE 3.76 Mcal 10.2 Mcal 5.07 Meal 8.91 Meal .41 Mcal .73 Meal
CP 310 g 690 g 140 g 370 g 260 g 20.7 g
Ca 22.8 g 13.6 g .75 g 2.50 g 8.78 g 2.37 g
P 4.37 g 13.6 g 4.05 g 10.6 g 6.50 g .322 g
Mg 5.82 g 4.91 g 1.65 g 4.38 g 4.55 g 1.01 g
Cu 29.3 mg 119 mg 5.55 mg 18.7 mg 26 mg 17 mg

 

Table 2. Calculated total daily nutrients for mares on an as fed basis

 

Nutrient Nutrient Content Calculated Total Daily
Intake

DE 2.24 Meal/kg 29.1 Meal

CP 13.7% 1790 g

Ca .39% 50.8 g

P .30% 49.4 g

Mg .17% 22.3 g

Cu 14.6 mg/kg 216 mg

 

alfalfa mix. The concentrate consisted of 54.5% oats, 26.1% corn, 11.4% protein pellets
(Kent Feeds Inc., Muscatine, IA), 4.0% wet molasses and 4.0% EQUI-SiTM if treated.
The mares were individually fed 5.7 kg of concentrate and 7.3 kg of hay daily divided
into two equal feedings at 0700 and 1600. All concentrate was consumed at each meal,
and when hay was not consumed during the morning meal, additional hay was provided,
and all hay was eaten overnight. The horses were maintained on their respective diets for
45 d post-foaling. Water and mineral blocks were provided to the horses at all times.
Mares and foals were housed in stalls that were 3.1 m x 3.1 m or larger from 1600

until 0800 the next morning. Beginning d 2 post-foaling, mares and foals were turned out

28

Mares and foals were housed in stalls that were 3.1 m x 3.1 m or larger from 1600
until 0800 the next morning. Beginning d 2 post-foaling, mares and foals were turned out
during the daytime into paddocks with an average stocking rate of three pairs per
paddock. Paddocks were approximately 2384.5 m2, with brome, orchard, timothy and
Kentucky blue forage. Mares and foals received health care routine for Michigan,
including deworming and farrier work prior to study.

Growth Measurements

Foal weight was determined using a portable scale (Tm-Test, Inc., San Antonio,
TX), and mare weight was estimated using a weight tape placed in the heart girth region
at the wither on d 0 and d 45. Height at the withers and hip for all animals was found
using a leveled measuring stick at the point of the withers and the tallest portion Of the
croup, respectively.

Sample Collection

Blood samples were collected nine h post-feeding (1600) from the mares and
foals by jugular venipuncture on d 0, 15, 30 and 45. Day 0 represented the ﬁrst 24 h after
parturition. Two 10-mL Vacutainer" tubes with no coagulant and one 7-ml tube with
KZEDTA (Becton Dickinson and Company, Franklin Lakes, NJ), and either 22 g x 2.5 cm
disposable needles for the foals or 18 g x 3.8 cm disposable needles for the mares
(Becton Dickinson and Company, Franklin Lakes, NJ) were used to collect blood.

Finally, approximately 10 mL of colostrum or milk samples were collected on d 0, 15, 30

29

and 45 from the mares. The udder was milked into acid washed plastic narrow-mouth
250—mL collection bottles (Nalgene Company, Rochester, NY).

The blood samples were centrifuged for 10 min for serum and for 20 min for
plasma at 754 x g for both. After centrifugation, the serum samples were aliquotted into
plastic 1.5-mL microcentrifuge tubes (DOT Scientiﬁc, Inc., Burton, MI) using disposable
glass Pasteur pipettes (VWR Scientiﬁc, West Chester, PA). The plasma samples were
withdrawn using acid wasth plastic Samco brand transfer pipettes (Corning Samco
Corporation, San Fernando, CA), and transferred into plastic 1.5-mL microcentrifuge
tubes in a plastic-lined box to avoid dust contamination. This box was only Open to the
front, and this Opening had a sheet of plastic wrap to serve as a curtain to prevent dust
contamination. Milk and colostrum were handled and stored in a similar manner as
plasma Serum, plasma and milk samples were then stored at -20°C until analysis.
Minimizing Silicon Contamination and Special Cleaning of Labware

Glass vacuum collection tubes with one needle were used to reduce trauma to the
foal during sample collections. Various collection methods were used on two mares in
this study, to collect plasma for Si analysis. These results are discussed in appendix E.

Two to three d prior to collection, all plastic equipment, including the milk
collection bottles and the Samco transfer pipettes, was specially washed using a 10%
nitric acid wash, using double distilled deionized water (dd H20) to dilute the nitric acid

(HNO3, EM Science, Gibbstown, NJ). The equipment was then rinsed three times with

30

dd H20, and stored in a fume hood to allow for drying and to minimize Si contamination
from dust. The plasticware was shielded from dust when not in use.
Biochemical Marker Analysis

Osteocalcin was determined using an enzyme-linked immunosorbent assay
(ELISA) (NovoCalcinQ, Metra Biosystems, Inc., Mountain View, CA) following
manufacturer’s instructions. Semm was diluted at 1:4 for mares and 1:15 for foals.
Dilutions were determined by testing samples at dilutions of 1:2, 1:4, and 1:6 for mares,
and 1:10, 1:15, and 1:20 for foals. These values were compared to the standard curve
provided by Metra Biosystems. When the samples fell along the linear portion Of the
curve (from .4 to 1.5 Optical density - OD), the corresponding dilution was chosen to run
the OC assay for all samples. Standards and controls were reconstituted with .5 mL of
IX wash buffer. Twenty-ﬁve 11L of standard, control, or sample was added to the
osteocalcin-coated strips. Samples were tested in duplicate. Anti-osteocalcin ( 125 uL)
was also added to each well. A two-h incubation at room temperature followed, and then
the wells were washed with 300 ILL of IX wash buffer three times, with three min
between each washing. Reconstituted enzyme conjugate (150 uL) was added to the wells
and incubated for one h at room temperature. After incubation, the wells were then
washed in triplicate with 300 LIL each time, and three min between each wash. Finally, a

working substrate was added (150 11L) to each well and incubated at room temperature

31

until the A standard read 1.5 OD, at which point 50 ML of stop solution was added, and
the plate was read at 405 nm optical density on a Spectra Max 340 plate reader
(Molecular Devices Corp., Sunnyvale, CA).

Serum PYD was quantiﬁed using an ELISA (Serum PYD”, Metra Biosystems,
Inc., Mountain View, CA) according to manufacturer’s directions, but modiﬁed for
dilutions. Dilutions Of 1:8 for d 0 and 1:5 for d 15, 30 and 45 for the foal were required
to ﬁt the standard curve. Mare samples were not diluted. Standard was prepared by
diluting the 120 nmol/L PYD standard 1:10 with assay buffer (12 nmol/L), and then
making a 1:2 dilution series for standards 6, 3, 1.5, and .75 nmol/L, with the zero
standard being only assay buffer. Controls were diluted 1:10 with assay buffer. Care
was taken to avoid sample exposure to light, since PYD is light sensitive. Samples were
centrifuged in 30k MWCO Spinﬁlter tubes at 3000 x g for 30 min. After equilibrated to
room temperature, coated strips were ﬁlled with 50 [IL of Reagent 1, 25 ILL of standard,
control or ﬁltered sample, and 75 uL of cold PYD antibody. Samples were tested in
duplicate. The plate was then gently mixed on a plate mixer for one min and incubated in
the refrigerator (2-8°C) in a box overnight (21.5 h). The wells were then emptied and
washed with 250 1.1L of IX wash buffer three times with three min between washes.
Enzyme conjugate (150 ILL) was then added to all wells, and the plates were incubated

for one h at room temperature in a box. Another triplicate wash with 250 ILL of IX

32

buffer in three min intervals followed, and then 150 ILL of working substrate solution was

added to each well. A ﬁnal incubation of 40 min at room temperature in a box followed,

and then 100 ILL of stop solution was added to each well. The plates were read at 405 nm
Optical density on a Spectra Max 340 plate reader (Molecular Devices Corp., Sunnyvale,
CA).

A radioimmunoassay (RIA) was used to measure serum ICT P (ICTP, 125I RIA
Kit”, Diasorin, Stillwater, MN), and was performed according to manufacturer’s
directions. All samples, standards, controls , nonspeciﬁc binding tubes, and total count
tubes were run in duplicate. The tubes contained the following: nonspeciﬁc binding
tubes — 100 111.. sample and 200 ILL dd H20; standards (0 and A-F) — 100 ILL standard and
200 [AL ICTP Antiserum; controls and samples - 100 ILL serum and 200 11L ICTP
antiserum. Tracer (200 uL of 125I ICTP) was added to all tubes, including the total count
tubes. Tubes were mixed and incubated for two h at 37°C. Next, 500 ILL of separation
reagent was added to all tubes except the total count tubes. Tubes were vortexed and
incubated for 30 min at 20-25°C. After incubation, tubes were centrifuged for 30 min at
2000 x g at 4°C, and immediately decanted for the supernatant. Tubes were then read
using the 1290 GammaTrac Gamma Counting System (Tm Analytic, Elk Grove Village,

IL).

33

Silicon Analysis

Plasma and milk samples were analyzed by Trace Elements Laboratory, LHSC
(University Campus, 339 Windermere Road, London, Ontario N6A 5A5, Canada) for Si
concentrations using electrothermal atomic absorption spectrometry as adapted from
Leung and Edmond ( 1997).

Calcium and Phosphorus Analysis

Approximately 2.5 g Of milk was weighed into a 250-mL, acid-washed, conical
beaker, and ashed with 15 ml. of nitric acid (Trace Element Grade, Fisher Scientiﬁc, St.
Louis, MO) and 3 mL of perchloric acid (Trace Element Grade, Fisher Scientiﬁc, St.
Louis, MO). Samples were diluted with dd H20, and stored in polystyrene test tubes
(Sarstedt Inc., Newton, NC).

Calcium concentrations were determined using ﬂame atomic absorption
spectrometry (Unicam 989 AA Spectrophotometer, Thermo J arrell Ash Solutions,
Franklin, MA) following a 1:250 dilution with lanthanum chloride (1% La). Inorganic
standards (Fisher Scientiﬁc, St. Louis, M0) were prepared similarly to samples, and a
standard reference material (Peach Leaves Standard, National Institute of Standards and
Technology, Gaithersburg, MD) was analyzed as per the samples to insure accuracy Of
the instrument. Each sample was run in duplicate.

Phosphorus concentrations of milk samples were determined using spectrometry.

PhOsphorus standards (Sigma-Aldrich Chemical CO., St. Louis, MO) and samples were

34

diluted by a factor Of 75 using dd H20. For the assay, 2.5 mL of molybdate-sulfuric
solution and .25 mL of elon were added to all standards, blanks and sample tubes. After
the incubation at room temperature for 45 min, standards were read at 700 nm on a DU
7400 Spectrophotometer (Beckman Instruments, Palo Alto, CA) using the sipper
assembly to establish a standard curve. A standard reference material (Peach Leaves
Standard, National Institute of Standards and Technology, Gaithersburg, MD) was
analyzed as per the samples to insure accuracy of the assay. Each sample was run in
duplicate, with each day being run twice, for a total Of four concentrations, which were
pooled to report average milk P concentrations.
Statistical Analysis

Treatment effects on Si, Ca, P, OC, ICTP, PYD and growth measurements were
analyzed as a two-factor AN OVA for repeated measures using the Proc Mixed procedure
of SAS (6.12) (SAS, 1997). The model included treatment, day and the interaction
between treatment and day. Least Square Means (LSMeans) and Standard Error Of the
Mean (SEM) were determined. Correlations between plasma Si concentrations (both
mare and foal) and milk Si concentrations were made using the Proc Corr procedure of
SAS (6.12) (SAS, 1997). The correlations were made by day, and were reported as r2
with P-values. For all analyses, a P-value of less than .05 was considered signiﬁcant, and

trends were noted at a P-value of less than or equal to .l.

35

Milk Ca concentrations were different at d 0 between treatment groups, and
therefore each day was standardized by subtracting the day from 0, allowing each horse
to serve as its own control. Proc Mixed was then utilized to report the change from d 0,
and LSMeans and SEM were reported. P-values of less than .05 were still considered

signiﬁcant, and trends were noted at a P-value of less than or equal to .1.

36

RESULTS AND DISCUSSION: EXPERIMENT l

EQUI-SiTM and Plasma Silicon Concentrations of Mares

Our ﬁrst objective was to determine if EQUI—SiTM supplementation would
increase plasma Si concentrations in Arabian mares. Plasma Si concentrations in the S
mares increased from d 0 to d 15 (P = .04), to d 30 (P = .001) and to d 45 (P = .05) (Table
3). Additionally, there was a trend for plasma Si concentrations to be higher at d 30 than
(1 15 in S mares (P = .07). Sodium zeolite A supplementation in weanling Quarter Horses
(Frey et al., 1992) and yearling Quarter Horses (Nielsen et al., 1993) resulted in higher
plasma Si concentrations in treated animals when compared to control horses. In the
current study, no change in plasma Si concentrations was seen in C mares (Table 3). On
(1 30, the S mares had higher Si concentrations than C mares (P = .01), and on d 45, S
mares tended to have higher plasma Si concentrations than C mares (P = .10). These
results illustrate that supplemental EQUI-Sim, at 2% of the diet, increases plasma Si
concentrations in S mares when compared to C animals after 30 d of supplementation.

Table 3. Means Of plasma Si concentrations (ug/L) in mares

 

 

 

Day
Treatment 0 15 30 45 SEM
Control (C) n=6 1339 1398 1335" 1263 108
Si Treated (S) n=6 1201' 1488b 1739by 1515b 108

 

at’days with different superscripts within rows differ (P g .05)
”treatments with different superscripts within columns differ (P 5 .01)

37

We concluded that EQUI-SiTM increased plasma Si concentrations in S mares (d 30)
when compared to C mares in this study.
EQUl-SiTM and Milk Silicon, Calcium and Phosphorus Concentrations

We did not know whether supplemental Si would increase milk Si concentrations
in horses, or if so, how long this process would take. Therefore, the next objective Of this
study was to determine if supplemental Si would increase the Si content in mares' milk.
NO studies to date have addressed this issue. In one study of mineral supplementation in
10 mares fed a high Cu and Zn diet, no differences were seen in milk Cu levels or mare
and foal serum Cu levels between treatment groups (Baucus et al., 1989). In the current
study, milk Si concentrations increased from d 0 to d 45 in the S mares (P = .001) (Table
4). Furthermore, on d 45, S mares’ milk had higher Si concentrations then C mares’ milk
(P = .002). This would indicate that Si was transferred into mammary tissue, and
subsequently being passed through to the milk. Therefore, supplemental Si does increase
milk Si concentrations. In the current study, Si concentrations in the milk Of control

Table 4. Means of milk Si concentrations (ug/L)

 

 

 

Day
Treatment 0 l 5 30 45 SEM
Control (C) n=6 224 460 393 265x 128
Si Treated (S) n=6 242‘ 468' 547'” 896by 128

 

“days with different superscripts within rows differ (P 5 .05)
“treatments with different superscripts within columns differ (P _<_ .01)

38

mares were not statistically different by day (Table 4). However, there was a numerical
increase in milk Si concentrations from d 0 on d 15 and 30. Due to high variability, this
difference was not signiﬁcant. I hypothesize that supplemental Si is necessary to change
milk Si concentrations, and with higher experimental units, this numerical increase in the
control group would not be present. For example, in human milk, Anderson (1993) also
found no changes in Si concentrations over time in women who were not supplemented
with Si. Therefore, unless horses are provided with an increased amount of Si during
lactation, Si concentrations in milk will remain unchanged.

Adler et al. ( 1986) found that when male rats were injected with 31Si, Si was
quickly cleared from plasma and deposited in tissues. Labeled Si was found in highest
concentrations in the kidney, followed by moderate deposition in the liver, lungs, bone,
skin and Spleen. Finally, low deposition of 3"Si was detected in skeletal and cardiac
muscle and testes, and no deposition of labeled Si was found in the brain. Because all
rats in this study were male, no deposition of 31Si in mammary tissue was analyzed (Adler
et al., 1986). In the current experiment, I hypothesized that Si was circulated through the
body and transferred to mammary tissue, thereby increasing milk Si concentrations. The
S mares may have passed this Si on to the foal via milk to allow the foal to deposit Si in
tissues or excrete the excess. In mares, the mineral concentration of milk is highest
during early lactation (ﬁrst 4 to 6 weeks), and then decreases over time (Schryver et al.,

1987). Based on previously cited literature, as the mineral content in milk decreases,

39

minerals may be utilized in other tissues, such as bone and collagen, rather than the
mammary system.

To further determine the inﬂuence of EQUI-SiTM of milk mineral concentrations,
Ca and P levels were determined for all milk samples. Milk Ca concentrations were
different between S and C groups on d 0. This may be due to limited numbers of
experimental units in this study, however, randomization was not of concern, as mares
were assigned treatment groups prior to lactation. Calcium concentrations on d 15, 30
and 45 were standardized by subtracting each day from d 0, to allow each mare to serve
as her own control. Milk Ca concentrations were then reported as mean differences from
d 0. Actual data for milk Ca is reported in Appendix A Table 1A. Treated mares (S) had
higher Ca concentrations on d 15 (P = .002), and tended to be higher on d 30 (P = .08),
relative to d 0 (Table 5) than C mares. These differences in milk Ca concentrations
between S and C mares may be a result of a potential Ca—Si interaction. Additionally, Si
has been found to increase the bone mineral content in tibias of chicks fed a low Ca diet

Table 5. Means of differences from d 0 for milk Ca concentrations (mg/g)

 

 

 

Day
Treatment 15 30 45 SEM
Control (C) n=6 .27 " .14 .24 .08
Si Treated (S) n=6 .66'ly .35" .36" .08

 

‘bdays with different superscripts within rows differ (P 5 .01)
“treatments with different superscripts within columns differ (P _<_ .01)

(Carlisle, 1974), suggesting that Si and Ca may compete or interact to alter bone
mineralization. This difference between S and C groups for milk Ca may also indicate
that S mares absorbed more Ca from the gut, allowing for increased milk Ca output. In
post-partum mares, milk Ca concentrations were found to increase for the ﬁrst two (1 after
foaling (Rook et al., 1997), and were highest in mares' milk the ﬁrst wk of lactation, and
then began to decrease as lactation progressed (Schryver et al., 1986). Milk Ca decreased
20, 30, 27, 22 and 12%, respectively, from the ﬁrst wk through wk six of lactation
(Schryver et al., 1986).

The role of EQUI-Sim on milk P concentrations was also investigated. On (1 15,
S mares' milk contained higher P concentrations than C mares' milk (P = .05) (Table 6).
This could perhaps be due to increased P absorption from the gut, or increased efﬁciency
of P by the mare. This may also suggest a potential Si-P interaction, however, more
research is needed to examine this potential interaction. Milk P levels were found to
increase through d 2 of lactation (Rook et al., 1997) and were highest for the ﬁrst wk,
followed by a decline in P concentrations over time (Schryver et al., 1986). Milk P levels

Table 6. Means of P concentrations (mg/g) in milk

 

 

Day
Treatment 0 15 30 45 SEM
Control (C) n=6 .97'I .68” .61‘ .62"c .03
Si Treated (8) n=6 .92' .76"y .61c - .59c .03

 

I""days with different superscripts within rows differ (P 5 .05)
"treatments with different superscripts within columns differ (P _<_ .05)

41

decreased to the same degree as milk Ca levels from the ﬁrst wk of lactation through wk
9 six (Schryver et al., 1986).
EQUI-SiTM and Plasma Silicon Concentrations of Foals

Since Si was increased in milk concentrations from the S treated mares, the next
Objective was to determine if an increase in foal plasma Si concentrations could be
Observed. Plasma Si concentrations increased from d 0 to d 45 in S foals (P = .02) (Table
7). By d 45, S foals had higher plasma Si concentrations than the C group (P = .004).
Plasma Si concentrations in S foals did not show a Signiﬁcant increase over C foals until
d 45, which was the last day of the study. Coupled with these data are the results Of the
milk Si concentrations, where a treatment effect between S and C was not seen until (1 45
(Table 4). There was a Strong positive correlation between milk Si levels and foal plasma
Si concentrations on d 45 (r2 = .834) (P = .0007). NO change in plasma Si concentrations
was seen in C foals (Table 7). This is in agreement with the results of the plasma Si
concentrations Of control mares in the current study. Therefore, supplemental Si fed to
mares increased Si concentrations in plasma of nursing foals.

Table 7. Means of plasma Si concentrations (pg/L) in foals

 

 

 

Day
Treatment 0 15 30 45 SEM
Control (C) n=6 1406 1239 1366 1146" 166
Si Treated (S) n=6 1337‘ 1463’I 1658‘” 1887"y 166

 

“days with different superscripts within rows differ (P s .05)
“treatments with different superscripts within columns differ (P _<_ .01)

42

Foal plasma Si concentrations were not different between treatment groups until d
45. It is possible that plasma Si concentrations in the S foals were not increased until (1
45 because, intake from milk of S treated mares was not great enough to elicit an increase
in the plasma Si concentration in the S treated foals. A longer Study would be required to
determine if elevated plasma Si concentrations could be maintained over time in the S
foals.
EQUI-SiTM and Bone Metabolism of Mares and Foals

Silicon supplementation has been found to have effects on bone metabolism in
vitro, however limited research has been done with Si supplementation in vivo and its
role in bone metabolism. Researchers have reported that Si enhanced osteoblastic
activity in vitro (Brady et al., 1991; Keeting et al., 1992), and inhibited osteoclastic
activity in vitro (Schiitze et al., 1995). Additionally, the researchers reported that bone
metabolism was enhanced in rats when supplemented with sodium metasilicate, as
evidenced by increased alkaline and acid phosphatase activity (Seaborn and Nielsen,
1994). However alkaline phosphatase activity is an indicator for a variety of metabolic
activities, and therefore may not be an accurate marker of bone formation in vivo. Still,
these studies indicate that Si might have a signiﬁcant role in bone metabolism, and
therefore our final Objective of the current study was to determine if supplemental Si
altered bone metabolism in lactating mares and foals. Osteocalcin in S mares tended to

increase from d 0 to d 30 (P = .08) (Table 8), and to d 45 (P = .07). Treated mares tended

43

Table 8. Means of OC concentrations (ng/mL) in mares

 

 

Day
Treatment 0 15 30 45 SEM
Control (C) n=6 14.9 13.3 11.1" 10.8" 2.9
Si Treated (S) n=6 10.7‘ 16.4‘” 18.0"y 18.4by 2.9

 

“days with different superscripts within rows tend to differ (P _5 .1)
"treatments with different superscripts within columns tend to differ (P g .1)

to have higher OC levels than C mares on d 30 (P = .1) and d 45 (P = .07). This trend
suggests that EQUI-SiTM may have a role in accelerating bone formation by enhancing
osteoblastic activity, however higher numbers of animals, more replicates, or a longer
study may be necessary to determine a statistically signiﬁcant effect.

In women, bone mineral density and markers Of bone formation, including OC,
increased after termination Of lactation (T ropeano et al., 1992). The effect of lactation
remained unclear in the Tropeano et al. (1992) study. However, the increase in bone
mineral density after lactation had stopped indicated that lactation yielded a negative
effect on alteration Of bone mineral density in postpartum women (T ropeano et al., 1992).
Further research should be conducted to investigate the inﬂuence Of lactation on bone
mineral density, and if supplemental Si may be utilized to prevent bone mineral loss
during lactation.

The rate at which Si effects OC in mares when given orally has not been
established. In a study where six Arabians were orally dosed with dexarnethasone at .2

mg/kg of body weight, OC concentrations were lower at two h through 72 h, when

compared to pretreated values (Geor et al., 1995). Dexamethasone is a glucocorticoid
that decreases OC concentrations in humans and laboratory animals, and oral
supplementation produced a rapid change in OC values in the Geor et al. (1995) study. It
has not been established how quickly oral supplementation Of Si effects OC in mares. In
the current experiment, the study length or animal variability may have prevented
treatment effects on OC in S mares.

We then analyzed the effects of supplemental Si on bone turnover in newborn
foals. All foals, regardless of their dam's treatment, had increased OC over the 45 d study
(P = .0008), by increasing from d 0 to d 15 and remaining high through the rest of the
study (Table 9). This increase in OC for both S and C foals is seen immediately after
birth and is evident by d 15, as the foals begin to apply load, or strain, on their skeletal
system. Loading on bone has been shown to increase bone turnover, and yields greater
bone mineral density and strength (Woo et al., 1981; Raub et al., 1989). This rapid rate
Of bone turnover will occur in young animals until they have reached skeletal maturity.
Osteocalcin has been Shown to decrease with age in humans (Kruse and Kracht, 1986),

Table 9. Means of OC concentrations (ng/mL) in foals

 

 

 

Dag
Treatment 0 15 30 45
Control (C) 69.4“: 12.9 112.0”: 11.9 108.2“: 12.8 1169"": 12.9

Si Treated (S) 61.4“: 12.9 122.7": 11.9 111.5“: 11.9 98.1””: 11.9

 

“days with different superscripts within rows differ (P 5 .01)
“n = 5; 'n = 6

45

dogs (Allen etal., 1998) and horses (Lepage et al., 1990, 1992), once the skeleton has
reached maturity. Over time, OC values will decrease due to a slowdown in metabolic
bone growth after skeletal maturation has peaked.

Osteocalcin is not affected by gender in horses (Lepage et al., 1992) or dogs
(Allen et al., 1998). Therefore, the gender of the foals was not Of concern in the current
study. Concentrations of OC in this study were similar to that reported by Black et al.
(1997) for one wk Old Standardbred foals.

To further analyze the role of supplemental Si in bone metabolism, ICT P was
studied. Postpartum ICTP concentrations for both C and S mares were constant through
d 15, but decreased by d 30 and 45 (Table 10). There were no differences due to
treatment in ICTP concentrations. This decrease may result from bone resorption
required to mobilize Ca for milk production.

Table 10. Means Of ICTP concentrations (uglmL) in mares

 

 

Day
Treatment 0 15 30 45 SEM
Control (C) (n=6) 17.0" 17.4‘ 10.6“" 9.0" 1 .7
Si Treated (S) (n=6) 15.0‘ 15.7" 11.0” 92" 1.7

 

at’clays with different superscripts within rows differ (P 5 .01)
'SEM = 1.8; n = 5
“n = 4
In a study involving postpartum dairy cows, ICTP concentrations increased from

d 1 after calving through d 4, and peaked at d 5. This peak was followed by a decrease in

ICT P concentrations from d 5 tO d 14 postpartum (Liesegang et al., 1998). The results of

46

ICTP in the current study follow a similar pattern as reported by Lisegang et al. (1998),
although the mares continued resorption through d 15. It was hypothesized that the
resorption seen in the mares was caused by the need to utilize the Ca pool in bone to meet
the increased Ca demands during early lactation. In a study analyzing plasma and milk
concentrations of various minerals prior to foaling, as mares approached parturition, milk
Ca concentrations increased, while serum Ca concentrations remained constant (Rook et
al., 1997). Although Ca absorption rates were not discussed in this paper, this may
explain the need for increased bone resorption from d 0 through d 15 in mares, thereby
utilizing Ca from the resorbed bone to meet this increased Ca demand. However, over
time this need for high Ca may be lessened or the mares are more efﬁcient, and therefore
bone resorption is diminished, which corresponds to the decrease from d 0 to d 30 and d
45 in ICTP concentrations in the current study. It has been established that milk Ca
concentrations decrease after the ﬁrst week of lactation in horses (Schryver et al., 1986).

To determine the role of supplemental Si in bone metabolism of foals, ICTP
concentrations were measured to indicate bone degradation. The concentrations of ICT P
did not change over time in foals from either C or S mares (Table 11). A longer
experiment is necessary to determine if Si would alter bone resorption when
supplemented for an extended period of time. However, overall bone turnover in this
study indicates net bone formation, with an increase in OC and no change in ICTP

cOncentrations over time for all foals. Brady et al. (1991), Keeting et al. (1992) (in vitro)

47

Table 11. Means of ICTP concentrations (uglmL) in foals

 

 

Day
Treatment 0 15 30 45 SEM
Control (C) n=6 34.5 35.7 32.8 29.9 2.6
Si Treated (S) n=6 33.2 36.4 34.9 37.3' 2.6

 

'n = 5; SEM = 2.8

and Seabom and Nielsen (1994) (in rats) indicate that silicon supplementation increased
osteoblastic activity. Additionally, Schiitze et a1. (1995) indicated a decrease in in vitro
osteoclastic activity (bone resorption) with zeolite A supplementation.

Previous reports of ICTP measurements for the young growing foal (ﬁrst 45 d of
life) had not been reported in the literature. The skeletal system is actively turning over
in order to form bone during the rapid grth period of young animals (Parﬁtt, 1984).
The variability Observed in OC and ICTP in foals in the current study may be too great to
determine a treatment effect. A follow-up study with a larger number Of foals should be
performed. A negative correlation was found between ICT P and age (r = -.76) (P <
.0001), in a study analyzing alterations in bone metabolism with age in horses (Price et
al., 1995). Price et al. (1995) found that as horses increased in age, ICTP decreased,
especially through the first four years Of growth. This relationship between ICTP and age
was also found with dogs (Allen et al., 1998). Although age does affect ICT P levels,
gender did not inﬂuence ICTP concentrations in dogs (Allen etal., 1998), and therefore

gender of the foals was not of concern in the present experiment.

48

To further test the role Of Si in skeletal metabolism of horses, PYD and DPD
crosslinks may be analyzed to indicate collagen degradation (Black et al., 1999). Control
mares had an increase in PYD from d 0 to d 15 (P = .04) (T able 12) and tended to
increase from d O to d 30 (P = .09). The S group increased in PYD from d 0 to d 15 (P =
.05) and decreased from d 15 to d 30 (P = .05). On (1 30, C mares showed a trend for
higher PYD levels than 8 treated mares (P = .08) (Table 12).

Table 12. Means of PYD concentrations (ng/mL) in mares

 

 

Day
Treatment 0 1 5 30 45 SEM
Control (C) (n=6) 1.2' 2.0" 2.0"“ 1.5“ .3
Si Treated (S) (n=6) 1.1II 1.9b 1.2'y 1.0‘ .3

 

"’days with different superscripts within rows differ (P 5 .05)
"treatments with different superscripts within columns differ (P g .1)

An increased urinary pyridinoline/creatinine ratio was found in cows at Six (1
postpartum until 14 d to 16 d (Kaidi et al., 1991). On d 20, the ratios were normal for
these cows. The increased excretion Of PYD crosslinks correlated with the decrease in
the size of the uterus during uterine involution, which took four weeks to complete (Kaidi
et al., 1991). Additionally, Kaidi et al. (1991) found that PYD:DPD ratios were highest
at five (1 postpartum, the peak involution period. Since the collagen content of the uterus
increases during gestation, it is possible that the postpartum increase in PYD
concentrations through d 30 in C mares and d 15 in S mares in the current study was due

to uterine involution. The collagen in the uterus is being degraded as the uterus returns to

49

normal size, which releases PYD crosslinks. However, the period for involution for the
mares in the current study appears to be longer than that necessary for the uterus to return
to normal size in dairy cows.

Analysis of PYD in this study also included DPD concentrations.
Deoxypyridinoline is bone speciﬁc, as it is most prevalently found in type I collagen
(James et al., 1996). Therefore, increased PYD concentrations in this study may have
been associated with increased bone resorption, which would be similar to the increase in
resorption observed with ICTP values in the mares through the ﬁrst 15 (1. Additionally,
Liesegang et al. (1998) found an increase in DPD levels in lactating dairy cows at d 2
through d 9 followed by a decrease from d 9 to d 14. The mares in this study showed
elevated PYD concentrations through d 30 in C mares and d 15 in S mares.

Supplemental Si may decrease bone resorption, as an in vitro experiment using puriﬁed
avian osteoclasts treated with zeolite A demonstrated a three-fold decrease in osteoclastic
pits at 24 h after treatment (Schiitze et al., 1995). This would indicate that Si may have
an inhibitory effect on bone resorption. Furthermore, in the current study, S mares may
have had less of a need for bone resorption at that point in lactation, perhaps due to
enhanced mineral utilization within bone. EQUI-SiTM supplementation may have
reduced the need for a high Ca concentration this far into lactation, by utilizing available

Ca more efﬁciently, and help reduce the need for bone degradation.

50

Finally, we determined the effects of EQUI-Sim supplementation on collagen
turnover in foals. In all foals, PYD concentrations decreased from d O to d 15 and
remained low throughout the rest of the study (Table 13). No differences in PYD levels
were seen between foal treatment groups. It is possible that supplemental Si does not
affect collagen crosslink release in growing foals. Silicon may have alternative
mechanisms in collagen, such as increasing the collagen content. However, it is likely
that because it took 45 d for plasma Si levels to be higher in S treated foals (Table 6), a
lack of treatment effect for PYD concentrations in foals may result.

Table 13. Means of PYD concentrations (ng/mL) in foals

 

 

Day
Treatment 0 15 3O 45 SEM
Control (C) (n=6) 16.4' 9.2" 6.2" 6.6" 1.7
Si Treated (S) (n=6) 20.7‘ 7.0" 6.1" 6.7" 1.7

 

“’days with different superscripts within rows differ (P5 .01)

A decrease in PYD concentrations for all foals was observed for the 45 d of this
experiment, and therefore a decrease in collagen degradation was seen over time. This
will allow for net bone deposition since OC levels are increasing. AS horses mature
skeletally, concentrations Of PYD and DPD decrease with time (Black et al., 1997, 1999),
and indicate a logical slowdown in growth as the horse matures. Black et a1. ( 1999)
found that PYD and DPD concentrations were higher in weanling horses than in adults.
This relative slowdown in bone and collagen metabolism corresponds to that seen for

other biochemical markers (Lepage et al., 1990, 1992; Price et al., 1995).

5.1

EQUI-SiTM Supplementation and Foal Growth

Throughout the current study all foals grew taller at the withers and hip and
increased in weight over time (P = .0001) (Table 14). No differences in growth
indicators existed between groups. Therefore, EQUI-SiTM treatment did not cause foals
to grow more rapidly, alleviating any increased risk for growth-related diseases. The
supplementation of feed additives could lead to developmental orthopedic diseases if
foals grow more rapidly than their non-supplemented counterparts. The foals in the
current study seemed to grow at normal rates, as growth measurements in other young
horses follow the same patterns for wither and hip height growth (Black et al., 1997).

Table 14. Means of growth measurements for foals
Growth Measurement Control (C) Si Treated (S) SEM (n=6 per group)

 

Wither Height

D 0 Wither Height (cm) 94.2‘ 91 .6' 1.3

D 45 Wither Height (cm) 109.5b 107.8" 1.3
Gain 15.3 16.2

Hip Height

D 0 Hip Height (cm) 95.8' 93.8‘ 1.5

D 45 Hip Height (cm) 114.3b 112.2" 1.5
Gain 18.5 18.4

Weight

D 0 Weight (kg) 433 42' 2.2

D 45 Weight (kg) 93" 93" 2.2
Gain 50 51

 

“’days with different superscripts per measurement within columns differ (P = .0001)

52

CHAPTER 3

EXPERIMENTAL PROCEDURE: EXPERIMENT 2

Assignment of Treatments and Management of Animals

Ten Arabian and 10 Quarter Horse yearlings were randomly assigned to one of
two treatment groups of 10 horses each. All Arabians were maintained at the Michigan
State University Horse Teaching and Research Center, and all Quarter Horses were
maintained at the Michigan State University Merillat Equine Center. Prior to the study,
the Arabian yearlings were housed on pasture continually, and were therefore brought
into stalls for a three d acclimatization period. During this period, the Arabian yearlings
were stalled at night, fed individually, and turned out during the day. However, prior to
the study, the Quarter Horse yearlings were already housed in stalls, and were being
longe-trained daily, and therefore did not need an acclimatization period.

The treated group (S) received approximately 2% of the diet of EQU'I-SiTM as fed
(.13 kg/day for Arabians and .16 kg/day for Quarter Horse yearlings). The control group
(C) received 0% EQUI-SiTM in their diet. Horses were fed to maintain body condition.
The Arabian diet was approximately 60% roughage (by weight) and 40% total
concentrate, and met NRC (1989) requirements for long yearlings (Table 15 and 16).
The hay was a grass and alfalfa mix. The concentrate consisted of 61.2% oats, 29.5%

corn, 5.0% dairy protein pellets (Kent Feeds Inc., Muscatine, IA), 4.3% dry molasses,

53

and their respective EQUI-SiTM treatment. The Arabians were individually fed 5.6 kg Of
concentrate and 7.3 kg of hay daily divided into two equal feedings at 0700 and 1600.
The Quarter Horse diet was approximately 60 % roughage (by weight) and 40 % total
concentrate, and met NRC (1989) requirements for long yearlings (Table 117 and 18).
The hay was an alfalfa and grass mix. The concentrate consisted Of 1.8 kg of StrategyTM
(Purina Feeds, St. Louis, MO), .45 kg of Equine Energy (Buckeye Nutrition, Dalton,
Ohio), .15 kg of molasses, and their respective EQUI-SiTM treatment. The horses were
individually fed, with each daily ration divided into two equal feedings at 0730 and 1530.
The molasses enhanced palatability, and therefore feed refusal did not occur. All horses

Table 15. Calculated daily feed composition table for Arabian yearlings on an as fed
basis

 

Nutrient Alfalfa Grass Hay Oats Corn Protein Molasses
Hay & Forage Pellet
DE 3.43 Mcal 9.31 Mcal 9.72 Mcal 5.53 .17 Mcal .73 Mcal
Mcal
CP 282 g 627 g 402 g 149 g 108 g 20.7 g
Ca 20.6 g 12.5 g 2.73 g .82 g 3.65 g 2.37 g
P 3.98 g 12.5 g 11.6 g 4.43 g 2.7 g .32 g
Mg 5.31 g 4.48 g 4.77 g 1.80 g 1.89 g 1.01 g
Cu 26.7 mg 109 mg 20.5 mg 6.05 mg 10.8 mg 17.0 mg

 

Table 16. Calculated daily nutrient table for Arabian yearlings on an as fed basis

 

 

Nutrient Nutrient Content Calculated Total Dailj Intake
DE 2.35 Meal/kg 28.9 Meal

CP 12.9% 1590 g

Ca .35% 43 g

P .29% 36 g

Mg .16% 19 g

Cu 15.45 mg/kg 190 mg

54

Table 17. Calculated daily feed composition table for Quarter Horse yearlings on an as
fed basis

 

Nutrient Alfalfa Hay Grass Hay Equine Stategy Molasses
& Forage Athlete
DE 3.43 Mcal 9.31 Mcal .572 Mcal 1.24 Mcal .73 Mcal
CP 282 g 627 g 54 g 255 g 20.7 g
Ca 20.6 g 12.5 g 3.83 g 13.7 g 2.37 g
P 3.98 g 12.5 g 1.8 g 10.9 g .32 g
Mg 5.31 g 4.48 g - - 1.01 g
Cu 26.7 mg 109 mg 9.09 mg 100 mg 17.0 mg

 

Table 18. Calculated daily nutrient table for Quarter Horse yearlings on an as fed basis

 

Nutrient Nutrient Content Calculated Total Daily Intake
DE 1.72 Meal/kg 15.3 Mcal

CP 12.8% 1240 g

Ca .55% 53 g

P .32% 30 g

Mg .11% 10 g

Cu 27 mg/kg 262 mg

 

were maintained on their respective diets for 45 d. Water and mineral blocks were
provided for the horses at all times.

The Arabian yearlings were housed in Stalls that were 3.1 m x 3.1 m or larger
from 1600 until 0800 the next morning. They were turned out during the daytime into
paddocks with an average stocking rate of 3.3 per paddock. Paddocks were
approximately 2385 m2, with brome, orchard, timothy and Kentucky blue grass forage.
The Arabians were dewormed at the start of the study using a commercially available
anthelmintic, but were not dewormed for the duration of the study, and received farrier

work prior to commencing the study. The Quarter Horse yearlings were housed in stalls

55

that were 3.7 m x 3.7 m. They were exercised daily with the following exercise protocol:
wk 1 - 30 min on mechanical walker at walk; wk 2 - 30 to 45 min at trot on mechanical
walker; wk 3 and 4 — 10 circles at the trot and 10 circles at the lope in each direction on
the longe line; wk 5 and 6 — 30 min total, with approximately 10 circles at trot and 20
circles at the lope in both directions on the longe line. The Quarter Horses were on
Strongid" C (Pﬁzer Animal Health) Daily Dewormer prior to the start of the study, but
were not dewormed for the duration of the study, and had their feet trimmed prior to the
start of the study.
Growth Measurements

Weight was determined using a weight tape placed in the heart girth region at the
withers. Wither and hip height for all animals was determined using a leveled measuring
stick at the point of the withers and the tallest point of the croup, respectively. Hoof
measurements were made by using a measuring tape, from the coronet to the bottom of
the toe, to measure toe length, and from the coronet to the bottom of each heel bulb, for
the right and left heel measurements. Results of hoof measurements can be found in
Appendix D Tables 1 through 3. All measurements were taken on d 0 and d 45.
Sample Collection

Blood was collected as reported in experiment one, following protocol for
collection time, days, and method. In addition, approximately 10 mL of blood was

Withdrawn using 18 g disposable sterile needles into a syringe (Luer-Lok TM, Becton

56

Dickinson and Company, Franklin Lakes, NJ) that was specially cleaned in a 10% nitric
acid wash. The blood was then immediately transferred from the syringe into a
polyurethane, Si-free collection tube (Falconc’, Becton Dickinson and Company, Franklin
Lakes, NJ) with 115 ILL of KZEDTA in the bottom. The needle was punctured through
paraﬁlm covering the collection tube to prevent contamination, and the blood was mixed
gently. This blood was used to obtain plasma for Si analysis. In the laboratory, the
paraﬁlm was removed from the polyurethane tubes within a special plastic-lined box, and
the tubes were capped for centrifugation.

Sample centrifugation was performed as reported in experiment 1. After
centrifugation, the serum samples were aliquotted into plastic 1.5-mL microcentrifuge
tubes (DOT Scientiﬁc, Inc., Burton, MI) using disposable glass Pasteur pipettes (VWR
Scientiﬁc, West Chester, PA). The plasma samples were withdrawn using acid washed
plastic Samco brand transfer pipettes (Corning Samco Corporation, San Fernando, CA),
and transferred into plastic 1.5-mL microcentrifuge tubes (DOT Scientific, Inc., Burton,
MI) in a plastic-lined box to avoid dust contamination. Serum and plasma were then
stored at —20°C until analysis.

Fecal samples were collected on d 45 by taking one fresh fecal ball out of each
horse's Stall. Fecal samples were then refrigerated until parasite analysis. Fecal samples

were analyzed for parasites by the Animal Health Diagnostic Laboratory (College of

57

Veterinary Medicine, Michigan State University, PO Box 30076, Lansing, MI 48909).
Samples were ﬁrst qualitatively analyzed, and if eggs were present, then quantitative
analysis was completed. The McMaster test was utilized for parasitic analysis. Results
of the parasite analysis can be found in Appendix D Tables 4 and 5.
Minimizing Silicon Contamination and Special Cleaning of Labware

The procedure used in experiment 1 was also followed in this experiment to
prevent Si contamination. The plasma samples from all yearlings were handled in the
same manner as the plasma samples taken with the vacuum tubes in experiment 1.
Biochemical Marker Analysis

Serum was analyzed for OC, ICTP and PYD as reported in experiment 1.
Silicon Analysis

Plasma Si was again analyzed by Trace Elements Laboratory, London Health
Sciences Centre (refer to experiment 1).
Statistical Analysis

Data were summarized as LSMeans and SEM. Treatment effects on Si, OC,
ICTP, PYD and growth measurements were analyzed as a two-factor ANOVA for
repeated measures using the Proc Mixed procedure of SAS (6.12) (SAS, 1997). The
class included horse, day, treatment and farm, and the model included treatment, day,
farm, time by treatment, treatment by farm, time by farm, and day by treatment by farm

interactions. If the day by treatment by farm interaction was greater than P 2 .3, then data

58

were pooled from both farms and summarized as LSMeans and SEM. However, if this
interaction was less than P < .3, as was the case with plasma Si levels, then a Student’s t-
test was applied to examine differences between the farms by day. If (1 O was different
for the two farms, then farms were pooled and d 15, d 30 and d 45 values were
standardized by subtracting them from d 0, allowing each horse to serve as its own
control. Once pooled, the differences from d 0 were reported as LSMeans and SEM for
each treatment group. For all analyses, a P-value of less than .05 was considered

signiﬁcant, and trends were studied at a P-value of less than or equal to .1.

59

RESULTS AND DISCUSSION: EXPERIMENT 2

EQUI-SiTM Supplementation and Plasma Silicon Concentrations

Supplemental Si has been shown to increase plasma Si concentrations in horses
(Frey et al., 1992; Nielsen et al., 1993). Therefore, out ﬁrst objective was to determine if
Equi-SiTM supplementation would increase plasma Si concentrations in Arabian and
Quarter Horse yearlings. Plasma Si concentrations were found to be different on d 0 at
the two farms (P < .0001 ), and therefore, all plasma Si concentrations for all yearlings
were standardized by subtracting them from d 0. This allowed each horse to serve as its
own control, and then all yearlings were pooled to determine treatment effects. Plasma Si
levels showed a greater difference from d 0 on d 15, 30 and 45 (P = .0001 for all days) in
S yearlings when compared to the C yearlings (Table 19). Equi-SiTM supplementation
increased plasma Si concentrations in S yearlings, which indicates that Equi-SiTM is an
absorbable form of Si.

Table 19. Means of differences from day 0 for plasma Si concentrations (rig/L) in

 

 

 

yearlings

Day
Treatment 1 5 3O 45 SEM
Control (C) (n=6) -70.9"‘ 84.0“ 0.1“ 49.7
Si Treated (S) (n=6) 404.8‘Iy 509.9by 326.9“y 49.7

 

“days with different superscripts within rows differ (P 5 .05)
x"treatments with different superscripts within columns differ (P 5 .01)

60

Interestingly, the Arabian and Quarter Horse yearlings in this study had different
plasma Si values on d O. This difference on d 0 may Simply have been due to variable
environmental or management factors. Silicon is very abundant in the environment, and
is naturally found in grains and grain fractions (Pennington, 1991), and in soil and grass
(Hays and Swenson, 1984). The Quarter Horses had a higher basal plasma Si
concentration (353 jig/L _1; 29) than the Arabians (176 jig/L i 56) (P < .0001) on d O. The
Arabians were on pasture prior to the study, and were turned out daily during the
experiment. Conversely, the Quarter Horses were housed in stalls prior to and during the
study. Therefore, differences in management may have led to different basal plasma Si
concentrations.

EQUI-SiTM Supplementation and Bone and Collagen Metabolism

Since plasma Si concentrations were elevated in S yearlings, our second objective
was to determine if increased plasma Si concentrations would alter bone and collagen
metabolism. Osteocalcin concentrations tended to decrease from d 0 to d 15 (P = .07)
and tended to increase from d 15 to d 30 (P = .06) and increased from d 15 to d 45 (P =
.04) in C yearlings (Table 20). These results were surprising that C increased in OC
concentration on d 45 when compared to d 15. This increase could be due to high
variability, and would likely not be seen with larger numbers of animals. The S group
had no change in OC concentrations over time, and there were no differences in serum

OC between treatment groups.

61

Table 20. Means of OC concentrations (ng/mL) in yearlings

 

 

 

Day
Treatment 0 15 30 45
Control (C) 77.7‘” i 22.9 57.6'” i 23.0 79.8“”i 23.2 90.9”“: 23.4
(n=10)
Si Treated (S) 567‘: 22.9 43.6‘: 22.9 45.1‘“ i 23.1 48.9“: 23.8
(n=10)
“’days with different superscripts within rows differ (P 5 .05)
“n = 9

Markers for bone formation and resorption were increased when weanling
Sprague-Dawley rats were fed normal physiological amounts of Si (Seaborn and Nielsen,
1994). An increase in alkaline and acid phosphatase activity was seen in homogenized
femurs of rats, however Seabom and Nielsen (1994) concluded that the mechanism for
the altered metabolism is still unclear. Additionally, in vitra, zeolite A increased
proliferation of human osteoblast-like cells, thereby enhancing osteoblastic activity
(Brady et al., 1991). Keeting et al. (1992) found that zeolite A increased Transforming
Growth Factor - B production, human osteoblast-like cell proliferation and cellular
alkaline phosphatase activity in vitra. Furthermore, an increase in OC was seen with
zeolite A treatment in vitra (Keeting et al., 1992). However, in vitra studies using zeolite
A may not be a true indicator of how this molecule would behave in viva. In the current
Study, I hypothesized that Si was not deposited in bone for a long enough period of time
to alter osteoblastic activity. This increase in activity may occur by increased cell

proliferation or differentiation, upregulated protein sythesis or other potential

62

mechanisms. Alternatively, it is possible that Si has a role in bone and collagen structure
other than altering bone metabolism, or may have no inﬂuence on bone.

Next we examined if EQUI-SiTM would alter bone degradation by measuring
ICTP concentrations. In the current study, ICTP was lower in S yearlings when
compared to C on d 45 (P = .04) (Table 21). This may indicate that EQUI-SiTM has a role
in decreasing bone resorption. Although resorption is a crucial component Of bone
remodeling, these treated animals appear to be resorbing bone, but to a lesser degree than
C yearlings. Therefore, since OC remains unchanged in S yearlings, net bone formation
should be higher in S. Net bone formation may be seen in S, regardless of the increase
seen for CC levels in C from d 15 to d 45.

Table 21. Means of ICTP concentrations (uglmL) in yearlings

 

 

 

Day
Treatment 0 15 30 45 SEM
Control (C) (n=10) 16.7 16.2 17.1 16.9" .6
Si Treated (S) 16.5“” 16.6“ 16.4'" 15.2by .6

(n=10)

 

“days with different superscripts within rows differ (P 5 .05)
"’ treatments with different superscripts within columns differ (P 5 .05)
m=9
Supplemental Si might affect a mediator in bone that activates osteoclastic
activity. Once this has begun, the osteoclastic enzymes may activate osteoblastic

enzymes, by employing the extensions of the osteoclast cell that reach into the matrix and

sending messages to the osteoblast to stimulate activity. Enzymatic activities in the

63

osteoblast and osteoclast alter each other’s activity (Mohan et al., 1984). Chambers
(1982) found that as osteoclastic activity increases, osteoblasts are activated and begin tO
increase in osteoblastic activity. Therefore, supplemental Si may first act on bone
resorption through altering osteoclastic activity, and this in turn may activate osteoblastic
action. Altering both osteoblastic and osteoclastic activity will alter total bone
metabolism. However, in the current study it took 45 d to alter bone resorption in the
yearlings, and there was not enough time to determine if this decreased resorption would
stimulate bone formation. A longer study would be required to determine if OC release
would be initiated after ICTP release has progressed. In a study demonstrating the role of
supplemental Si in bone resorption, rats were implanted with mineralized or
demineralized bone, taken from other Si supplemented or Si deﬁcient rats (Seabom and
Nielsen, 1994). Seabom and Nielsen ( 1994) found an interaction between Si
supplementation and the implant bone source for both mineralized and demineralized
bone implants in the implanted rats. Less acid phosphatase activity was found in the
implanted bone portion of Si supplemented rat tibias, when compared to the implanted
bone portion from Si deﬁcient rat tibias, indicating decreased bone resorption in the Si
supplemented rats (Seaborn and Nielsen, 1994). In Seabom and Nielsen’s (1994) study,
bone formation was not affected by the implant, as evidenced by no change in alkaline

phosphatase activity.

Since OC concentrations remained unchanged in S yearling horses in the current
study, and ICTP levels decreased in the S group, net bone formation may result. This
could potentially reduce the incidence of bone-related injuries later on during exercise, as
there would be a greater amount of bone. In one study, a trend was Observed in weanling
Quarter Horses treated with sodium zeolite A to have a greater percent change in
radiographic bone aluminum equivalence (RBAE) on the lateral side of the third
metacarpal (P < .1), when compared to the non-supplemented control (Frey et al., 1992).
Frey et a1. (1992) concluded that this trend for a change in RBAE may have been caused
by an interaction of Si and other minerals of the hydroxyapatite during early bone
calciﬁcation. Carlisle (1970) found this type of Si-Ca interaction in young mice and rats,
as Si was concentrated in calciﬁcation sites early during skeletal growth, but as the bone
mineral reached maturity and Ca levels increased, Si levels were only present at minimal
levels. A second study supporting that Si supplementation may potentially decrease the
incidence of bone-related injuries was done on young Quarter Horses in race training
(Nielsen et al., 1993). Nielsen et al. (1993) found that horses supplemented with sodium
zeolite A had faster race times. It was felt that the faster horses in the non-supplemented
group sustained more injuries, and hence, were removed from the study, thereby reducing
the average speed of the non-supplemented group. Nielsen et al. (1993) also found that
supplemented horses ran a greater amount of strides before experiencing a bone-related

injury, when compared to non-supplemented horses and lower-dose supplemented.

65

Therefore, Si may make intensely exercised horses less susceptible to bone-related
failure.

To further test Si's role in bone and collagen metabolism, measurements of semm
PYD concentrations were made. Silicon has increased the collagen content of bone in
vitra (Carlisle and Alpenfels, 1978), and Si deﬁcient chicks had malformed Skulls and
long bones, and smaller or absent combs, which all normally contain a high collagen
content. Therefore, we also studied the role of EQUI-SiTM supplementation in collagen
metabolism. No differences in PYD concentrations were seen by day in the C or S
treated group (Table 22). Furthermore, there were no differences in PYD concentrations
between treatment groups on any day. This lack of treatment effect may be explained by
one of several potential reasons. First, Si may have been deposited in collagen, but did
not have time to alter the turnover of the collagen molecules. This may be possible, even
though an increase in Si was seen between S and C on d 15 (Table 19), because it is
unknown how long it would take to detect differences between groups for PYD, and 45 (1
may not have been long enough to determine differences. Secondly, the increased Si
levels may have been deposited into other tissues first, and therefore, never accumulated

Table 22. Means of PYD concentrations (ng/mL) in yearlings

 

 

Day
Treatment 0 15 30 45 SEM
Control (C) (n=10) 4.8 4.2 4.1“ 4.1 .3
Si Treated (S) (n=10) 4.7 4.1 4.4 4.0 .3

 

“n=9

66

in collagen. Collagen might not have a need for Si at this point in development, and
therefore was not incorporated into the collagen tissue. In growing chicks, when the Si
content of the diet was deﬁcient, collagenous tissues were smaller and abnormal;
however, supplemental Si did not increase collagen content in the normal chicks
(Carlisle, 1972).
EQUI-Sim Supplementation and Yearling Growth

All yearlings in the current study grew taller at the withers and hip, and increased
in weight over time (P 5 .002) (Table 23). No differences existed between treatment
groups for the three growth measurements. This is important, as S treatment did not alter
growth as compared to C yearlings, illustrating that EQUI-SiTM does not increase the risk
of growth-related diseases. Yearlings are still growing animals, and similar growth
measurements are supported by measurements taken in other yearling horses (Wall et al.,

1998).

67

 

Table 23. Means Of growth measurements in yearlings

 

Growth Measurement Control (C) Si Treated (S) SEM (n=10 per group)

Wither Height

D 0 Wither Height (cm) 1378‘ 139.3’ 1.5

D 45 Wither Height (cm) 140.0” 141 .2" 1.5
Gain 2.2 1.9

Hip Height

D 0 Hip Height (cm) 1425' 145.0’ 1.4

D 45 Hip Height (cm) 1431" 146.0” 1.4
Gain .6 1.0

Weight

D 0 Weight (kg) 332‘ 341‘ 9

D 45 Weight (kg) 352° 360° 9
Gain 20 19

 

at’days with different superscripts per measurement within columns differ (P = .002)

68

CHAPTER 4

CONCLUSIONS AND IMPLICATIONS

Enhancing net bone formation may provide a more superior skeleton. Modeling
and remodeling is increased through manipulation of hormonal stimulators, microfracture
repair, or exogenous factors, such as nutritional supplements. Therefore,
supplementation with certain feed additives may have the potential to alleviate bone-
related failure or diseases involving bone loss.

In the present studies, EQUI-Sim, a Si containing feed additive, was fed to horses
of various production levels in hopes of altering bone turnover. Silicon has been shown
in the literature to increase metabolism of cells in vitra (Brady et al., 1991; Keeting et al.,
1992; Schiitze et al., 1995), and to stimulate bone formation and resorption in rats
(Seaborn and Nielsen, 1994). The current studies were designed to examine potential
alterations of bone metabolism due to Si supplementation in the horse. Biochemical
markers were utilized to determine the effect of EQUI-SiTM in achieving altered bone
turnover. While, these markers are indices of systemic trends of bone turnover, they do
not provide evidence as to the quality and/or quantity of the altered bone. Terminal
studies using animal models are required to fully understand the role of supplemental Si

in viva.

69

 

In the current study, EQUI-SiTM increased plasma Si concentrations in mares and
yearlings, and increased milk Si concentrations, thereby increasing plasma Si
concentrations in foals. These increases in plasma Si concentrations illustrate EQUI-
Sim's bioavailability, as it was actively absorbed into blood. The bioavailabilities of Si
sources differ, but EQUI-SiTM appears to be efﬁcacious in increasing plasma Si
concentrations of horses of various ages. Furthermore, this plasma Si is transported
throughout the body, as evident by the increased milk Si concentrations of S mares.

EQUI-SiTM increased milk Ca and P concentrations in S mares on d 15. Increased
Ca and P concentrations in milk can be utilized by the foal for bone mineral, if foals are
consuming adequate milk. However, milk yield was not determined in this study. The
increase in Ca and P may suggest that EQUI-SiTM mobilized Ca and P from bone during
lactation. A second potential reason for the increased Ca and P concentrations in milk on
d 15 may be that EQUI-SiTM increases absorption of these minerals from the gut, thereby
allowing for more output in the milk. Dietary absorption was not examined in these
studies. Further experiments examining the role of EQUI-SiTM in lactating mares are
required to determine supplemental Si's inﬂuence on lactation and bone turnover.

Bone turnover appeared to be altered in mares and yearling horses supplemented
with EQUI-Sim. The trends seen for increased osteoblastic activity and decreased bone

and collagen degradation in S mares may indicate net bone formation. Therefore,

70

stronger bone may result, allowing the mares to be exercised or rebred with little adverse
effects on the skeletal system. The decrease in bone degradation seen in S yearlings may
also indicate net bone formation, as DC was not changed in the S group. A higher degree
of bone formation may equate with stronger bone, allowing for less bone-related injuries
and increasing the longevity of horses. The lack of treatment effect in foals was because
it took 45 d for plasma Si concentrations to be increased in foals, and this marked the end
of the study. A longer study may produce different results than those found in the current
study. Furthermore, normal bone turnover is very rapid during growth, and may be
masking treatment effects in the bone turnover markers in foals, because of high
individual variability.

Silicon may have an alternate mechanism in bone and collagen physiology and
metabolism. One alternative may be that Si changes the hydroxyapatite crystal in size,
structure or content. When other minerals (besides Ca and P) incorporate into
hydroxyapatite, its structural and functional characteristics may be altered (Hays and
Swenson, 1984; Wasserman, 1984). This may impact the strength and integrity of bone.
A second alternative is that Si may alter mineralization by activating or inhibiting the
calciﬁcation process. Carlisle (1970, 1974) proposes that Si may inﬂuence the
mineralization process, but the exact mechanism for this has not been established.

Silicon may alter the collagen crosslinks themselves, by increasing their number or

structure. Additionally, Si may alter the glycosaminoglycan content of connective tissue,

71

by increasing the number or structure of the molecules. Carlisle (1976) found that Si
supplementation in chicks increased the glycosaminoglycan levels in cartilage.

Additional research studying the histology of bone and collagen is necessary to
understand the complete role of Si in these tissues. Silicon may be altering the bone
mineral density, strength and porosity. Furthermore, the degree of Si deposition in bone
and collagen tissue at various stages of growth and maturation still needs to be
determined. Further research is required to examine these alternative mechanisms of Si in
bone and collagen.

Currently, Si requirements for the horse are not included in the NRC (1989).
Continued research may provide evidence for such recommendations. While Si is
naturally found in grains and dirt (Pennington, 1991), not all of this Si is available.
Bioavailable supplements, such as EQUI-Sim, may prove to be beneﬁcial in enhancing
skeletal health, and may aid in alleviating bone-related injuries or diseases involving
bone loss. While many factors govern bone metabolism, supplemental Si may provide a
nutritional means of enhancing skeletal health. More research in the area of Si and bone
is needed to answer questions regarding potential biochemical mechanisms and other

roles for Si in bone physiology.

72

APPENDIX A

73

APPENDIX A

Non-normalized milk Ca concentrations data and AN OVA table (Project 1)

Appendix Table 1A. Non-normalized means of milk Ca concentrations (mg/g) in
Arabian mares

 

 

Day
Treatment 0 15 30 45 SEM
Control (C) (n=6) 1.13“ 1.40bx 1.28“ 1.37” .06
Si Treated (S) (n=6) .928"y 1.59by 1.28c 1.29C .06

 

‘bcdays with different superscripts within rows differ (P 5 .05)
"treatments with different superscripts within columns differ (P 5 .05)

Appendix Table 1B. Proc mixed table for non-normalized milk Ca concentrations (mg/ g)
for Arabian mares

 

Source NDF DDF F-value P-Value
Treatment 1 10 .19 .6753
Day 3 30 3 1.54 .0001
Day*Treatment 3 30 5.68 .0033

 

N on-normalized plasma Si concentrations for yearlings (Project 2)

Appendix Table 3A. Non-normalized means of plasma Si concentrations (jig/L) in

 

 

 

yearlings

Day
Treatment 0 l 5 30 45 SEM
Control (C) (n=10) 249.7” 178.8“ 333.7'“ 249.8“ 32.7
Si Treated (S) (n=10) 278.8’ 683.6"y 788.7cy 605.7by 32.7

 

albcdays with different superscripts within rows differ (P 5 .05)
”treatments with different superscripts within columns differ (P 5 .01)

Appendix Table 4A. T-test for D 0 for plasma Si concentrations for yearlings

 

 

Breed Mean Plasma Si concentration Std Dev Std Error
Arabians (n=10) 175.8 28.7 9.08
Quarter Horses (n=10) 352.7 55.7 17.6

74

APPENDIX B

75

u . ._.»'. b1II-'—'.Iu . .

 

APPENDIX B

Proc Mixed Tables

Appendix Table 1B. Proc mixed table for plasma Si concentrations (pg/L) for Arabian

 

 

mares
Source NDF DDF F-value P-Value

Treatment 1 10 2.90 .1 192 a
Day 3 30 2.35 .0925 i
Day*Treatment 3 30 2.29 .0987

 

Appendix Table 2B. Proc mixed table for milk Si concentrations (pg/L) for Arabian

 

 

 

mares

Source NDF DDF F-value P-Value i 4
Treatment 1 10 4.82 .0528

Day 3 30 2.61 .0699

Day*Treatrnent 3 30 2.60 .0703

 

Appendix Table 3B. Proc mixed table for milk Ca concentrations (mg/g) for Arabian

 

 

mares

Source NDF DDF F-value P-Value
Treatment 1 10 6.75 .0266
Day 2 20 10.30 .0008
Day*Treatment 2 20 2.75 .0877

 

Appendix Table 4B. Proc mixed table for milk P concentrations (mg/g) for Arabian

 

 

mares

Source NDF DDF F-value P-Value
Treatment 1 10 0.00 .9517
Day 3 29 63.90 .0001
Day*Treatment 3 29 3.47 .0287

 

76

Appendix Table 5B. Proc mixed table for plasma Si concentrations (ug/L) for Arabian
foals

 

Source NDF DDF F-value P-Value
Treatment 1 10 3.26 . 1013
Day 3 30 0.67 .5780
Day*Treatment 3 30 2.44 .0840

 

Appendix Table 6B. Proc mixed table for serum OC concentrations (ng/mL) for
Arabian mares

 

Source NDF DDF F-value P-Value
Treatment 1 10 2.57 . 1400
Day 3 30 0.22 .8827
Day *Treatment 3 30 1 .77 . 1732

 

Appendix Table 7B. Proc mixed table for serum OC concentrations (ng/mL) for
Arabian foals

 

Source NDF DDF F-value P-Value
Treatment 1 10 0.08 .7769
Day 3 26 8.91 .0003
Day*Treatment 3 26 0.71 .5570

 

Appendix Table 8B. Proc mixed table for serum ICTP concentrations (til/L) for
Arabian mares

 

Source NDF DDF F-value P-Value
Treatment 1 10 0.1 1 .7436
Day 3 26 21 . 16 .0001
Day*Treatment 3 26 0.61 .6130

 

Appendix Table 9B. Proc mixed table for serum ICTP concentrations (ul/L) for
Arabian foals

 

Source NDF DDF F—value P-Value
Treatment 1 10 0.90 .3660
Day 3 29 0.56 .6468
Day*Treatment 3 29 0.97 .4199

 

77

Appendix Table 10B. Proc mixed table for serum PYD concentrations (ng/mL) for

Arabian mares

 

 

Source NDF DDF F-value P-Value
Treatment 1 10 2.45 . 1486
Day 3 27 3.78 .0220
Day*Treatment 3 27 0.65 .591 3

 

Appendix Table 11B. Proc mixed table for serum PYD concentrations (ng/mL) for

 

 

Arabian foals

' Source NDF DDF F -value P-Value
Treatment 1 10 0. 1 8 .6773
Day 3 30 23.61 .0001
Day *Treatment 3 30 1 .35 .2782

 

Appendix Table 12B. Proc mixed table for plasma Si concentrations (jig/L) for yearlings

 

 

Source NDF DDF F-value P-Value
Treatment 1 16 128.04 .0001
Day 3 48 32.16 .0001
Farm 1 16 4.19 .0575
Day*Treatment 3 48 28.25 .0001
Treatment*Farm 1 16 10. 16 .0057
Day*Farm 3 48 5.17 .0036
Day*Trt*Farm 3 48 3.48 .0229

 

Appendix Table 13B. Proc mixed table for normalized plasma Si concentrations (ng/mL)

for yearlings

 

Source NDF DDF F-value P-Value
Treatment 1 1 8 46.48 .0001
Day 2 36 l 8.23 .0001
Day*Treatment 2 36 2. 19 . 1263

 

78

Appendix Table 14B. Proc mixed table for serum OC concentrations (ng/mL) for

 

 

yearlings

Source NDF DDF F-value P-Value
Treatment 1 16 0.84 .3732
Day 3 42 2.54 .0692
Farm 1 16 1 .72 .2088
Day*Treatment 3 42 0.7 1 .55 1 8
Treatment*Farm 1 16 0.50 .4896
Day*Farm 3 42 1.97 .1336
Day*Trt*Farm 3 42 0.44 .7273

 

Appendix Table 15B. Proc mixed table for serum ICTP concentrations (ill/L) for

 

 

 

yearlings

Source NDF DDF F-value P-Value
Treatment 1 16 0.81 .3808
Day 3 47 1.09 .3637
Farm 1 16 l 1.70 .0035
Day *Treatment 3 47 1 .48 .23 12
Treatment*Farm 1 16 0.45 .5136
Day*Far'm 3 47 6.49 .0009
Day*Trt*Farm 3 47 1.28 .2936

 

Appendix Table 16B. Proc mixed table for serum PYD concentrations (ng/mL) for

 

 

yearlings

Source NDF DDF F-value P-Value
Treatment 1 16 0.02 .8878
Day 3 46 2.23 .0970
Farm 1 16 0.21 .6566
Day*Treatment 3 46 0.23 .8727
Treatment*Farm 1 16 0.01 .9072
Day*Farm 3 46 2.53 .0685
Day*Trt*Farm 3 46 0.66 .5818

 

79

Appendix Table 17B. Proc mixed table for foal wither height (cm)

 

 

 

 

 

 

 

 

 

 

Source NDF DDF _ F-value P-Value
Treatment 1 10 2.65 . 1 348
Day 1 10 142.05 .0001
Day*Treatment 1 10 0.09 .7683
Appendix Table 18B. Proc mixed table for foal hip height (cm)

Source NDF DDF F-value P-Value
Treatment 1 10 1.94 .1943
Day 1 10 159.06 .0001
Day*Treatment l 10 0.00 .9734
Appendix Table 19B. Proc mixed table for yearling wither height (cm)

Source NDF DDF F-value P-Value
Treatment 1 16 0.33 .5759
Day 1 16 4.20 .0572
Farm 1 16 38.34 .0001
Day*Treatment 1 l6 0. 1 1 .7496
Treatment*Farm 1 16 3.17 .0940
Day*Farm 1 16 0.03 .8687
Day*Trt*Farm 1 16 0.04 .8514
Appendix Table 20B. Proc mixed table for yearling hip height (cm)

Source NDF DDF F-value P-Value
Treatment 1 16 0.68 .4223
Day 1 16 5.10 .0382
Farm 1 16 62.27 .0001
Day*Treatment 1 16 0.02 .8960
Treatment*Farm 1 16 2.80 . 1 136
Day*Farm l 16 0.37 .5497
Day*Trt*Farm l 16 0.01 .9375

 

80

 

Appendix Table 21B. Proc mixed table for yearling weight (kg)

 

Source NDF DDF F-value P-Value
Treatment 1 16 0. 1 l .7474
Day 1 16 7.46 .0148
Farm 1 16 92.85 .0001
Day*Treatment 1 16 0.13 .7228
Treatrnent*Farm l 16 0.79 .3879
Day*Farm l 16 0.00 .9994
Day*Trt*Farm 1 16 0.07 .7962

 

81

 

‘3'.
«use».

APPENDIX C

82

APPENDIX C

Proc Corr Tables

Appendix Table 1C. Proc corr table for D 0 milk Si concentration and foal plasma Si
levels
Milk Si Concentration Plasma Si Concentration

Milk Si Concentration 1.0000 -0.2808
P-Value 0.0 .3766

Foal Plasma Si Concentration . -0.2808 1.0000
P-Value .3766 0.0

 

Appendix Table 2C. Proc corr table for D 15 milk Si concentration and foal plasma Si
levels
Milk Si Concentration Plasma Si Concentration

t.

 

 

Milk Si Concentration 1.0000 0.2150
P-Value 0.0 .5022

Foal Plasma Si Concentration 0.2150 1.0000
P-Value .5022 0.0

 

Appendix Table 3C. Proc corr table for D 30 milk Si concentration and foal plasma Si
levels

Milk Si Concentration Plasma Si Concentration

 

Milk Si Concentration 1.0000 0.3419
P-Value 0.0 .2766

Foal Plasma Si Concentration 0.3419 1.0000
P-Value .2766 0.0

 

Appendix Table 4C. Proc corr table for D 45 milk Si concentration and foal plasma Si
levels
Milk Si Concentration Plasma Si Concentration

 

Milk Si Concentration 1.0000 0.8342
P-Value 0.0 .0007

Foal Plasma Si Concentration 0.8342 1.0000
P-Value .0007 0.0

 

83

Appendix Table 5C. Proc corr table for D 0 milk Si concentration and mare plasma Si

levels
Milk Si Concentration

Milk Si Concentration 1.0000
P-Value 0.0
Mare Plasma Si Concentration 0.7313

P-Value .0069

Plasma Si Concentration
0.73 l 3

.0069

1.0000

0.0

 

Appendix Table 6C. Proc corr table for D 15 milk Si concentration and mare plasma Si

levels
Milk Si Concentration

Plasma Si Concentration

 

Milk Si Concentration 1.0000
P-Value 0.0
Mare Plasma Si Concentration 0.0637

P-Value .8441

0.0637
.8441
1.0000
0.0

 

 

Appendix Table 7C. Proc corr table for D 30 milk Si concentration and mare plasma Si

levels
Milk Si Concentration

Plasma Si Concentration

 

Milk Si Concentration 1.0000
P-Value 0.0
Mare Plasma Si Concentration 0.3370

P-Value .2840

0.3370
.2840
1.0000
0.0

 

Appendix Table 8C. Proc corr table for D 45 milk Si concentration and mare plasma Si

levels
Milk Si Concentration

Plasma Si Concentration

 

Milk Si Concentration 1.0000
P-Value 0.0
Mare Plasma Si Concentration 0.0204

P-Value .9499

0.0204
.9499

1 .0000
0.0

 

84

APPENDIX D

85

APPENDIX D

Hoof Measurement Data Tables: Experiment 2

Appendix Table 1D. Mean Toe Length (cm) for Yearlings

 

 

Day
0 SEM 45 SEM
Control (C) 4.87 .3 5.47 .3
Si Treated (S) 4.78 .3 5.33 .3 .....;

 

Appendix Table 2D. Mean Right Heel Length (cm) for Yearlings

 

 

 

Day ;

0 SEM 45 SEM J
Control (C) 2.95 .22 3.63 .22
Si Treated (S) 2.77 .22 3.5 .22

 

Appendix Table 3D. Mean Left Heel Length (cm) for Yearlings

 

 

DaL
0 SEM 45 SEM
Control (C) 2.9 .22 3.62 .23
Si Treated (S) 2.83 .23 3.62 .23

 

Parasite Data Tables: Experiment 2

Appendix Table 4D. Mean Parasite EgLNumber for Arabian Yearlings on d 45

 

Mean Parasite SEM
Egg Number
Control (C) 30.0 24.24
Si Treated (S) 0 0

 

Appendix Table SD. Mean Parasite Egg Number for Quarter Horse Yearlin gs

 

 

on d 45
Mean Parasite SEM
Egg Number
Control (C) 690 330.49
Si Treated (S) 635 294.17

 

86

. it. .. .474,

.L b. 3.1...

r—v-
I.-

APPENDIX E

87

APPENDIX E

Comparison of Blood Sample Collection Methods
and Their Effect on Plasma Si Concentrations

Introduction
Silicon (Si) analysis using electrothermal atomic absorption spectrometry was l
reported by Leung and Edmond (1997). Proper sample collection appears to be critical

for accurate results. Special care and handling of sampling materials is also critical.

 

Glass pipettes, syringes or vacuum tubes should be avoided when obtaining samples, as
glass will contaminate your sample. Plastic labware should be used instead (Gitelman
and Alderman, 1990). Silicon is a component of glass, as glass is made of sand. Sand
and dirt contain large amounts of Si.

Little is known about the extent that blood leaches Si from glass before the blood
is centrifuged and stored. A potential leaching effect may be aggravated by pH
differences. Additives in the collection tube, such as ethylenediamine tetracetic acid
(KZEDTA) or heparin may also chelate Si.

Because of the existing questions regarding Si collection and sampling, three
methods of blood collection were originally utilized in this study. However, only glass
plasma collection tubes were used in the current study to reduce stress to the foals during

sample collections, as both serum and plasma were collected, and vacuum tubes made for

88

ease of collection. Plasma Si was analyzed by Trace Elements Laboratory, LHSC
(University Campus, 339 Windermere Road, London, Ontario N6A 5A5, Canada).
Because of a limited amount of funds, the three methods were analyzed for only two
mares. Our objective was to determine differences in Si concentrations, as taken by
plastic collection tubes containing KzEDTA, glass tubes containing KZEDTA and glass
tubes containing heparin.

Methods

Minimizing silicon cantaminatian:

Two to three d prior to collection, all plastic equipment, including syringes,
plastic collection tubes and the transfer pipettes, was specially washed using a 10% nitric
acid wash. The equipment was then rinsed three times with double distilled water (dd
H20) and stored in a fume hood to allow for drying and to minimize Si contamination
from dust. The plasticware, when taken on the farm, was kept in a cooler until needed.
and replaced immediately after completion of the collection to avoid contamination.

In the laboratory, all samples were transferred to storage tubes within a plastic
wrap lined box. This box was only open to the front, and this opening had a sheet of
plastic wrap to serve as a curtain to prevent dust contamination. Plasma samples were
only transferred with the specially cleaned plastic pipettes within the box. The Si-free

storage microcentrifuge tubes were closed tightly, and then stored at -2°C.

89

 

KzEDTA Preparation for Plastic Collection Tubes:

Fifteen g of KZEDTA and 100 mL of dd H20 were mixed and stored in a plastic
acid washed bottle in the refrigerator. Then, KZEDTA (115 11L) was added to the
polyurethane tubes.

Sample Collection:

One 7-mL Vacutainer“ tube with KzEDTA additive and one 10-mL Vacutainer®
tube with heparin additive (Becton Dickinson and Company, Franklin Lakes, NJ) were
collected using the same 18 g x 3.8 cm disposable needles (Becton Dickinson and
Company, Franklin Lakes, NJ) from two adult mares. Additionally, 10 mL of blood from
each mare was withdrawn using 18 g disposable sterile needles into a syringe (Lucr-
Lokm, Becton Dickinson and Company, Franklin Lakes, NJ) that was specially cleaned
in a 10% nitric acid wash. The blood was then immediately transferred from the syringe
into a polyurethane Si-free collection tube (Falcon’, Becton Dickinson and Company,
Franklin Lakes, NJ) with 115 uL of KZEDTA in the bottom. The needle was punctured
through paraﬁlm covering the collection tube to prevent contamination, and the blood
was mixed gently. In the laboratory, the paraﬁlm was removed from the polyurethane
tubes, and the tubes were capped for centrifugation.

The blood samples were centrifuged for 20 min at 754 x g. After centrifugation,

plasma was withdrawn using specially cleaned plastic Samco brand transfer pipettes

 

du.

Si!

14

Sr

,
SIC

USll

Nit.

Dal;

Res

Piitil

Cont

(Corning Samco Corporation, San Fernando, CA), and transferred into plastic 1.5-mL
microcentrifuge tubes (DOT Scientiﬁc, Inc., Burton, MI) in a plastic lined box to avoid
dust contamination. Plasma was then stored in a —20°C freezer until analysis.
Silicon Analysis:
Plasma Si concentrations were analyzed by Trace Elements Laboratory, LHSC
(University Campus, 339 Windermere Road, London, Ontario N6A 5A5, Canada). ! I

Plasma Si concentrations were determined via electrothermal atomic absorption

 

spectrometry as adapted from Leung and Edmond (1997). J
Statistical Analysis:
Plasma Si concentrations for the three methods of blood collection were analyzed
using the Proc Anova procedure of SAS (6.12). The model included the type of method.
Mean Si concentrations for each type of collection were compared to each other at the or
= .05 level. Statistical signiﬁcance was reported at a P-value of less than or equal to .05.
Data were reported as mean Si concentration i standard deviation for each method.
Results
Samples collected with the glass tubes had higher Si concentrations than the
plastic collection tubes (Table 1B). Glass tubes with KZEDTA had higher Si

concentrations than tubes with heparin.

91

Appendix Table 1E. Mean Si concentrations (pg/L) for three blood collection methods
Mean Si Concentration Standard Deviation

 

Plastic Tubes with KZEDTA (n=2) 203.0‘ 31.1
Glass Tubes with KZEDTA (n=2) 1010.0” 66.5
Glass Tubes with Heparin (n=2) 691.5 ° 23.3

 

"’cMeans with different superscripts within columns are different (P 5 .05)
Discussion

Glass tubes yielded higher plasma Si concentrations than the plastic collection
tubes. This suggests that the blood does draw Si from the glass, and contaminates the
sample. Why the glass tubes containing KZEDTA yielded higher plasma Si levels than
the tubes containing heparin iS unclear. Further research is required to analyze the
differences seen between KZEDTA and heparin in this study. The results of this study
indicate that collection method does affect plasma Si concentrations.
Conclusions and Future Research

Blood collection using glass equipment does contaminate the sample and yields
inaccurate Si analyses. Comparisons of relative Si concentrations due to treatment
effects can still be determined using glass tubes, however the concentrations themselves
are inaccurately elevated. It is our suggestion, that plastic tubes be utilized for plasma Si
analysis.

There may be many factors contributing to the increased Si concentrations from
glass collection tubes. Leaching may be accentuated by pH shifts and time of sample

contact with the glass tubes. Further studies on these interactions Should be conducted to

92

 

optimize collection technique. Anticoagulants may affect Si analysis as well. For
example, KZEDTA is a known chelator of some minerals in blood. Further comparison
of KZEDTA versus heparin is required. Silicon research is continually being conducted

and collection methods should be solidiﬁed to allow for meaningful results.

93

 

APPENDIX F

94

 

ﬁlm-1r I

APPENDIX F

Osteocalcin and Carboxy-terminal Pyridinole Cross-link Telopeptide Region of
Type I Collagen Measurements in Arabian Horses for the First Year of Life

Introduction

Biochemical markers are becoming an increasingly popular tool in equine
research, and these markers provide valuable data on changes in various metabolic
activities. In the area of equine exercise physiology, systemic markers of bone
metabolism are continually used to indicate changes in turnover rates. Additionally,
these markers are useful in providing evidence of alterations in the turnover rates as a
result of various treatments. Young growing animals are rapidly turning over bone until
they reach skeletal maturity around age four (Parﬁtt, 1984). This turnover is designed to
optimize bone strength and integrity. However, this rapid growth may mask changes that
are actually occurring as a result of research-related treatment or growth-related diseases.
Therefore, having an understanding of the typical pattern of growth, as indicated by
systemic markers of bone turnover, will enable scientists, veterinarians and managers to
detect deviations from this general norm.

One such biochemical marker is osteocalcin (OC), which is an indicator of bone
formation. Osteocalcin is predominately secreted by the osteoblast, or bone-forming cell

(Lian and Gundberg, 1988; Kannus et al., 1996). It is a non-collagenous protein that is

95

released into circulation during new bone formation (Price, 1982). A second biochemical
marker, carboxy-terminal pyridinoline cross-linked telopeptide region of type I collagen
(ICTP), serves as an indicator of bone degradation. Type I collagen is the most
predominant form of collagen molecule in bone (Risteli et al., 1993; Lawrence and
Fowler, 1997), and ICTP is released into circulation when the collagen in the bone's
extracellular matrix is degraded (Hassager et al., 1994; Risteli et al., 1993). Together,
these two markers can be utilized to determine the rate of bone turnover in growing
animals. Therefore, the Objective of our study was to determine normal bone turnover in
foals through the ﬁrst two years of life by using OC and ICT P as systemic indicators of
bone metabolism. With this information, a growth curve will then be formulated.
Methods
Animal Management

Twelve Arabian foals were housed with their dam, for the ﬁrst 45 d of life, in
Stalls that were 3.1 m x 3.1 m or larger from 1600 until 0800 the next morning. During
the day, foals and mares were turned out in a paddock that averaged 2384.5 m2, with a
stocking rate Of three mares/foal units per paddock. Horses had ad libitum access to a
mixed grass paddock (brome, orchard, timothy and Kentucky blue grasses) in the
paddocks during the day. After 45 d, all mares and foals were turned out onto a mixed
grass pasture that was 432.8 m x 53.6 m. Four foals were removed from the study over

time due to unrelated death or other reasons.

96

 

Foals were introduced to creep feed at approximately three mon, and were
weaned between three and four mon. Routine health care, including vaccinations, farrier
work, and deworming, was maintained throughout the study. Creep feed was ad libitum,
in a group, however the foals did not always ﬁnish the ration. Once weaned, horses were
group fed, and each weanling ate approximately 1.70 kg oats, .51 kg corn and .33 kg
protein pellets (Kent Feeds, Muscatine, IA) at each meal. All horses were fed this ration
twice daily at 0700 and 1600. Hay was a grass and alfalfa mix, and each weanling ate

approximately 5.68 kg of hay during the winter mon, and, as yearlings, ate approximately

 

5.0 kg during spring mon. The creep feed, weanling ration and yearling ration were
approximately the same, and all horses were group fed. The daily nutrient values for
these diets are summarized in Table 1F.

Appendix Table 1F. Calculated daily nutrient table on an as fed basis

 

Nutrient Nutrient Content Calculated Total Daily Intake
DE 2.05 Mcal/kg 22.07 Mcal

CP 14.26% 1540 g

Ca .37% 40 g

P .32% 30 g

Mg .18% 20 g

Cu 14.23 mg/kg 178.71 mg

 

Blood Sample Collection
Blood samples were taken via jugular venipuncture from all foals on d 0,
following parturition, d 15, 30, 45, 60 and approximately every 30 d thereafter. Two 10-

, mL Vacutainer" tubes with no additive (Becton Dickinson and Company, Franklin Lakes.

97

NJ) were ﬁlled from all animals using 22 g x 2.5 cm disposable needles (Becton
Dickinson and Company, Franklin Lakes, NJ) to collect serum for biochemical marker
analysis. Blood samples were then centrifuged within two h of collection for 10 min at
754 x g. After centrifugation, serum samples were removed using disposable glass
Pasteur pipettes (VWR Scientiﬁc, West Chester, PA), aliquotted into plastic 1.5-mL
microcentrifuge tubes (DOT Scientiﬁc, Inc., Burton, MI) and frozen at -20°C until
analysis of OC and ICTP.
Osteocalcin Assay

Samples were analyzed for CC within 26 weeks of collection using an enzyme-
linked immunosorbent assay (ELISA) (NovoCalcinQ, Metra Biosystems, Inc., Mountain
View, CA) following manufacturer’s instructions. Serum was diluted at 1:15 for foals on
d 0 through d 120 and 1:10 for all days thereafter. Dilutions were determined by testing
samples at dilutions of 1:10, 1:15, and 1:20 for foals. These values were compared to the
standard curve provided by Metra Biosystems. When the samples fell along the linear
portion of the curve (from .4 to 1.5 optical density - OD), the corresponding dilution was
chosen to run the OC assay for all samples. Standards and controls were reconstituted
with .5 mL of IX wash buffer. Twenty-ﬁve pl. of standard, control, or sample was
added to the osteocalcin-coated strips. Samples were tested in duplicate. Anti-

osteocalcin (125 ILL) was also added to each well. A two-h incubation at room

98

temperature followed, and then the wells were washed with 300 [AL of IX wash buffer
three times with three min between each washing. Reconstituted enzyme conjugate (150
ILL) was added to the wells, and incubated for one h at room temperature. After
incubation, the wells were then washed in triplicate with 300 uL each time, and three min
between each wash. Finally, a working substrate was added (150 III.) to each well and
incubated at room temperature until the A standard read 1.5 OD, at which point 50 uL of
stop solution was added, and the plate was read at 405 nm optical density on a Spectra
Max 340 plate reader (Molecular Devices Corp., Sunnyvale, CA).
I C TP Assay

Samples were analyzed for ICTP using a radioimmunoassay (RIA) (ICT P, ”’1
RIA Kit“, Diasorin, Stillwater, MN) according to manufacturer’s directions. All samples,
standards, controls, nonspeciﬁc binding tubes, and total count tubes were run in
duplicate. The tubes contained the following: nonspeciﬁc binding tubes - 100 ILL sample
and 200 ILL double distilled water; standards (0 and A-F) - 100 11L standard and 200 uL
ICTP Antiserum; controls and samples — 100 11L serum and 200 11L ICTP antiserum.
Tracer (200 uL of I2"I ICT P) was added to all tubes, including the total count tubes.
Tubes were mixed and incubated for two h at 37°C. Next, 500 ILL of separation reagent
was added to all tubes, except the total count tubes. Tubes were vortexed and incubated

for 30 min at 20-25°C. After incubation, tubes were centrifuged for 30 min at 2000 x g at

99

4°C, and immediately decanted for the supernatant. Tubes were then read using the 1290
GammaTrac Gamma Counting System (Tm Analytic, Elk Grove Village, IL).
Statistical Analysis

Osteocalcin and ICTP were statistically analyzed as a two-factor ANOVA for
repeated measures using Proc Mixed procedure of SAS (6.12). The model included
sample day. Osteocalcin and ICTP concentrations for all days were reported as LSMeans
1: SEM. Statistical signiﬁcance was reported at a P-value of equal to or less than .05.
Trends were investigated at P-values equal to or less than .1.
Results

Results of the ﬁrst 150 d for most yearlings for CC and ICTP are reported in
Table 2F. Sample collections and assay analysis are in progress for the remaining
animals and days. Osteocalcin appeared to increase immediately after birth, and
remained higher through d 60. Osteocalcin values from d 15 through d 150 seemed

Appendix Table 2F. Mean OC and ICT P concentrations (i SEM) in horses from birth
through d 150

 

 

Day Mean OC (ng/mL) SEM Mean ICTP (ug/mL) SEM

0 65.376‘ 11.5 (n=10) 33.850‘ 1.7 (11:12)
15 117.378 b° 10.6 (n=12) 36.060a 1.7 (n=l2)
30 110.115be 11.0(n=11) 33.823“ 1.7 (11:12)
45 107.311 °° 11.0 (n=11) 33.226‘ 1.7 (n=1 1)
60 122.135 ‘ 11.0 (n=11) 24.105 b 1.9 (n=9)
90 95.534” 11.1 (n=11) 23.928 " 1.7 (n=11)
120 90.017abc 12.7 (n=8) 25.946” 2.1 (11:7)
150 102.017 ”‘6 15.8 (n=5) 23.160” 2.8 (n=4)

 

, ’°°°Means with different superscripts within columns differ (P 5 .05)

100

 

variable, as evident by Table 2F. A decrease in ICT P values was seen on d 60, and ICTP
remained lower through d 150.
Discussion
The increase in OC values seen after (1 0 is logical, as bone begins to be loaded
even on the ﬁrst day of life. Loading of bone causes an increase in bone turnover and
yields greater bone mineral density and strength (Woo et al., 1981; Raub et al., 1989;

Bell, 1999). The decrease in ICTP concentrations was not seen until (1 60. This decrease

 

in bone resorption, coupled with increased bone formation, results in net bone formation. ’
It is possible that it took 45 d for bone resorption to be signiﬁcant in newborn foals,

because until then a cell mediator was not acitvated in the osteoclasts until bone

formation was increased. Studies show that enzymatic activities of bone forming and

bone degrading cells alter each other's activity (Mohan et al., 1984). Chambers (1982)

found that as osteoclastic activity (bone degradation) increases, osteoblasts are activated

and begin to increase in osteoblastic activity (bone formation). Therefore, osteoblast and

osteoclast cells appear to be capable of initiating activity for each other.

It is expected that over the short term, OC and ICTP levels will ﬂuctuate. The
results of this study will provide an indication of variability for the bone growth markers.
Over the long term, the values for these markers are expected to decrease, as reduced
levels were seen in horses, dogs and humans as age increased (Allen et al., 1998; Black et

al., 1999; Kruse and Kracht, 1986; Lepage et al., 1990, 1992; Price et al., 1995). Once all

101

horses have been analyzed for OC and ICT P over 720 d more conclusive results will be

published.

Conclusion and Implications

The use of biochemical markers provides indices of systemic indicators for bone
metabolism. Young growing animals have high bone turnover to produce a stronger
skeletal system. Therefore, the need for a growth curve for CC and ICTP has arisen.
This curve can then be utilized to monitor normal skeletal growth and health, and can be
used in research studies to determine treatment differences. The reported results only
cover 150 d of the 720 d study. Once all data are collected, a growth curve will be
established for OC and ICTP. With this information, skeletal health can be monitored,

and research of young horses will be more meaningful.

102

 

 

LITERATURE CITED

103

LITERATURE CITED

Adler, A. J ., Z. Etzion, and G. M. Berlyne. 1986. Uptake, distribution and excretion of
3‘silicon in normal rats. Am. J. Physiol. 251:E670-E673.

Allen, M. J ., W. E. Hoffmann, D. C. Richardson, and G. J. Breur. 1998. Serum markers
of bone metabolism in dogs. Am. J. Vet Res. 59(3):250-254.

Anderson, R. R. 1993. Longitudinal changes of trace elements in human milk during the
ﬁrst 5 months of lactation. Nutr. Res. 13:499-510.

Ashcraft, A., and W. J. Tyznik. 1976. Effect of diet on volume and composition in
mare’s milk. J. Anim. Sci. 43(Suppl. 1):248 (Abstr.).

Baucus, K. L., S. L. Ralston, G. A. Rich, and E. L. Squires. 1989. The effect of copper
and zinc supplementation on mineral content of mares’ milk. Equine Vet. Sci.
9(4):206-209.

Benke, G. M., and T. W. Osborn. 1979. Urinary silicon excretion by rats following oral
administration of silicon compounds. Food Cosmetics Tox. 17:123-127.

Black, A., S. L. Ralston, S. A. Shapses, L. Suslak—Brown, and P. A. Schoknecht. 1997.
Skeletal growth patterns in Standardbred foals from birth to 1 year. Proc.15"’
ENPS, Fort Worth, TX, May 28—31. p. 326-327.

Black, A., P. A. Schoknecht, S'. L. Ralston, and S. A. Shapses. 1999. Diurnal variation
and age differences in the biochemical markers of bone turnover in horses. J.
Anim. Sci. 77:75-83.

Brady, M. C., P. R. M. Dobson, M. Thavarajah, and J. A. Kanis. 1991. Zeolite A
stimulates proliferation and protein synthesis in human osteoblast-like cells and
the osteosarcoma cell line MG 63. J. Bone Min. Res. 6:S139.

Burger, E. H., J. Klein-Nulend, and J. P. Veldhuijzen. 1994. Mechanical stress and bone
development. In: F. Lyall and A. J. El Haj (Eds) Society for Experimental
Biology Seminar Series 54: Biomechanics and Cells. pp 187-196. Cambridge
University Press, Cambridge.

104

 

Carlisle, E. M. 1970. Silicon: A possible factor in bone calciﬁcation. Science. 167:279-
280.

Carlisle, E. M. 1972. Silicon, an essential element for the chick. Science. 178:619-621.
Carlisle, E. M. 1974. Silicon as an essential element. Fed. Proc. 33(6):]758-1766.

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

Carlisle, E. M. 1978. Essentiality and function of silicon. In: G. Bendz, and I. Lingvist
(Eds) Biochemistry of Silicon and Related Problems. pp 231-253. Plenum
Press, London.

Carlisle, E. M. 1982. The nutritional essentiality of silicon. Nutr. Rev. 40(7): 193-198.

Carlisle, E. M., and W. F. Alpenfels. 1978 . A requirement for silicon for bone growth
in culture. Fed. Proc. 37:787.

Chambers, T. J. 1982. Osteoblasts release osteoclasts from calcitonin-induced
quiescence. J. Cell. Sci. 57: 247-260.

El Haj, A. J ., and G. P. Thomas. 1994. Cellular modelling of mechanical interactions
with the skeleton. In: F. Lyall and A. J. E] Haj (Eds) Society for Experimental
Biology Seminar Series 54: Biomechanics and Cells. pp 147-163. Cambridge
University Press, Cambridge.

Eyre, D. R., I. R. Dickson, and K. Van Ness. 1988. Collagen cross-linking in human
bone and articular cartilage. Biochem. J. 252:495-500.

Fraher, L. J. 1993. Biochemical markers of bone turnover. Clin. Biochem. 26:431-432.

Frey, K. S., G. D. Potter, T. W. Odom, D. M. Senor, V. D. Reagan, V. H. Weir, J.
Elslander, S. P. Webb, B. L. Morris, W. B. Smith, and K. E. Wiegand. 1992.
Plasma silicon and radiographic bone density in weanling Quarter Horses fed

sodium zeolite A. J. Equine Vet. Sci. 12(5):291-295.

Fujimoto, D., and T. Moriguchi. 1978. Pyridinoline, a non-reducible crosslink of
collagen. J. Biochem. 83:863-867.

105

 

Geor, R., E. Hope, L. Lauper, S. Piela, J. Klassen, V. King, and M. Murphy. 1995.
Effect of glucocorticoids on serum osteocalcin concentrations in horses. Am. J.
Vet. Res. 56(9): 1201-1205.

Gibbs, P. G., G. D. Potter, R. W. Blake, and W. C. McMullan. 1982. Milk production of
Quarter Horse mares during 150 days of lactation. J. Anim. Sci. 54(3):496-499.

Gitelman, H. J ., and F. R. Alderman. 1990. Determination of silicon in biological
samples using electrothermal atomic absorption spectrometry. J. Anal. Atomic
Spectr. 52687-689.

Glade, M. J. 1993. Effects of gestation, lactation, and maternal calcium intake on
mechanical strength of equine bone. J. Am. College Nutr. 12(4):372-377.

Gomez, B., C. A. Bally, D. K. Jenkins, R. J. Kelm, Jr., and S Seyedin. 1994. An enzyme
immunoassay for intact, newly synthesized osteocalcin: A marker of bone
formation. 1994 Int. Conf. Progress Bone Min. Res. October 14-16. Vienna,
Austria.

Gundberg, C. M. and S. K. Nishimoto. 1999. Vitamin K-dependant proteins of bone and
cartilage. In: (M. J. Seibel, S. P. Robins, and J. P. Bilezikian, Eds.) Dynamics of
Bone and Cartilage Metabolism. pp. 43-57. Academic Press, New York.

Hassager, C., L. T. Jensen, J. Podenphant, K. Thomsen, and C. Christiansen. 1994. The
carboxy-terrninal pyridinoline cross-linked telopeptide of type I collagen in serum
as a marker of bone resorption: The effect of nandrolone decanoae and hormone
replacement therapy. Calcif. Tissue Int. 54:30-33.

Hatori, M., K. J. Klatte, C. C. Teixeira, and I. M. Shapiro. 1995. End labeling studies of
fragmented DNA in the avian growth plate: evidence of apoptosis in terminally
differentiated chondrocytes. J. Bone Min. Res. 10:1960-1968.

Hays, V.W., and M. J. Swenson. 1984. Minerals. In J. M. Swenson (Ed.) Dukes’
Physiology of Domestic Animals (10"’ Ed). pp 449-485. Comstock Publishing
Associates, Ithaca, NY.

Hope, E., S. D. Johnston, R. L. Hegstad, R. J. Geor, and M. J. Murphy. 1993. Effects of

sample collection and handling on concentration of osteocalcin in equine serum.
Am. J. Vet. Res. 54(7):1017-1019.

106

Jacobson, M. D., M. Weil, and M. C. Raff. 1997. Programmed cell death in animal
development. Cell. 88:347-354.

Jacobson, N. L., and A. D. McGilliard. 1984. In: J. M. Swenson (Ed.) Dukes’ Physiology
of Domestic Animals (10* Ed). pp 863-880. Comstock Publishing Associates,
Ithaca, NY.

James, I. T., A. J. Walne, and D. Perrett. 1996. The measurement of pyridinium
crosslinks: a methodological overview. Ann. Clin. Biochem. 33:397-420.

Jee, W.S. 1988. The skeletal tissues. In: L. Weiss (Ed.) Cell and Tissue Biology: A
Textbook of Histology. pp 218-254. Urban & Schwarzenberg, Baltimore, MD.

Kaidi, R., P. J. Brown, J. S. E. David, D. J. Etherington, and S. P. Robins. 1991. Uterine
collagen during involution in cattle. Matrix. 11:101-107.

Kannus, P., L. Jozsa, M. Kvist, T. L. N. Jarvinen, V. Maunu, T. Hurme, and M. Jarvinen.
1996. Expression of osteocalcin in the patella of experimentally immobilized and
remobilized rats. J. Bone Min. Res. 11(1):79-87.

Keeting, P.E., M.J. Oursler, K.E. Wiegand. S.K. Bonde, T.C. Spelsberg and BL. Riggs.
1992. Zeolite A increases proliferation, differentiation, and transforming growth
factor-B production in normal human adult osteoblast-like cells in vitra. J. Bone
Min. Res. 7: 1281-1289.

Kimmel, D. B. 1993. A paradigm for Skeletal strength homeostasis. J. Bone Min. Res.
8(Suppl. 2):S515-S522.

Kruse, K., and U. Kracht. 1986. Evaluation of serum osteocalcin as an index of altered
bone metabolism. Eur. J. Pediatr. 145:27-33.

Lanyon, L. E. 1994. Mechanically sensitive cells in bone. In: F. Lyall and A. J. El Haj
(Eds) Society for Experimental Biology Seminar Series 54: Biomechanics and
Cells. pp 178-186. Cambridge University Press, Cambridge.

Last, J. A., L. G. Armstrong, and K. M. Reiser. 1990. Biosynthesis of collagen
crosslinks. Int. J. Biochem. 22(6):559-564.

107

 

Lawrence, L.A., E. A. On, G. J. Miller, P. W. Poulos, G. Piotrowski, and R. L. Asquith.
1994. The mechanical properties of equine third metacarpals as affected by age.
J. Anim. Sci. 72:2617-2623.

Lawrence, T. L. J ., and V. R. Fowler. 1997. Growth of Farm Animals. CAB
International, New York, NY.

Lepage, O. M., M. Marcoux, and A. Tremblay. 1990. Serum osteocalcin or bone Gla-
protein, a biochemical marker for bone metabolism in horses: Differences in
serum levels with age. Can. J. Vet. Res. 54:223-226.

Lepage, O. M., L. DesCoteaux, M. Marcoux, and A. Tremblay. 1991. Circadian
rhythms of osteocalcin in equine serum. Correlation with alkaline phosphatase,
calcium, phosphate and total protein levels. Can. J. Vet. Res. 55 :5-10.

Lepage, O. M., M. Marcoux, A. Tremblay, and G. Dumas. 1992. Sex does not inﬂuence
serum osteocalcin levels in Standardbred horses of different ages. Can. J. Vet.
Res. 56:379-381.

Leung, F. Y., and P. Edmond. 1997. Determination of silicon in serum and tissue by
electrothermal atomic absorption spectrometry. Clin. Biochem. 30:399-403.

Lian, J. B., and C. M. Gundberg. 1988. Osteocalcin: Biochemical considerations and
clinical applications. Clin. Orth. Rel. Res. 226:267-291.

Liesegang, A., M. -L. Sassi, J. Risteli, R. Eicher, M. Wanner, and J. -L. Riond. 1998.
Comparison of bone resorption markers during hypocalcemia in dairy cows. J.

Dairy Sci. 81:2614-2622.

Lind, M., B. Deleuran, K. Thestrup—Pedersen, K. Soballe, E.F. Eriksen and C. Biinger.
1995. Chemotaxis of human osteoblasts. APMIS. 103:140-146.

Marks, S. C., Jr., and S. N. Popoff. 1988. Bone cell biology: The regulation of
development, structure, and function in the skeleton. Am. J. Anat. 183:1-44.

Merkley, J. W., and E. R. Miller. 1983. The effect of sodium ﬂuoride and sodium
silicate on growth and bone strength of broilers. Poult. Sci. 62:798-804.

108

Miles, R. D., R. H. Harms, and S. M. Laurent. 1986. Inﬂuence of sodium zeolite A on
laying hen performance. Nutr. Rep. Int. 34(6):]097-1103.

Mohan, S., T. Linkhart, J. Farley, and D. Baylink. 1984. Bone-derived factors active on
bone cells. Calcif. Tissue Int. 36:S139-Sl45.

Mumpton, F. A., and P. H. Fishman. 1977. The application of natural zeolites in animal
science and aquaculture. J. Anim. Sci. 45:1188-1203.

Najda, J ., M. Goss, J. Gminski, L. Weglarz, K. Siemianowicz, and Z. Olszowy. 1994.
The antioxidant enzymes activity in the conditions of systemic hypersilicemia.
Biol. Trace Elem. Res. 42:63-70.

Nejweide, P. J ., E. H. Burger, and J. H. M. Feyen. 1986. Cells of bone: Proliferation,
differentiation, and hormonal regulation. Phys. Rev. 66(4):855-886.

 

 

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 E. H. Bird. 1993. Training distance to failure in
young racing quarter horses fed sodium zeolite A. J. Equine Vet. Sci.
13(10):562-567.

Nixon, A. J. 1991. Articular cartilage surface structure and function. Equine Vet. Ed.
3(2):72-75.

NRC. 1989. Nutrient Requirements of Horses (5" Ed.). National Academy Press,
Washington, DC.

Oftedal, O. T., H. F. Hintz, and H. F. Schryver. 1983. Lactation in the horse: milk
composition and intake by foals. J. Nutr. 113:2196-2206.

Ogawa, T., T. Ono, M. Tsuda, and Y. Kawanishi. 1982. A novel ﬂuor in insoluble
collagen: a crosslinking moiety in collagen molecule. Biochem. and Biophys.
Res. Com. 107(4): 1252-1257.

Parﬁtt, A. M. 1984. The cellular basis of bone remodeling: The quantum concept
reexamined in light Of recent advances in the cell biology of bone. Calcif. Tissue

Int. 36:837-S45.

Pennington, J. A. T. 1991. Silicon in foods and diets. Food Add. Contam. 8(1):97-118.

109

Price, J. S., B. Jackson, R. Eastell, A. E. Goodship, A. Blumsohn, 1. Wright, S. Stoneharn,
L. E. Lanyon, and R. G. G. Russell. 1995. Age related changes in biochemical
markers of bone metabolism in horses. Equine Vet. J. 27(3):201-207.

Price, P. 1982. Osteocalcin. In: W. A. Peck (Ed.) Bone and Mineral Research. Vol. 1.
pp 157-190.

Pritchard, J. J. 1972. General histology of bone. In: G. H. Boume (Ed.) The
Biochemistry and Physiology of Bone (2" Ed.). pp 1-20. Academic Press, New
York.

Rabon, H. W., Jr., D. A. Roland, Sr., and M. M. Bryant. 1995. Absorption of silicon and
aluminum by hens fed sodium zeolite A with various levels of dietary
cholecalciferol. Poult. Sci. 74:352-359.

Raub, R. H., S. G. Jackson, and J. P. Baker. 1989. The effect of exercise on bone growth
and development in weanling horses. J. Anim. Sci. 67:2508-2514.

Risteli, J., I. Elomaa, S. Niemi, A. Novamo, and L. Risteli. 1993. Radioimrnunoassay
for the pyridinoline cross-linked carboxy-tenninal telopeptide of type I collagen:
A new marker of bone collagen degradation. Clin. Chem. 39(4):635-640.

Robins, S. P., P. Stewart, C. Astbury, and H. A. Bird. 1986. Measurement of the cross
linking compound, pyridinoline, in urine as an index of collagen degradation in
joint disease. Ann. Rheum. Dis. 45:969-973.

Robins, S. P. 1999. Fibrillogenesis and maturation of collagens. In: (M. J. Seibel, S. P.
Robins, and J. P. Bilezikian, Eds.) Dynamic of Bone and Cartilage Metabolism.

pp. 31-42. Academic Press, New York.

Roland, D. A.,Sr., S. M. Laurent, and H. D. Orloff. 1985. Shell quality as inﬂuenced by
zeolite with high ion-exchange capability. Poult. Sci. 64:1177-1187.

Roland, D. A. 1988. Further studies of effects of sodium aluminosilicate on egg shell
quality. Poult. Sci. 67:577-584.

110

 

Rook, J. S., W. E. Braselton, R. F. Nachreiner, J. W. Lloyd, M. E. Shea, J. E. Shelle, and
P. R. Hitzler. 1997. Multi-element assay of mammary secretions and sera from
periparturient mares by inductively coupled argon plasma emission spectroscopy.
Am. J. Vet. Res. 58(4):376-378.

Schryver, H. F., O. T. Oftedal, J. Williams, L. V. Soderholm, and H. F. Hintz. 1986.
Lactation in the horse: The mineral composition of mare milk. J. Nutr.
116:2142—2147.

Schiitze, 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(1):39-
46.

Schwarz, K. 1978. Signiﬁcance and functions of silicon in warm-blooded animals.
Review and outlook. In: G. Bentz and I. Lindqvist (Eds) Biochemistry of
Silicon and Related Problems. pp 207-230. Plenum Press, New York.

Seabom, C. D., and F. H. Nielsen. 1994. Dietary silicon affects acid and alkaline
phosphatase and ‘5 calcium uptake in bone of rats. J. Trace Elem. Exp. Med.
7:1 1-18.

Tahtela, R., and E. Tholix. 1996. Serum concentrations of type I collagen
carboxyterrninal telopeptide (ICTP) and type I procollagen carboxy- and
aminoterrninal propeptides (PICP, PINP) as markers of metastatic bone disease in
breast cancer. Anticancer Res. 16:2289-2294.

SAS. 1997. SAS” System for Mixed Models. SAS Institute Inc., Cary, NC.
Tropeano, G., P. Scirpa, A. Tricerri, D. Paine, A. Barini, M. Magaro, and S. Dell'Acqua.
1992. Changes in maternal bone metabolism after delivery: A prospective study.

J. Matem. Fetal Invest. 2:241-243.

van der Rest, M. and R. Garrone. 1991. Collagen family of proteins. FASEB J. 5:2814-
2823.

Van Soest, P. J ., M. S. Allen, and M. I. McBumey. 1983. Silicon, chromium, the rare

earth elements and the remainder of the periodic table. Nat. Feed Ingred. Assoc.
Nutr. Inst. Program, April 4-7, Chicago, IL.

111

Vaughan, J. 1970. The Physiology of Bone. Cambridge Press, Oxford.

Vaughan, J. 1980. Bone growth and modelling. In: T. L. J. Lawrence (Ed.) Growth in
Animals. pp 83-99. Butterworths, Boston.

von der Mark, K. 1999. Structure, biosynthesis, and gene regulation of collagens in
cartilage and bone. In: (M. J. Seibel, S. P. Robins, and J. P. Bilezikian, Eds.)
Dynamics of Bone and Cartilage Metabolism. pp. 3-29. Academic Press, New
York.

Wall, L. H., 111, G. D. Potter, P. G. Gibbs, and G. W. Brumbaugh. 1998. Growth of
yearling ﬁllies fed alfalfa or soybean meal. J. Equine Vet. Sci. 18(4):266-269.

Wasserman, R. H. 1984. Bones. In: J. M. Swenson (Ed.) Dukes’ Physiology of
Domestic Animals (10‘h Ed). pp 467 -485. Comstock Publishing Associates,
Ithaca, NY.

W00, 8. L.-Y., S. C. Kuei, D. Amiel, M. A. Gomez, W. C. Hayes, F. C. White, and W. H.

Akeson. 1981. The effect of prolonged physical training on the properties of
long bone: A study of Wolff’s law. J. Bone Joint Surg. 63-A(5):780—787.

112