Div.
‘..31

. I O‘Pl.

‘ til}

1;.

 

x9!
Ital-.113!
tielllvxi

5. 0.3. o
9323...
"

'rHan

lllllllllllllllllllllllllllllllllllillllllllllllllllllllllllll

31293 00880 0629

This is to certify that the

dissertation entitled
Investigation of Autoantibodies to Thyroglobulin,

Thyroxine, Triiodothyronine and Thyroid Peroxidase

in Dogs
presented by

Eileen L. Thacker, DVM

has been accepted towards fulfillment
of the requirements for

PhD 0 degree in Pathology

 

 

“fitmwi

Major professor

Date ‘3 0C) 92

MSU is an Affirnmtivt‘ Action ”Equal Opportunity Institution 0-12771

F' “H
LIBRARY

Michigan State

University J

 

 

K

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.

H DATE DUE DATE DUE DATE DUE

93ng 00412001 [l

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

_J

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

MSU Is An Affirmative Action/Equal Opportunity Institution
oMmS—m

 

 

 

 

 

INVESTIGATION OF AUTOANTIBODIES
TO THYROGLOBULIN, THYROXINB, TRIIODOTHYRONINE
AND THYROID PEROXIDASE IN DOGS
BY
Eileen L. Thacker, DVM

A Dissertation
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

Doctor of Philosophy

Department of Pathology

1992

ABSTRACT
INVESTIGATION OF AUTOANTIBODIES
TO THYROGLOBULIN, THYROXINE, TRIIODOTHYRONINB
AND THYROID PEROXIDASE IN DOGS
BY
Eileen L. Thacker, DVM

Assays were developed to detect autoantibodies to
thyroglobulin (Tg), thyroxine (T4), triiodothyronine (T3), and
thyroid peroxidase (TPO) in dogs. An enzyme-linked
immunosorbent assay (ELISA) was used to detect autoantibodies
to Tg. Autoantibodies to T4 and T3 were detected using
radioisotope labeled T4 and T3 in an agarose gel
electrophoresis assay. An initial study of 119 canine sera
from dogs with clinical signs of hypothyroidism including
alopecia, lethargy and obesity found 58 (48.7%) had
autoantibodies to Tg, T4, or T3. The presence of Tg
autoantibodies was seen more frequently (P<0.05) in samples of
low serum T4 concentrations. In a second study of a group of
135 Golden Retrievers, with no clinical signs of
hypothyroidism, 16.3% of the sera tested had autoantibodies to
one of the 3 antigens. Forty-two Golden Retrievers related to
a dog with autoantibodies from the initial study population
were tested for the presence of autoantibodies to Tg, T4, or
T3. Of these dogs, 6 (14.3%) had autoantibodies.
Autoantibodies to Tg, T4, or T3 were present in significantly
higher frequency in Golden Retrievers 1 year of age or younger

(P<0.05) .

Canine TPO was isolated to determine whether
autoantibodies to TPO occur in canine hypothyroidism, as is
the case in human autoimmune thyroiditis. The enzyme was
solubilized from the thyroid using a deoxycholate-trypsin
digestion, ammonium sulfate precipitation, differential
centrifugation and DEAE sephadex chromatography. A monoclonal
antibody to canine Tg coupled to a immunoaffinity column was
used to eliminate the contaminating Tg from the TPO
preparation. SDS-polyacrylamide gel electrophoresis (SDS-
PAGE) under non-reducing conditions indicated protein bands at
107 and 47 kDa. SDS-PAGE, under reducing conditions
eliminated the 107 kDa band which was replaced with components
at 60, 45, and 39 kDa. A Western immunoblot assay was
developed using the isolated TPO. Fifty canine sera from dogs
with autoantibodies to T9, T4, or T3 were screened by Western
blot for evidence of autoantibodies to TPO. No antibodies
were detected in these sera suggesting a different
pathogenesis for canine autoimmune thyroid disease than that

hypothesized for human autoimmune thyroiditis.

Dedicated with love to
my husband, Brad, and

my children, Stacy and Paul.

ACKNOWLEDGEMENTS

I wish to express my gratitude to Dr. R.W. Bull, my major
professor, for his encouragement, support and patience during
my studies.

I am also grateful to my committee members, Dr. K. R.
Refsal, Dr. R. Dunstan, Dr. G. Padgett, and Dr. P. Coussens
for their advice and willingness to share their knowledge and
time.

I am also indebted to Peggy Coffman and J.D. for their
constant patience and their help. Without these 2 my research
would never have been completed.

Finally, my deepest appreciation goes to my family; my
husband, Brad, who tolerated, encouraged, and supported me
through all these years of studies and a special thanks to my
children, Stacy and Paul, for understanding and always saying

that I could do this.

ii

TABLE OF CONTENTS

Page

LIST OF TABLES ..................................... v
LIST OF FIGURES .................................... vi
INTRODUCTION.... ................................... 1
LITERATURE REVIEW. ............................. .... 7
Thyroid Anatomy ............................... 7
Thyroid Physiology ............................ 11
Thyroglobulin. ................................ 14
Thyroid Regulation ............................ 20
Thyroid Hormones...... ............... ......... 22
Mechanisms of Action .......................... 25
Thyroid Tests ................................. 26
Autoimmune Thyroiditis ........................ 31
References .................................... 45
EXPERIMENT 1: ...................................... 73

Prevalence of autoantibodies to thyroglobulin,
thyroxine, triiodothyronine and relationship to
autoantibodies and serum concentrations of
iodothyronines in dogs.

a) Summary .................................... 73
b) Materials and Methods ...... ' ............... . 73
c) Results .................................... 74
d) Discussion ................................. 76
e) References ................................. 77

iii

Page

EXPERIMENT 2:... ................................... 78
Golden Retriever study.
a) Summary... ................................. 78
b) Materials and Methods ...................... 79
c) Results .................................... 80
d) Discussion ................................. 81
e) References ................................. 84

EXPERIMENT 3 ....................................... 85

Isolation of thyroid peroxidase and investi-
gation of thyroid peroxidase in the dog.

a) Summary .................................... 86
b) Introduction ............................... 87
c) Materials and Methods ...................... 88
d) Results .................................... 94
e) Discussion ................................. 96
f) References ................................. 102
EXPERIMENT 4: ...................................... 106

Studies of autoantibodies to canine TPO by
enzyme-linked immunosorbent assay and inhibition

of
a)
b)
C)
d)
e)

SUMMARY

enzyme activity.

Summary .................................... 106
Materials and Methods ...................... 107
Results .................................... 110
Discussion ................................. 110
References ................................. 112
AND CONCLUSIONS ............................ 114

iv

Table

LIST OF TABLES

Distribution of autoantibodies
antigens in serum samples from

Distribution of autoantibodies
antigens in serum samples from

Retrievers ................ .....

Distribution of autoantibodies
antigens in serum samples from

Retrievers ................ .....

Activity of canine TPO at various stages of

purification ...................

Page
to thyroid
119 dogs....... 75
to thyroid
135 Golden
............... 83
to thyroid
42 Golden
..... .......... 83
............... lOO

LIST OF FIGURES

Figure Page

1 The effect of serum dilution on absorbance

in the TgELISA ................................. 74
2 Effect of increasing concentrations of Tg

in the TG ELISA ................................ 75
3 Autoradiographs of agarose gels for'T; and T3

autoantibody determination ..................... 75
4 Distribution of autoantibodies to T4 and T3 in

relationship to serum T4 and T3

concentrations ................................. 76
5 SDS-polyacrylamide gel electrophoresis of

canine TPO ..................................... 101

vi

1
INTRODUCTION

Hypothyroidism in dogs is a frequently diagnosed (or
misdiagnosed) disorder. Because the clinical symptoms;
alopecia, lethargy, obesity and reproductive failure are non-
specific, hypothyroidism is difficult to differentiate from
other dermatologic, neurologic or reproductive disorders.

Currently, the diagnosis of canine hypothyroidism is
based on the measurement of thyroid hormone concentrations in
the serum. However, thyroid hormone concentrations are
frequently not affected until the later stages of thyroid
disease and as such, are of limited value in diagnosing early
stages of the disease“ Compounding the diagnostic problem are
the many factors that affect thyroid hormone concentrations.
Furthermore, many drugs and metabolic changes lower thyroid
serum concentrations while the dog is physiologically
euthyroid. In other words, the dog is not hypothyroid even
though the serum thyroid hormone concentrations are low.

Primary diseases of the thyroid are the most common cause
of hypothyroidism in dogs. Idiopathic thyroid atrophy and
lymphocytic thyroiditis account for the majority of canine
primary autoimmune thyroid diseases.‘ Idiopathic thyroid
atrophy is characterized by loss of thyroid parenchyma with
adipose and connective tissue replacement. The cause of this
disorder is unknown and there has been speculation that this

is actually an end-stage result of

lymphocytic 'thyroiditis. Lymphocytic thyroiditis is
characterized by the infiltration of the thyroid gland by
mononuclear cells, lymphocytes being the most predominate type
which results in the destruction of the thyroid follicles and
formation of lymphoid follicles. Fibrosis of the thyroid
parenchyma is seen in advanced cases with.minimal inflammatory'
cell infiltrate. inn a colony of laboratory beagles, focal
lymphocytic thyroiditis was observed with no significant
changes in thyroid function or accompanying clinical signs.2
However in pet dogs, lesions tend to be more diffuse and
extensive with clinical signs of hypothyroidism occurring as
the thyroid gland is progressively destroyed.3

While the initiating cause of autoimmune lymphocytic
thyroiditis is unknown, both the cellular and humoral immune
systems play a role in the destruction of the thyroid gland.
The formation of autoantibodies to various thyroid antigens is
indicative of an alteration in the immune system.
Autoantibodies to thyroglobulin (Tg), thyroxine (T4),
triiodothyronine (13), and the microsomal antigen which has
been identified as thyroid peroxidase (TPO) in humans have
been identified in dogs and human beings with thyroid
disease.“‘6 In studies of humans with Hashimoto's thyroiditis,
the presence of immune complexes composed of Tg and anti-Tg
antibodies have been found on the thyroid follicular basement
membrane. The formation of these complexes may lead to

thyroid damage by fixation of complement or by activation

3

of killer cell activity.6 Increased activity of cytotoxic T
cells and/or the cytotoxic effects of lymphokines released by
activated T«cells responding to thyroid antigens are potential
cell mediated immune mechanisms that damage the thyroid.7

The detection of autoantibodies to the various thyroid
antigens is an important tool in the diagnosis of thyroid
disease in human beings. Greater than 90% of patients with
Hashimoto's thyroiditis have autoantibodies to TPO.
Autoantibodies to thyroid antigens have also been studied in
the dog.“‘ The incidence of these autoantibodies have varied
with the type of assay used for their detection and the
population of dogs tested” 'The incidence of autoantibodies to
T9 in various populations of dogs has ranged from 27 to 73
percent. Autoantibodies to T4 and T3 were found in 0.2 percent
of the canine sera samples submitted to Auburn University for
the assay of thyroid hormone concentrations.8 In addition,
microsomal autoantibodies were detected in 4 to 86 percent of
various studies of hypothyroid dogs. Because of the wide
variation in ranges of autoantibodies detected many questions
remain as to the role and importance of these autoantibodies
in dogs with lymphocytic thyroiditis and clinical
hypothyroidism.

Previous studies in dogs employed complement fixation or
indirect immunofluorescence assays to identify microsomal
autoantibodiesfl‘5 Because in human beings, the microsomal

autoantibodies had been demonstrated to fix complement,

4
complement fixation using a crude thyroid extract was
performed. Indirect immunofluorescence was also used in the
dog to investigate thyroid autoantibodies. It has since been
documented in human beings that.the microsomal antigen is‘TPO.

This study looked at the various thyroid antigens and
autoantibodies in the dog and their relationship to each other
and to the clinical signs of hypothyroidism. The development
of assays for these various autoantibodies may help diagnose
autoimmune lymphocytic thyroiditis in the dog. Looking at
autoantibodies.to the various thyroid antigens may improve our
ability to determine the presence or absence of thyroid
disease with improved accuracy. In addition to improved
diagnostic assays, determination of the thyroid antigens that
induce thyroid autoantibody formation may be beneficial to
investigating the pathogenesis of thyroid disease and answer
the question of what is responsible for the body no longer
recognizing the thyroid as a part of itself.

The specific objectives of this study were to develop
assays to detect autoantibodies to canine Tg, T“ 15 and the
microsomal antigen using purified canine TPO. After these
assays had been developed, the relationship between thyroid
autoantibodies and the clinical signs of hypothyroidism in
dogs and their relationship to serum thyroid hormone

concentrations were investigated.

5
REFERENCES
Gosselin SJ, Capen CC, Martin SL. Histologic and
ultrastructural evaluation of thyroid lesions associated

with hypothyroidism in dogs. Vet Pathol 18:299—309, 1981.

Fritz TE, Norris WP, Kretz WB. Influence of lymphocytic
thyroiditis on iodine metabolism in the beagle. Proc Soc

Biol Med 134:450-458, 1970.

Gosselin SJ, Capen CC, Martin SL, et al. Autoimmune
lymphocytic thyroiditis in dogs. Vet Immunol Immunopathol

3:185-201, 1982.

Mizejewski GJ, Baron J, Poissant G. Immunologic
investigations of naturally occurring canine thyroiditis.

J Immunol 107:1152-1160, 1971.

Haines DM, Lording PM, Penhale WJ. The detection of
canine autoantibodies to thyroid antigens by enzyme-
linked immunosorbent assay, hemagglutination and indirect

immunofluorescence. Can J Comp Med 48:262-267, 1984.

Pinchera A, Fenzi G, Chiovato L, et al. Autoimmune
hypothyroidism. In: Drexhage HA. (ed) The Thyroid and
Autoimmunity. Amsterdam, Elseiver Sci Pub. BV (Biomed

Div), 1986, pp. 73-80.

6
Schleusener H, Bogner U. Cytotoxic mechanisms in
autoimmune thyroiditis. In: Walfish PG, Wall JR, Volpe R.
(eds) Autoimmunity and the Thyroid. Toronto, Academic

Press, 1985, pp. 95-107.

Young DW, Sarten JL, Kemppainen RJ. Abnormal canine
triiodothyronine-binding factor characterized as a
possible triiodothyronine autoantibody. Am J Vet Res

46:1346-1350, 1985.

7

LITERATURE REVIEW

Thyroid Anatomy

The thyroid gland was named by the Greeks because of the
resemblance the human thyroid gland has to a shield. In the
dog, the thyroid gland is not shield-like, but rather is 2
separate lobes lying dorsal-lateral to the first five to eight
tracheal rings. An isthmus, connecting the two lobes is
occasionally found on the ventral surface of the trachea. In
most dogs the cranial pole of the right lobe lies opposite the
first tracheal ring or the caudal border of the cricoid
cartilage of the larynx. The left lobe usually lies one to
three tracheal rings caudal to the right lobe.

The thyroid glands lie in fascia which is attached to the
overlying musculature and the tracheal The glands in dogs are
not as tightly attached to the trachea as they are in humans.
The size of the lobes are variable, the mass of one of the
lobes occasionally being as much as 50% larger than the other.
Generally, younger subjects have larger thyroid glands which
decrease in size with age. The glands have been reported to
be larger in females than in males, and may have a transitory
enlargement during estrus and pregnancy.I A thyroid lobe in
a 10 to 15 kg dog is approximately 2 X 1.5 X 0.5 cm.2 The

normal thyroid gland ranges from 40 to 400 mg/kg body

8
weight. The absolute weight of the thyroid glands may be
increased in areas of iodine deficiency, where weights of 80
to 1600 mg/kg have been reported.3

The dorsal edge of the thyroid glands lie in close
proximity to the carotid sheath carrying the carotid artery,
vagosympathetic nerve trunk, internal jugular vein and the
recurrent laryngeal nerve. The lobes lie in the fascial
compartment composed laterally of the deep lamina of the
middle cervical fascia of the sternocephalicus muscles, and
medially of the prevertebral fascia which communicates with
the thoracic cavity.

There is a richer blood supply to the thyroid glands then
almost any other organ in the body. The cranial thyroid
artery arising from the common carotid artery is the principal
blood supply to the thyroid glands. Occasionally a caudal
thyroid artery originating from the brachycephalic artery is
also present. Beneath the fibrous capsule of the thyroid
gland, the blood vessels form a vascular plexus. This plexus
gives rise to a capillary network which surrounds each
follicle, the capillary endothelium being contiguous with the
follicular epithelium. The principal venous return from the
thyroid gland is the caudal thyroid vein which leaves the
caudal end of the lobe and joins the internal jugular vein.

The thyroid glands are innervated by fibers from the
sympathetic component of the cranial laryngeal nerve. There

are also parasympathetic nerve fibers from the vagus. Hormone

9

secretion is indirectly influenced by blood flow through the
gland, which is controlled by nerve plexuses in the walls of
the arteries. Stimulation of the thyroid nerve slows the flow
of blood through the gland, whereas stimulation of the
homolateral vagus accelerates the blood flow. However, the
most important factor in the control of thyroid hormone
secretion is the balance maintained between the thyroid
hormones and the thyrotrophic hormones of the anterior
hypophysis.

Microscopically, the thyroid gland has two capsules. ‘The
outer capsule is continuous with the cervical fascia. The
inner capsule is composed of fibroelastic tissue from which
septa extend into the thyroid parenchyma. The septa provide
the internal support for innervation, blood supply and
lymphatics and separate the thyroid into lobules. .Because the
septa are not connected, the thyroid gland is not actually
lobulated, but pseudolobulated.

The thyroid gland consist of follicles which are the
structural and functional unit of the thyroid. A fbllicle
consists of a single layer of epithelial cells enclosing a
cavity, which is filled with colloid. The follicles are
surrounded by a thin basement lamina and a network of
capillaries and nerves. The shape of the follicular cells
varies from follicle to follicle and with the activity of the
thyroid gland. In general, cuboidal cells are characteristic

of hypoactive follicles and columnar cells of hyperactive

10
follicles. Follicle diameter ranges from 30 to 160 microns,
with epithelial height ranging from 8 to 20 microns.4

Thyroid follicular cells, when examined with an electron
microscope, are similar to other secretory cells, but have
several unique features. Numerous microvilli extend into the
colloid from the apical surface of the follicular cell. It is
at or near this surface of the cell that iodination,
exocytosis, and the initial phase of hormone secretion
occurs.6 The nucleus of the follicular cell is located in the
basal portion of the cell and has with no‘ distinctive
features, but is poor in heterochromatin, and has 1 or more
nucleoli. The cytoplasm is laden with extensive endoplasmic
reticulum with associated polysomes containing the precursor
of thyroglobulin (T9). The Golgi apparatus is located
apically in the cell and resembles other secretory cells. It
is in the Golgi complex that the peripheral monosaccharides
are incorporated into the carbohydrate units of the Tg
molecule.

In addition to the follicle cells, the thyroid glands
contain parafollicular or C cells that produce the hormone
calcitonin. These cells are found either in the follicle
wall or in the interfollicular spaces.’

Usually there are 2 parathyroid glands associated with
each thyroid lobe. The external parathyroid lies in the
fascia at the cranial pole of the thyroid. Internally, the

parathyroid lies beneath the thyroid capsule on the medial

ll
aspect of the lobe; in some cases it may be embedded in the
thyroid parenchyma. The internal parathyroids cannot be
surgically separated from the
thyroid glands. The location of the parathyroids is variable
in each individual dog and may vary between the right and left

lobes.

Thyroid Physiology
The physiological processes of the thyroid gland are:

synthesis in the follicular cells of the specific thyroid
glycoprotein, Tg and it's secretion into the follicular lumen;
iodination of Tg, including the formation of hormones; and the
storage of Tg in the follicular lumen, reabsorption of Tg into
the follicular cell, degradation of Tg and release of hormones
into the adjacent vasculature.

The follicle is the smallest functional unit of the
thyroid gland. Ihi the active follicular cells, the rough
surfaced endoplasmic reticulum (RER) is a prominent component.
Most of the protein synthesis occurring in the RER is directed
toward the production of Tg. The synthesis of the peptide
chain of Tg occurs in the polysomes attached to the RER
membranes as the Tg is transported across the RER membrane.7~8

The Tg is transported via transitory vesicles to the
Golgi apparatus and involves energy dependent membrane
fissions and fusions.9 It.is in the Golgi apparatus that the

majority of peripheral monosaccharides are incorporated into

12

the carbohydrate units of the Tg molecules.lo From the Golgi
apparatus, Tg is transported to the apical surface in
vesicles. The vesicles fuse with the apical surface by
exocytosis and. empty their contents into the follicular
lumen." Upon stimulation with thyroid stimulating hormone
(TSH), apical microvilli extend into the colloid in the
follicle lumen and fuse, leading to endocytotic uptake of Tg
and formation of colloid droplets in the cytoplasm. These
colloid droplets fuse with lysosomes to form phagolysomes in
which acid proteases and peptidases degrade the Tg into
iodothyronines including thyroxine (Tu) and triiodothyronine
(13) the active thyroid hormones. The thyroid hormones, T4
and T3, are released from the phagolysosomes at the base of
the cell into blood vessels while the remaining products of Tg
breakdown wo form new thyroid hormone molecules.

Iodide transport occurs with the pumping of iodide from
the plasma into the thyroid, where it is concentrated,
oxidized and used in thyroid hormone synthesis. Iodide (I)
is immediately oxidized within the thyroid to the organic form
(I) by H55 and 0,. The transport of iodide is temperature
dependent and requires an intact cellular membrane but not an
intact follicule as isolated thyroid cells can concentrate
iodide.‘2 The iodide concentrating mechanism is located in the
basal membrane, requires external K+ and appears to be a Na+
transport mechanism involving’Na*—KE.ATPase:as demonstrated by

3

inhibition by ouabain.I Iodide transport is an active process

13

as the iodide is transported against an electrochemical
concentration gradient. Thyroid stimulating hormone (TSH)
plays a major role in iodide accumulation. With an increase
in TSH, there is an increase in the influx of iodide modulated
by cAMP.14

The transport of iodide across the plasma membrane
involves phospholipid fractions which complex the iodide and
make it more soluble» Thyroid.phospholipid binding is similar
to an enzyme reaction in regard to saturation kinetics,
structural specificity, and the competition by other anions.
Other proteins, especially albumin have been suggested as
carriers for iodide to the thyroid. In addition to iodide
ions, the thyroid can concentrate other halides and complex
anions which exhibit competitive inhibition of iodide uptake.‘2

The activity of the iodide transport system in the
thyroid glands varies inversely with the circulating levels of
iodide. As a consequence of this autoregulation, plasma
iodine concentrations influence thyroid hormone synthesis by
altering the iodide transport in the thyroid.‘5 However,
normal individuals exposed to chronic iodide excess escape the
autoregulation as the iodide transport system adapts to limit
intracellular iodide. As a result, the intra-thyroidal iodide
concentration falls below the level necessary to inhibit
iodotyrosine formation.l6 Since cAMP production is inhibited
by iodides, the inhibition of adenylate cyclase has been

suggested as the escape mechanism.l7

l4
Thyroglobulin tTgl

Thyroglobulin is the iodinated protein which is the
biosynthetic precursor of the thyroid hormones and has an
essential role in thyroid hormone homeostasis. Thyroglobulin
allows the utilization of iodine to produce a high yield of
thyroid hormones“ The presence of high and low affinity sites
for iodination allows Tg to modulate hormone synthesis in
relation to iodine availability. Thyroglobulin allows the
storage of thyroid hormones in a non-active form in
extracellular space away from metabolic influence.

Structurally Tg is a dimeric glycoprotein with a
molecular weight of 660,000 daltons and a sedimentation
coefficient of 19S. Examination with an electron microscope
reveals 2 symmetric, identical subunits of 330,000 daltons.l8
The amino acid composition of Tg is not unusual and doesn't
contain an unusual amount of tyrosine, from which the thyroid
hormones, T4 and T3 are composed. N-glycosylation of the
peptide chain results in two types of oligosaccharide units
representing eight to ten percent by weight of the molecule
depending on the species.”20 After synthesis and glycosylation
of the peptide chains, the protein is iodinated and degraded
to produce the thyroid hormones.

Thyroglobulin has been found to have specific structural
domains involved in thyroid hormone formation. In human

beings, a single Tg molecule contains 110 tyrosine residues,

15

two-thirds of which are on the surface of the molecule, and in
vivo a maximum of
twenty are iodinated giving rise to two to five hormonally
active iodothyronines.2l Porcine Tg also contains several T4
forming sites which have different primary structure and
different hormone synthesizing capacity than human beings.22
The same hormonogenic structure can synthesize T4 or T3
depending on the amount of iodination of the thyroglobulin.

Non-iodinated, newly formed Tg dissociates freely into
123 subunits by dilution, alkaline pH or dissociating agents.
Prior to iodination, there are ten free sulfhydryl groups on
thyroglobulin.23 As the Tg molecule becomes more iodinated
the sulfhydryl groups disappear and it becomes more resistant
totdissociation into the 128 subunits. ‘This stability appears
to be due to the increasing number of disulfide bridges in the
Tg molecule as a result of iodination.24

Most iodination of Tg occurs post translation at the
microvilli-rich, cell-colloid interface very close to the
apical membrane by the enzyme thyroid peroxidase» The Tg then
diffuses into the lunenal colloid where some iodination can
occur.” Colloid is composed almost entirely of Tg with
greater than 90% consisting of the 19S Tg. Thyroglobulin
breakdown to the thyroid hormones occurs almost exclusively in
the follicle cell. The colloid material in the follicle lumen
is engulfed by endocytosis forming membrane bound colloid

droplets. There are few colloid droplets in normal, resting

l6

thyroid tissue, but their number rapidly increases following
the administration of TSH. During endocytosis, a very large
surface. of the apical plasma membrane is internalized.26
Exocytotic vesicles containing newly formed Tg may be a source
of the membrane engulfed by endocytosis; however the
endocytotic and exocytotic membranes are not the same as the
endocytotic vesicles are devoid of thyroid peroxidase (TPO)
implying that a prior reorganization of the apical membrane is
necessary.“ The mechanism by which the membrane changes is
unknown. Enzymes in the lysosomes that hydrolyze Tg include;
acid phosphatase, glycoside hydrolases, cathepsin D, several
thiol endopeptidases, and several exopeptidases.2l

Efficient hydrolysis of Tg resulting in release of the
thyroid hormones requires that the phagolysomes maintain an
acidic pH, which may be driven by an adenosine triphosphate
(ATP) pump.27 Among the iodinated compounds leaving the
thyroid gland, T“.fm, and iodide are the main constituents.
Other minor components include the inactive reverse T3 nag),
trace amounts of monoiodotyrosine (MIT), diiodotyrosine (DIT) ,
diiodothyronines, and Tg. Following the hydrolysis of Tg, the
majority of the MIT and DIT, which contain 80% of the Tg
iodine, are.deiodinated within the follicle cell.28 The iodide
released from the MIT and DIT is reused for hormone synthesis,
providing the follicle cell with two to three times the amount

of iodide transported from the blood stream.29

17

The mechanism of release of the thyroid hormones into the
blood stream following Tg hydrolysis is presently unknown.
Under physiological conditions about 90 per cent of the
thyroid hormones secreted are T4, of which 75 per cent are
deiodinated in peripheral tissues to form the active T3, the
remaining T4 being degraded and excreted.30

In some pathological states, the amount of T3 secretion
can increase. Iodine deficiency, for example results in
decreased deiodination and increased production of T3. The
high T3 content in Tg is accompanied by a high relative
content of T3 in thyroid secretion resulting in a eumetabolic
state being maintained.31 Normally, the T4/T3 ratio in
thyroidal secretion is related to the T4/T3 ratio in the Tg
hydrolysate, but this latter is not always an exact
representation of the iodine content of the gland.32 The
preferred secretion of T3 over T4 under these conditions may
also be due in part to increased deiodination of T4 within the
follicular cell and a faster secretion of T3 during Tg
hydrolysis.33

Iodination of Tg is catalyzed by the enzyme thyroid
peroxidase (TPO), a membrane-bound, glycosylated hemoprotein
with a molecular weight of 100,000 daltons.34 The amino acid
sequence of human TPO has been determined.” Thyroid
peroxidase has several important functions in the thyroid
which include: 1) the oxidation of iodide to a higher

oxidation state with H202 acting as the electron acceptor;

18
2)binding of the oxidized iodine to tyrosine residues of the
Tg molecule; and 3) the coupling of the resulting
iodotyrosines.

Because of its particulate nature, TPO has been difficult
to purify, and preparations with high specific activity have
been isolated only after first releasing the exzyme from the
membrane by proteolysis.“5 Although it is likely that the
native enzyme is altered by these procedures, the fragment
that is isolated is nevertheless extremely active and carries
the active site.

As previously discussed, iodide, the form in which iodine
enters the thyroid gland, must first be oxidized to a higher
oxidation state before it can act as an effective iodinating
agent. The exact mechanism of TPO-catalyzed iodination has
not been determined at this time, however various theories
have been discussed. One theory suggests that that both the I‘
and the tyrosyl residue undergo monoelectron oxidation
resulting in the formation of free radicals.37 Another theory
suggests a two-electron change producing an intermediate, the
iodinium ion, 1*.38 A third theory proposes that TPO degrades
Hg» catalytically in the presence of low concentrations of 1'
resulting in the formation of hypoiodite (HIO3) as the
iodinating intermediate.” Most investigators suggest that
iodination occurs on the enzyme implying that there is a
specific site for the acceptor (tyrosine or tyrosyl) Evidence

suggests that oxidized TPO has two substrate sites, one

l9

favoring 1? and the other favoring tyrosine. Iodination of
tyrosine appears to occur after the tyrosine has been
incorporated into thyroglobuin.40 The iodination of Tg
probably occurs on fully aggregated Tg rather than on its
subunits.41

Without a source of Hfih, TPO displays no activity.39 It
appears that Hg» plays an essential role in thyroid hormone
formation and it is possible that under some conditions, it
could be a rate-limiting step. The pathway for the H202
generation in the thyroid involves reduced pyridine
nucleotides coupled to a flavoprotein cytochrome reductase.
This is based on the fact that iodination in the thyroid is
stimulated by the addition of NAD(P)H and flavin nucleotides.42

Formation of T4 can occur nonenzymatically by iodination
of Tg and other proteins with molecular iodine.”’ The
catalytic role for TPO in the coupling reaction has been
demonstrated by observing that the amount of labeled'n,and T3
increased markedly when the labeled Tg was inclubated with
TPO. This increase could not be attributed to a change in the
level of iodination in Tg, because the coupling system in
contrast to the iodination system contained no added iodide.
Moreover, the increase in labeled'n,and T3 was accompanied by
a corresponding decrease in DIT, indicating conversion of DIT
to T4 and T3.“4

Analysis of the Tg gene has been performed in humans,

cattle, goats and rats and indicates a split gene of unique

20
organization composed of approximately 300 kilobases.”47 In
all species studied there is a predominance of small exons
150-200 base pairs in length homogeneous throughout the gene
suggesting this is a pattern under strong selective pressure.
A giant 64 kilobase intron is present in the middle of the
genes.” Hormonogenic sites have been assigned in four exons;
1 at the 5' end of the human gene and 3 at the 3' end of the

rat and human gene, although they are in different exons.

Thyroid Regulation

The primary regulatory mechanisms of the thyroid gland
are the hypothalamic-pituitary-thyroid control system and an
intrathyroidal autoregulatory system. Thyroid stimulating
hormone, released from thyrotrophs in the pituitary,
stimulates many aspects of thyroid activity including thyroid
hormone synthesis and secretion. The secretion of TSH is
regulated by two elements: neural control by the hypothalamus
and feedback control by thyroid hormones. Three different
neurohormones mediate neural control, a stimulating
neuropeptide (thyrotropin-releasing hormone [TRH]) and two

8 These

inhibitory factors, somatostatin and dopamine.4
regulating factors are synthesized by neurons in the
hypothalamus and reach the anterior pituitary through
specialized blood vessels. In addition to regulation by

thyroid hormone concentrations, TSH secretion is modified by

estrogens, which sensitize the pituitary to TRH, and

21

glucocorticoids, which reduce pituitary responsiveness to TRH
and possibly inhibit TRH secretionf“l

Thyroid hormones act on the pituitary to inhibit TSH
secretion.”‘However, uncertainty still exists as to*whether or
how the brain participates in the feed-back regulation of TSH
secretion by modulation of TRH and somatostatin secretion.48
In tissue cultures of either normal pituitary cells or mouse
thyrotropic tumors, both T4 and T3 cause a decrease in the
spontaneous release of TSH or that induced by TRH.53
Suppression of TSH secretion by thyroid hormones takes several
hours and does not occur if protein synthesis is blocked.52’53
Initially, thyroid hormones inhibit the release of TSH, but
later TSH synthesis is also reduced. Nuclear T3 and T4
receptors have been demonstrated on mouse thryotropic cells

3 High doses of

with T3 binding having higher avidity than T,.5
thyroid hormones will decrease the population of TRH receptors
in the pituitary. Another factor in TSH suppression is the
conversion of T4 to T3 which also occurs in the pituitary.54
TSH binds to specific receptor sites on the plasma
membrane of thyroid follicular cells stimulating adenylate
cyclase activity, thereby increasing generation of cyclic AMP
which appears to be the mediator of most of the hormonal
metabolic effect.” The adenylate cyclase system in the
thyroid resembles that in other tissues as there are three

components: the hormone receptor, a guanine nucleotide

regulatory protein, and the catalytic activity.“57 Cyclic AMP

22
actives protein kinases, which phosphorylate appropriate
substrates, altering their function.58 Two of the
phosphorylated substrates have been identified as the
catalytic unit of Na+, K*-ATPase and.a glycoprotein related to

9 Upon administration of TSH,

an aggregation-promoting factor.5
morphologic alterations occur in the thyroid cell that are
associated with enhanced secretion of thyroid hormones. TSH
also influences all phases of iodine metabolism from
stimulation of the iodide transport to secretion of'n,and.Trf°

In addition to the hypothalamus-pituitary-thyroid axis,
the thyroid is regulated by it's requirement for iodine. For
a continued flow of thyroid hormones to the tissues, the
thyroid has a mechanism that adapts to the iodine available,
and the level of thyroid hormones. Administration of excess
iodide reduces the formation of cAMP induced by TSH by unknown

' As previously discussed, excess iodide decreases

mechanisms.6
the iodide transport system and inhibits iodide incorporation

into Tg.62

Thyroid Hormones

Thyroxine is the primary secretory product of the thyroid
gland, however T3 is the active hormone in metabolic
regulation. Thyroid hormones are formed from some of the
iodotyrosine residues that are present in the thyroglobulin
molecule. In studies of the concentrations of T4 and T3 from

Tg metabolism, it has been demonstrated T3 and T3 accumulate

on'

C0

of

$1.1

th

86

23
only after the iodination reaction and DIT formation reach a
plateau. ‘2 This study also revealed that the concentration of
15 levels off after 5-10 minutes, while the I; concentration
continues to increase. This suggests'n,is not the precursor
of‘np The MIT/DIT ratio during the time-course of the study
supported that DIT is the precursor of T4.

In most vertebrates, the majority of the circulating
thyroid hormones are bound to serum proteins.‘53 The major
serum thyroid hormone—binding proteins are thyroxine-binding
globulin (TBG), thyroxine-binding prealbumin (TBPA), and
albumin. The interaction of the hormones with the various
binding proteins is noncovalent and reversible. All three
binding proteins are synthesized in the liver.64 They are
responsible for the maintenance of a large extrathyroid pool
of thyroid hormone, of which only a minute fraction of free
hormone is immediately available to tissues. One of the
functions of the hormone binding proteins is to safeguard the
body from the effect of abrupt fluctuations in hormonal
secretion. .A second function may be to serve as an additional
protection against iodine waste. Body metabolism is dependent
on the quantity of thyroid hormones available to peripheral
tissues. Since the intracellular hormone is in equilibrium
with the free or unbound hormone concentrations in the serum,
the free rather than the total hormone concentrations is a
more accurate indicator (M? the hormone-dependent metabolic

state.

24

Thyroxine is predominantly localized to the extracellular
compartment. The amount of T4 secreted by the thyroid is
fairly constant; however since the majority of extrathyroidal
T4 is bound to serum binding proteins, fluctuations in T4
secretion from.the thyroid.have relatively little influence on
the levels of T3 available to the peripheral tissues. The
production of T3 is primarily from the peripheral conversion
of T4 to T3.

The distribution of T42nuifg to the tissues in the body
is due to a complicated series of relationships between
cellular uptake and intracellular binding proteins. In humans
31 percent of T3 is taken up by the liver and kidney; 44 per
cent is found in muscle, brain, and skin; and the remaining 22
per cent is present in plasma. The distribution of T3 is
different, with only 5 per cent present in the liver, 18 per
cent in plasma, and 75 per cent in remaining tissues.65

The'transformaticw10f the iodothyronines require that the
hormones enter the cell or be closely bound to the cell
membrane. The unbound or "free" thyroid hormones enter the
cell. The peripheral deiodination of T4 results in the
formation of T3 and rT3. This deiodination is the action of
several deiodinases that display both substrate and tissue
specificity. In certain tissues such as brain and pituitary,
nuclear'n,is derived in large part from local deiodination of
T3 within the cell, whereas in other tissues such as liver,

kidney and heart, the principal source of T3 is plasma. This

25
allows certain tissues a degree of autonomy in the regulation
of their nuclear Trf°
Mechanisms of Action

Thyroid hormones act on tissues through various
mechanisms, including transport of amino acids, and
electrolytes from the extracellular environment to the
interior of the cell; synthesis or activation of specific
enzyme proteins within the cell; and enhancement of
intracellular events, including translation and transcription,
that lead to changes in cell size, cell number, and cell
function.

Although it has been assumed that thyroid hormones enter
the cell by passive diffusion, recent studies have suggested
that there is an active transport system into target cells of
the liver and kidney.67 To effect their actions, thyroid
hormones bind to a nuclear receptor and are capable of acting
directly on the genome to initiate transcription or
translation. Thyroid hormone receptors are nonhistone
chromatin proteins found in nuclear fractions.68 Nuclear
receptors for'L,have a high affinity but a limited number of
receptors and do not require cystolic transport for T3 to
gain access to the cell nucleus.69

The dramatic effects of T3 on 02 uptake by tissues
resulting in an increased "metabolic rate" indicates that
thyroid hormones are:a prime determinant.of metabolic activity

which differs qualitatively from tissue to tissue. This

26
increased. metabolic rate is reflected by increased
mitochondrial activity in liver, kidney and muscle. Thyroid
hormones stimulate the sodium pump which operates through
membrane-bound Na*,K*-ATPase and glycolytic pathways leading
to calorigenesis coupled with oxidative phosphorylation.
Other tissue effects include: increased lipolytic activity in
adipose tissue; modulation of gonadotropin secretion by the
pituitary; and altered concentrations of neurotransmitters in
the brain.70 'Tissue responsiveness to thyroid hormones is not
static throughout life, but decreases with aging. Thyroid
hormones ‘modulate the actions of other' hormones through
receptor interactions or by altering transcription or
translation events initiated by other hormones. Finally
thyroid hormones can regulate concentrations of their own
receptors and, under certain circumstances, modify, through

transamination, their own metabolism.”72

Thyroid nggg

Thyroid function testing‘ in veterinary' medicine has
improved in the last decade, however, there is still no single
laboratory test that will positively prove hypothyroidism.
Each of the currently used tests will measure one aspect of
thyroid function at a given time. Because of the complexity
of the interacting regulatory mechanisms which control thyroid
gland secretion, a clinician must understand what a specific

test measures, the technical and interpretative difficulties

27
encountered with the test, the clinical conditions that alter
the test, and what further testing may be necessary.

Thyroid testing' can. be classified according to ‘the
information provided at the functional, etiological or
anatomical levels. Tests such as thyroidal radioiodide
uptake, performed in vivo, directly assess the level of
thyroid gland activity and its ability to perform hormone
biosynthesis. Tests that measure hormone concentration and
its transport in the blood are an indirect assessment of the
level of metabolic activity of the thyroid gland.
Additionally, the presence of substances, such as thyroid
autoantibodies, usually absent in healthy individuals, is
useful in establishing the etiology of some thyoid disorders.

Currently, the most common screening test for dogs is the
measurement of the circulating iodothyronines, T4 and T3.
While many laboratory techniques were used in the past, such
as column chromatography and the protein-bound iodine test,
virtually all tests have been replaced by radioimmunoassays
(RIA) or enzyme-linked immunoassays (ELISA).73 The.T3-uptake
tests are still used in humans to measure the degree of plasma
protein binding.74 'The (Eruptake test is considered an
insensitive test in the dog as T3 and T3 bind less avidly to
the canine plasma proteins than does human T3 and T3. The
normal’n,uptake in the dog is 40% to 60% of the plasma versus

85% to 95% in human plasma.”

28

The measurement of total'n,and total'33is now done using
RIA modified and validated for the dog.76 The principle
behind RIA is the competition of the substance being measured
with a known quantity of radiolabeled substance. Specific
monoclonal or polyclonal antibodies are used to bind the
labeled and unlabeled substances. The antibodies can either
be fixed to a solid support, such as a tube lining, or the
antibody can be added to the antigen and the antibody—ligand
complexes precipitated out of the solution. A standard curve
is based on increasing quantities of the antigen being added
to the assay and a stoichiometric relationship between the
antigen-antibody and antibody-radiolabeled antigen determined.
This relationship can be converted to a straight line by
mathematical transformations, such as the logit-log plot. The
sensitivity of the assay is dependent upon the affinity of the
antibody and the specific activity of the radiolabeled
antigen.

Many nonthyroidal factors can affect concentrations of
serum thyroid hormones. Fasting, systemic illness, drugs and
other factors have been shown to lower total T4 and T3
concentrations“ In humans with either chronic or severe acute
non-thyroidal illnesses, inhibitory substancesfihave been found
in the circulation which impair the binding affinity of T4tx>
serum binding proteins.77 In conditions of inadequate dietary
protein and carbohydrate intake, the low T3 state appears

76.78

important in facilitating protein conservation. Various

29
drugs can alter the concentrations of thyroid hormones by
affecting any level of the hypothalamic-pituitary-thyroid
axis. In addition to altering thyroid hormone synthesis,
drugs can affect the transport, metabolism, and excretion of
Cm,and its derivatives. Some hormones and drugs may alter the
concentration of the binding proteins in the serum affecting
the thyroid transport in the circulation. Thyroid binding
globulin concentrations are increased by estrogens and
decreased by drugs such as glucocorticoids and androgens as
well as protein-losing enteropathies, hepatitis, malnutrition,

7’” In addition to

ovariectomy, and nephrotic syndrome.
decreasing the serum binding proteins, glucocorticoids also
inhibit the conversion of T4tx>T3 in the peripheral tissues.
This results in a decrease in T3 concentrations with an
increase in rT3 concentrations.83 Measuring the "free" or
unbound T4 and T3 is used to measure the concentration of
active hormone in the circulation unaffected by the protein-
binding. Recent studies in the dog indicate the measurement
of unbound T4 as measured by radioimmunoassay may not be
effective in separating hypothyroid dogs from euthyroid dogs
with systemic illnesses.84 Measurement of unbound T, by
equilibrium dialysis may be a more accurate guide to use to
determine concentrations of unbound Tp&‘

Another test to measure the ability of the thyroid gland

to function is the TSH stimulation test. In this test,

exogenous TSH is administered after baseline serum thyroid

30

hormone concentrations are established. Interpretation of the
response to TSH depends on the amount of TSH administered and
the length of the interval between serum samples. Stimulation
with TSH does not cause signs of hyperthyroidism since the
elevated concentrations of T3 are primarily protein bound.85
The maximum response of T4 and T3 concentrations to TSH
stimulation has been shown to be between eight and twelve
hours post TSH.86 The dosage and route of administration have
not seemed to effect the results significantly, however since
bovine TSH is a foreign protein, repeated administration could
cause the production of antibodies and an allergic reaction.
The normal response to TSH administration is a two to four-
fold increase in‘D,and'n,concentrations. ‘Usually there is no
problem differentiating the normal response from primary
hypothyroidism, but occasionally when there is a poor
response, or the resting values are very low, it may be
difficult to differentiate non-thyroidal illness from
hypothyroidism.

The thyroid releasing hormone (TRH) stimulation test is
used to determine if the pituitary can produce TSH to
stimulate the thyroid. Baseline thyroid hormone concentrations
are obtained, followed by the administration of TRH. In the
dog, the response to this test is so poorly documented, that
intrepretation of results is difficult. As a result, this

test is rarely performed.

31

Measurement of canine serum TSH concentrations would be
extremely helpful in the diagnosis of thyroid disorders.
Unfortunately, at this time no reliable assay is available for
commercial use.

A radioactive iodine uptake study can be used to
determine the thyroid gland's functional ability to take up
iodine against a gradient. Radioactive tracers including,
l‘23I,‘25I, or‘”I can be used. '”I is preferred.due to its lower
radiation hazard. Twenty four to seventy two hours following
administration of the radioactive tracer, radioactivity counts
are taken over the thyroid. This test is very expensive,
time-consuming and not generally available. Due to the

environmental hazards this test is rarely performed in dogs.

Autoimmune Thyroiditis

Autoimmune lymphocytic thyroiditis is an endocrine
disease characterized by either the infiltration of the
thyroid gland by lymphocytes or idiopathic follicular
atrophy.87 Spontaneous autoimmune lymphocytic thyroiditis is
present in humans (Hashimoto's thyroiditis), dogs, chickens,
and BUF rats.“89 Clinical signs of hypothyroidism appear as
the thyroid parenchyma is destroyed by either invasion of the
thyroid epithelium by lymphocytes or by the atrophy of thyroid
follicles. Currently it is believed that both of these
disorders are autoimmune in origin and involve both cellular

and. humerol mechanisms, although the exact etiology and

32

31%9' The role of humerol immunity is

pathogenesis are unknown.
suggested by the presence of circulating antibodies to various
thyroid antigens. In addition, immune complexes of Tg and
anti-Tg antibodies have been detected along the fbllicular
basement.membrane. ‘Ehese complexes may lead to thyroid damage
through complement fixation or by activation of killer cell
activity."0 Antibody-dependent cell mediated cytotoxicity
(ADCC) also has been observed in sera from patients with
Hashimoto's thyroiditis. Cytotoxic T cells or the cytotoxic
effects of lymphokines released by activated T cells
responding to thyroid antigens are potential cell mediated
immune mechanisms that may also potentially damage the
thyroid.92

Autoantibodies that react specifically with normal
thyroid gland constituents are detected in the serum of
patients with autoimmune thyroid disease. fflua presence of
these autoantibodies has been used to investigate the
pathogenesis of the disease and also for diagnostic purposes.

Autoantibodies to Tg have been detected in humans,
chickens, and dogs with autoimmune thyroiditis. Normally, Tg
is present in small quantities in the circulation of up to 90
per cent of euthyroid, healthy adult human beings.93
Thyroglobulin has also been detected in normal dog sera and
appears to be present in higher concentrations than in human
beings. The difference in the concentration of serum Tg

between human beings and dogs may be due to assay variability

33
or a different mechanism of release of Tg from the thyroid.94
Autoantibodies to Tg have been found in about 60 per cent of
adults with. Hashimoto's thyroiditis.95 131 one study of
autoantibodies to human Tg, 1 minor and 2 major epitopes were
found on the antigen indicating a polyclonal antibody response
with different isoelectric focusing points.96 Using monoclonal
antibodies to human thyroglobulin, it was found that human Tg
had.5 to 6 epitopes unique to primate and human Tg in addition
to a number of epitopes that cross react with other species.97
Autoantibodies to Tg appear to be directed against the normal
Tg molecule as they do not react with abnormal Tg or Tg that
has been heat denatured. This suggests that the
autoantibodies to Tg recognize epitopes on 2 or more peptide
chain. Complexes of Tg and antibody are deposited in the
thyroid gland which may cause thyroid damage and activate
killer cells. In human beings, Tg autoantibodies are
primarily of the IgG class of immunoglobulin, of the subclass
1.98 However a small percentag of the autoantibodies are IgG
subclass 4, which fixes complement.99 In addition to IgG,
anti-Tg IgA has been detected in 40 to 50 per cent of human
patients with Hashimoto's thyroiditis.loo This is important as
IgA.has been shown to be involved in complement induced tissue
damage in other organs. Generally, Tg autoantibodies do not
fix complement nor are they cytotoxic. The presence of Tg
autoantigens on the thyroid cell surface membranes makes them

susceptible for potential damage by autoantibodies. The

34

demonstration that partially glycosylated asialoagalacto Tg,
which bears the parent molecule's autoantigenic determinants,
is expressed on the thyroid follicular cell surface suggest a
potential pathogenetic route for this antigen in autoimmune
thyroid disease.ml

Autoantibodies to T4 and/or T3 have been reported in all
species reported to develop autoimmune thyroiditis.”89 As
discussed previously, the majority of circulating'n,and T3is
bound to serum binding proteins such as thyroxine-binding
globulin, prealbumin and albumin. Thyroid hormones can be
considered haptens and thus are unable to induce antibody
formation on their own. Much research has been done to study
the underlying immunologic mechanism of the production of
thyroid hormone autoantibodies. Thyroglobulin contains large
quantities of'n,and T3that could provide antigenic sites for
the formation of autoantibodies to ‘R, and Tr. Thyroid
autoantibodies have been found to cross-react with the
thyroxyl-containing sequences of Tg and thus may be a subset
of the Tg autoantibodies.102 However, because thyroid hormone
autoantibodies have been detected in sera samples with no
autoantibodies to Tg under certain pathologic conditions other
serum proteins may also be involved with the formation of
thyroid hormone autoantibodies.103 Clinically, thyroid
hormone autoantibodies in the patient's serum interfere with
RIA of thyroid hormones. This interference results in

confusing information about the patients' thyroid status.

35

Depending on the type of RIA used, the serum thyroid hormone
concentrations can appear to be very high or very low. With
a single-antibody RIA technique using either polyethylene
glycol or absorption or using free thyroid hormones with
charcoal dextran, the thyroid.hormone autoantibodies behave in
the same manner as the radiolabeled hormones added to the
assay. Thus, as levels of bound I25I-T4 or 12S’I-T3 increase, the
value of T3 or T3 determined from the standard curve shows a
lower than actual value. The opposite result is seen with
double-antibody assays, in which the binding of labeled
thyroid hormone by the autoantibodies results in less
radioactivity being bound by the assay antibodies which are
added first. The second antibody recognizes only the assay
antibodies, not the autoantibodies against thyroid hormones
resulting in less tracer hormone being precipitated causing
falsely higher results.

The degree of interference of thyroid autoantibodies on
the RIA is to some extent a reflection on their binding
potential, including their affinity and concentration.
Copping et al.(1989) used reverse-flow electrophoresis to
demonstrate that autoantibodies are capable of transporting
thyroid hormones in vivo in the presence of the other serum
binding proteins.“” We demonstrated that the autoantibodies
continue to bind thyroid hormones in the presence of 8-
anilino-l-naphthalene-sulfonic acid (ANS)in dogs, while

binding to the other serum binding proteins was blocked.”5

36
This confirmed that the thyroid autoantibodies had both the
affinity and the concentration required to be present in the
equilibrium between the thyroid hormones and the binding
proteins.

The failure of the thyroid hormone autoantibodies to
induce hypothyroidism by their binding to thyroid hormones in
circulation is not surprising as they function along the lines
of the other serum binding proteins, such as TBG, prealbumin
and albumin. As long as the patient has sufficient thyroid
reserve, enhanced plasma binding of 'D, (due to thyroid
autoantibodieS»or serum binding proteins) produces a transient
decrease in unbound'n,which serves as a stimulus for enhanced
T3 secretion via negative feedback resulting in a new steady
state with an increase in total T4 but no change in the
unbound TM. These observations have been confirmed in animals
with experimentally induced thyroid autoantibodies; the
unbound T, serum concentrations was reported normal while
total T3 serum concentrations were markedly elevated.106

The microsomal antigen has been shown to be the membrane
bound, glycosylated enzyme, TPO. These autoantibodies have
been shown to be highly correlated to thyroid pathology in
human beings, with approximately 95 per cent of adults with
Hashimoto's thyroiditis having anti-TPO autoantibodies.m7 It
has been shown that fusion of the exocytotic vesicles carrying
the newly synthesized Tg with the apical plasma membrane leads

to cell surface expression of the TPO. The exposed TPO

37
results in complement-mediated cytotoxicity in sera from
patients with autoimmune thyroid disease.”8 Using primary
cultures of human thyroid cells, it was shown that the
expression of TPO antigen in thyroid cells is dependent.on.TSH
stimulation, through pathways which involve cAMP production
and synthesis.109

It has also been reported that autoantibodies to TPO are
capable of inhibiting the catalytic activity of TPO.”0
Okamoto et al. (1989) found that the mean serum free T,
concentration was significantly lower in patients with TPO
autoantibodies suggesting that these autoantibodies inhibit
thyroid function, but no simple relationship between TPO
autoantibodies and thyroid function was demonstrated.“1 It is
difficult to assume that, TPO autoantibodies inhibit the
activity of the enzyme, because of its inaccessibility to the
cytoplasm of thyroid follicular cells.

A study by Kohno et al.(1986) found that IgG fractions
purified by ion exchange chromatography or protein A-Sepharose
column from human patients with chronic thyroiditis inhibited
the activity of TPO by both guaiacol assay method and iodide
assay method.”2 This same study suggests that the antigenic
determinants on the TPO molecules are heterogeneous with two
binding sites, one for the tyrosyl residue in Tg and one for
the aromatic donor. Recognition of the inhibition of TPO

activity by either the guaiacol assay or the iodide assay

38

might be caused by autoantibodies which recognize either one
of the sites, both sites or none.“3

The effector mechanisms for tissue damage in Hashimoto's
thyroiditis in human beings is fairly well defined with the
interplay of several separate factors. The important
cytotoxic effects of microsomal autoantibodies are
recognized.“‘ .Antibodies can also be demonstrated in immune
complexes deposited on the follicular basement membranes and
these can fix complement to mediate tissue damage. ADCC may
be important in Hashimoto's thyroiditis as supported by the
increase in killer cells which produce this form of
cytotoxicity.“5 Although the evidence points toward.ADCC, the
role of antibody-dependent complement-mediated cytotoxicity
and direct T cell killing as relevant pathogenetic mechanisms
in the human disease is more complex. What initiatesthe
process off appears to be related to the role of the antigen-
presenting cell.”6 It has been demonstrated that patients
with Hashimoto's thyroiditis have aberrant expression of class
II HLA antigens on thyroid cells, probably in response to
stimulation by lymphocyte-secreted interferon-7.”‘”9 It has
been further suggested that the thyroid follicular cells may
have the capacity to present their own self-autoantigens in
conjunction with the class II antigen to the immune system.‘20

Considering' that. most. patients with. Hashimoto's
thyroiditis express HLA-DR antigens in the thyroid epithelial

cells, and that TSH enhances the interferon-y-induced HLA-DR

39

expression in cultured human thyrocytes, TSH may pay a role in
the expression and/or maintenance of thyroid immune responses
in Hashimoto's thyroiditis by developing TPO antigens.‘2"122

Although the pathogenesis of autoimmune thyroiditis in
human beings has been extensively studied, the cause cannot be
attributed to abnormal T cell function. Lymphocytes of
patients with lymphocytic thyroiditis generally do not have a
positive blastogenic response to Tg, although in one study 71
per cent of human patients with lymphocytic thyroiditis and 9
per cent of the normal control patients had a positive

3 Another test for T cell function uses the

response to Tg.12
production of the lymphokine, migration inhibition factor
(MIF) released when T lymphocytes encounter an antigen to
which they are sensitized. The results of this test suggest
an antigen-specific defect in suppressor T cell function.‘24
When T cells are cloned from the thyroid tissue of patients
with Hashimoto's thyroiditis, a large number are reported to
have the T4+T8+ phenotype of cytotoxic/suppressor cells and
are cytolytic.”5

Investigations of B lymphocyte cell function in human
patients with Hashimoto's thyroiditis found no generalized B
cell hyper-reactivity to exogenous antigen. When cultured B
cells were stimulated with pokeweed mitogen (PWM) no
suppressor effect.by normal T cells on autoantibody production

was observed and normal T cells could provide efficient helper

function.126 In contrast, when normal B cells were combined

40
with T cells from patients with Hashimoto's thyroiditis no Tg

autoantibodies were made.“128

With the use of lymphocytes
freshly collected from the thyroid glands of patients with
Hashimoto's thyroiditis, spontaneous Tg autoantibody formation
synthesis occurred without PWM stimulation; thus, the target
organ is an important source of autoantibodies in autoimmune
thyroid disease.129

Hypothyroidism is an important endocrine disease in dogs
and is often associated with dermatologic abnormalities such
as alopecia, seborrhea and pyoderma. Other clinical signs
seen in hypothyroid dogs include obesity, lethargy,
hypothermia, bradycardia and neuromuscular disorders.

Autoantibodies to thyroid antigens have been detected in
canine serum. suggesting' the presence of immune mediated
thyroid disease. Previous studies have documented autoimmune
thyroiditis as the most common cause of canine
hypothyroidistUQD‘ While both ‘the idiopathic follicular
atrophy and the lymphocytic forms have been documented in the
dog, lymphocytic thyroiditis is more common.”°

Idiopathic thyroiditis is characterized by loss of the
thyroid parenchyma and replacement by adipose and connective
tissue. It lacks the inflammatory infiltrate seen with
lymphocytic thyroiditis. Degeneration of follicular cells may
be seen.132 Lymphocytic thyroiditis is characterized by a

focal to diffuse infiltration of the thyroid gland by

lymphocytes, plasma cells, and macrophages. Thyroid follicle

 

des

mi:

al
no:
de:
sec
in
exa
la:
is
par
th}

th§

pre
the
far
801
in
She
inh
her
tim.

In;

41
destruction and lymphocytic nodule formation may be present as
the disease progresses. In advanced cases fibrosis with
minimal inflammatory infiltrate remaining may be seen.133

A study of the progression of the disease by Conaway et
al (1985), indicated three stages. Early. changes were
noninflammatory and characterized by slight follicular
degeneration causing collapse of follicular lumens. The
second phase was inflammatory with infiltration of
lymphocytes, plasma cells and macrophages. Ultrastructural
examination at this time noted focal thickening of the basal
lamina, suggesting immune complex deposits. The final stage
is fibrous connective tissue replacement of the normal thyroid
parenchyma. One of the dogs in the study had end-stage
thyroiditis with histologic findings similar to idiopathic
thyroid atrophy.134

In human beings, lymphocytic thyroiditis is more
prevalent in women; however this does not appear to be true in
the dog with males and females equally affectedfi”137 A
familial tendency has demonstrated.in.Great Danes, beagles and
Borzois.‘3‘H38 An increased prevalence of thyroid autoantibodies
in some breeds including Doberman Pinchers, Old English
Sheepdogs, and Irish Setters suggests that thyroiditis is

inherited in these breeds also.”“”

A definite pattern of
heritability has not been established in the dog as of this
time, however it appears to be polygenic as in human beings.”0

In a study with Borzoi dogs lymphocytic thyroiditis appeared

42
to be an autosomal recessive trait.”5 Further studies are
needed to confirm the mode of inheritance.

In a colony of laboratory Beagles, a 16.2% incidence of
thyroiditis was detected in young adults of both sexes.136 The
presence of thyroid autoantibodies has also been detected in
dogs with lymphocytic thyroiditis.”““*“ Autoantibodies to
canine Tg have been detected by tanned cell hemagglutination,
chromic chloride passive hemagglutination and ELISA
techniques. The incidence of Tg autoantibodies varied with
the type of assay and the criteria used to select the
population of dogs to be tested. However, the incidence of Tg
autoantibodies detected in dogs with hypothyroidism is very
similar to the incidence of autoantibodies in humans with

HaShimoto I S theriditiS .105.l4l.l42.l44

In addition to Tg
autoantibodies, autoantibodies to the thyroid hormones,'n,and
T3 have been detected in dogs with lymphocytic thyroiditis.'°5'”5'
”7 Complement fixation and ELISA techniques have been used
previously to measure microsomal autoantibodies in the dog,“‘2‘I43
These studies found a low incidence of microsomal
autoantibodies in dogs, however, since these studies were
conducted, the techniques to isolate and purify 'thyroid
microsomes have been greatly improved.

Haines et al (1984) investigated the presence of
antithyroglobulin. (anti-Tg) autoantibodies. and. ‘thyroid

microsomal autoantibodies in 34 dogs with low basal serum T4

concentrations.H2 Autoantibodies to thyroid microsomal

43
autoantibodies were detected by ELISA in 29% of those dogs.
In contrast, autoantibodies to Tg were found in 59% of the
dogs. Autoantibodies to Tg have also been detected in the
sera of dogs with non-thyroidal endocrine disorders. In one
study, 43% of 65 dogs with nonthyroidal endocrine disease had
autoantibodies to Tg as detected by ELISA, however, Gosselin
et a1 (1980) did not detect autoantibodies to Tg using chromic
chloride hemagglutination in sera from a similar population of
dogs.141 Human patients with immune-mediated endocrine
disorders which resulted in diabetes mellitus or
hypoadrenocorticism are predisposed to the development of

8

thyroid autoantibodies.l4 Relatives of dogs with lymphocytic

thyroiditis commonly are reported to have antibodies to Tg,
and the same is true in people.“I48

In a study in our lab of canine sera collected from dogs
suspected of having autoimmune thyroiditis resulting in
clinical hypothyroidism, we detected autoantibodies to'n,and

T3jJ126.1% and 32.8% respectively.105

Of the 119 serum samples
used in the study, 2.5% had autoantibodies to only T3, 0.8%
had autoantibodies to only T“ and 3.4% had autoantibodies to
T3 and T,. This suggests that autoantibodies to the thyroid
hormones may bind to another protein occasionally in the dog
to induce the production of these autoantibodies. In an
earlier study of dog serum submitted for the measurement of

thyroid hormone concentrations, only 0.2% of the samples were

found to have autoantibodies to T3 and/or T3, however

 

id

e]

R]

at
dc
fc
5F
pr

CC

In
We

ho-

44

identification of these autoantibodies was based on a marked
alteration in the results of RIA of thyroid hormones, not the
direct demonstration of autoantibodies.”7 In studies of
thyroid hormone autoantibodies in human beings, the prevalence
of autoantibodies to T, and T3 was found to be dependent on
method of detection, ranging from 40% (6 of 15) detected by
electrophoresis techniques to 2.6% (1 of 38) from a modified
RIA technique."““5o

Mizejewski et al (1971), used complement fixation to
measure microsomal autoantibodies in a colony of beagle
dogs.M3 Using that assay, 7 of 11 dogs with thyroid pathology
had significantly high titers to complement fixing antibodies.
Fluorescent antibody staining was also performed on thyroid
biopsies from these dogs. A staining pattern characterized as
"cracked mirror" appearance as described by Balfour et al
(1961), suggesting the presence of both Tg and microsomal
autoantibodies was seen in the thyroid glands in the same 7
dogs.”1 In other studies, immunofluorescence methods were
found to be technically unreliable and time consuming.‘30 It
appears that in the dog, autoantibodies to Tg are more
prevalent than the microsomal antigen. This is in direct
conflict with results found in humans, but is similar to
chickens with spontaneous autoimmune thyroiditis and mice.“‘“50
In a study of Tg autoantibodies using both an ELISA and

Western blot assay, there was a significant correlation,

however, some sera were positive on the Western blot and

45
negative on the ELISA and some sera were positive with the
ELISA and negative on Western blot assay. This suggests a
heterogeneous population of antibodies.

The recognition that the major component of the
autoimmune response is generated within the thyroid and that
the thyroid has the capacity to invoke these responses now
makes the thyroid not only the target of the autoimmune
response but also the source.”8 What initiates these events
is unknown, but environmental influences are suspected.
Iodine ingestion is increasingly implicated in the possible

generation of the autoimmune response.l2

REFERENCES
1. Evans HE, Christensen GC.(eds) The Endocrine System. In:

Anatomy of the Dog. Philadelphia, W B Saunders. 1979,

pp.611-612.

2. Chastain CB, Ganjam VK.(eds) The Normal Thyroid and
Clinical Tests of Its Functions. In: Clinical
Endocrinology of Companion Animals. Philadelphia, Lea

and Febiger. 1986 pp. 113-134.

3. Ettinger SJ.(ed) Thyroid Diseases. In: Textbook of
Veterinary Internal.Medicine. Philadelphia,ivEBSaunders.

1989, p 1632.

46
Martin SL, Capen CC. The endocrine system. In: Catcott
R.(ed) Canine IMedicine. Santa Barbara, American

Veterinary Publications, 1979, pp. 1087-1205.

Ericson LE. Exocytosis and endocytosis in the thyroid

follicle cell. Mol Cell Endocrinol 22:1-24, 1981.

Pearse Age: The Characteristics of the C cell and Their
Significance in Relation to Those of Other Endocrine
Polypeptide Cells and to the Synthesis, Storage, and
Secretion of Calcitonin. In: Calcitonin. New York,

Springer-Verlag, 1970, p 125.

Spiro MJ, Spiro RG, Bhoyroo VD. Lipid-saccharide
intermediates in glycoprotein synthesis. I. Formation of
an oligosaccharide-lipid by thyroid slices and exaluation
of its role in protein glycosylation. J Biol Chem

251:6400-6408, 1976.

Godelaine D, Spiro MJ, Spiro RG. Processing of the
carbohydrate units of thyroglobulin. J Biol Chem

256:6409-6419, 1976.

10.

11.

12.

13.

14.

47
Bj5rkman U, Ekholm R. Intracellular transport and
secretion of thyroglobulin in isolated thyroid follicles.
In: Eggo MC, Burrow GN. (eds) Thyroglobulin. New York,

Raven Press, 1984, p 127.

Ring P, Bjorkman U, Ekholm R. Localization of the
incorporation of 3H-galactose and 3H-sialic acid into
thyroglobulin in relation to the block of intracellular
transport induced by monesin. Cell Tissue Res 250:149-

156, 1987.

Bjérkman U, Ekholm R, Ericson LE, et al. Transport of
thyroglobulin and thyroid peroxidase in the thyroid

follicle cell. Mol Cell Endocrinol 5:3-17, 1976.

Wolff J. Transport of iodide and other anions in the

thyroid gland. Physiol Rev 44:45—90, 1964.

Wolff J, Halmi N. Thyroidal iodide transport. V. The role
of Na+-K+ activated ouabain sensitive ATPase. J Biol Chem

238:847-851, 1963.

Wilson B, Raghupathy E, Toroue T, et al.‘TSH-like actions
of dibutyl-cAMP on isolated bovine thyroid cells.

Endocrinology 83:877-884, 1968.

15.

16.

17.

18.

19.

20.

48
Ingbar SH. Autoregulation of the thyroid. Response to
iodide excess and depletion. Mayo Clin Proc 47:814-823,

1972.

Braverman LE, Ingbar SH. Changes in thyroidal function
during adaptation to large doses of iodide. J Clin Invest

42:1216-1231, 1963.

Rappoport B, West MN, Ingbar SH. 0n the mechansim of
inhibition by iodine of the thyroid adenylate cyclase
response to thyrotropic hormone. Endocrinology 99:11-22,

1976.

van der Walt B, Kotze B, van Jaarsveld PP, et al.
Evidence that thyroglobulin contains non-identical half-

molecule subunits. J Biol Chem 253:1853-1858, 1978.

Tsuji T, Yamamoto K, Irimura T, et al. The structure of
carbohydrate unit A of porcine thyroglobulin. Biochem J

195:691-700. 1981.

Yamamoto K, Tsuji T, Irimura T, et al. The structure of
carbohydrate unit B of porcine thyroblobulin. Biochem J

195:701-713, 1981.

21.

22.

23.

24.

25.

26.

 

49
Rolland M, Montfort MF, Lissitzky S. Effiency of
thyroglobul in as a thyroid hormone forming protein.

Biochem Biophys Acta 303:338-347, 1973.

Marriq C, Arnaud C, Rolland M, et al. An approach to the
structure of thyroglobulin. Hormone-forming sequences in

porcine thyroglobulin. Eur J Biochem 111:33-47, 1980.

Edelhoch G, Carlomagno S, Salvatore G. Iodine and the
structure of thyroglobulin. Arch Biochem Biophys 134:264-

265, 1969.

Andreoli M, Sena L, Edelhech H, et al. The noncovalent
subunit structure of human thyroglobulin. Arch Biochem

Biophys 134:242-248, 1969.

Overholm T, Ericson LE. Intraluminal iodination of

thyroglobulin. Endocrinology 114:827-835, 1984.

Wollman SH, Loewenstein JE. Rates of colloid droplet and
apical vesicle production and membrane turnover during
thyroglobulin secretion and resorption. Endocrinology

93:248-252, 1973.

27.

28.

29.

300

31.

32.

50
Fouchier F, Mego JL, Damg J, et al. ATP-induced
stimulation. of ‘the intralysosomal hydrolysis of
thyroglobulin. Evidence for an ATP-driven proton pump in

thyroid lysosmes. Horm Metab Res 16:359—362, 1984.

Lissitzky S. Deiodination of iodotyrosines. In: Reinwein
D, Klein E. (eds) Diminished Thyroid Hormone Formation.
Possible Causes and Clinical Aspects. Stuttgart,

Schattauer Verlag, 1982, pp 49-61.

DeGroot LJ, Buhler U. Effect of perchlorate and
methimazole on iodine metabolism. Acta Endocrinol 68:696-

706, 1971.

Chopra IJ, Solomon DH, Chopra U, et al. Pathways of
metabolism of thyroid hormones. Recent Prog Horm Res

34:521-567, 1978.

Greer MA, Grimm Y, Studer H. Qualitative changes in the
secretion of thyroid homones induced by iodine

deficiency. Endocrinology 83:1193-1198, 1968.

Greer MA, Allen CF, Torresani J, et al. TSH stimulation
of iodothyronine formation in prelabeled thyroglobulin of

hypophysectomized rats. Endocrinology 94: 1224-1231, 1974 .

33.

34.

35.

36.

37.

38.

51
Laurberg P. Enhanced release offfi and FD,compared to T,
from thyroglobulin during autolysis of dog thyroid

homogenate. Acta Endocrinol 97:486-490, 1981.

Taurog A. Thyroid peroxidase and thyroxine biosynthesis.

Recent Prog Horm Res 26:189-247, 1970.

Kimura S, Kotani T, McBride 0W, et al. Human thyroid
peroxidase. Complete cDNA and protein sequence,
chromosome mapping and identification of two alternately

spliced mRNAs. Proc Natl Acad Sci USA 8:5555-5559, 1987.

Rawitch AB, Taurog A, Chernoff SB, et al. Hog thyroid
peroxidase: physical, chemical, and catalytic properties
of the highly purified enzyme. Arch Biochem Biophys

194:244-257, 1979.

Nunez J, Mauchamp J, Pommier J, et al. Relationship
between iodination and conformation of thyroglobulin.

Biochem Biophys Res Commun 23:761-768, 1966a.

Ohtaki S, Mashimo K, Yamazaki I. Hydogen peroxide
generating system in hog thyroid microsomes. Biochem

Biophys Acta 292:825-833, 1973.

39.

40.

41.

42.

43.

44.

52
Magnusson RP, Taurog A, Dorris ML. Mechanism of iodide-
dependent cataltic activity of thyroid peroxidase and

lactoperoxidase. J Biol Chem 259:197-205, 1984.

Taurog A. Biosynthesis of iodoamino acids. In: Greep RO,
Astwood EB (eds): Handbook of Physiology, Vol III,
Endocrinology . Washington, DC, American Physiological

Society, 1974, p.101.

Simon C, Roques M, Torresani J, et al. Effect of
propylthiouracil on the iodination and maturation of rat

thyroglobulin. Acta Endocrinol (Kbh) 53:271-285, 1966.

Nagasaka A, DeGroot LJ, Hati R, et al. Studies on the
biosynthesis of thyroid hormone: Reconstruction of a
defined in Vitro iodinating system. Endocrinology 88:486-

497, 1971.

DeCrombrugghe B, Edelhoch H, Beckers C, et al. Effects of
iodination on sedimentation and iodoamino acid synthesis.

J Biol Chem 242:5681-5685, 1967.

Johnson TB, Tewkesbury LB. The oxidation of 3,5-
diiodotyrosine to thyroxine. Proc Natl Acad Sci USA

28:73-77, 1942.

45.

46.

47.

48.

49.

53
Baas F, Van Ommen GJB, Bikker H, et al. The human
thyroglobulin gene is over 300 kb long and contains
introns of up to 64 kb. Nucleic Acid Res 14:5171-5186,

1986.

Christophe D, Pohl V, Van Henverswyn B, et al. Isolation
and characterization of a fragment of rat thyroglobulin

gene. Biochem Biophys Res Commun 105:1166-1175, 1982.

Van Ommen GJV, Stuka A, de Vijlder JJM. Structure of
coding and flanking sequences of normal and defective

goat thyroglobulin. Ann Endocrinol (Paris), 44:22A, 1983.

Hékfelt T, Elde R, Foxe K, et al. Aminergic and
peptidergic pathways in the nervous system with special
reference to the hypothalamus. In: Reichlin S,
Baldessarini RJ, Martin JB (eds) The Hypothalamus ARNMD,

Vol 16. New York, Raven Press, 1978, pp 69-136.

Carlson HE, Jacobs LS, Daughaday, WH. Growth hormone,
thyrotropin and prolactin response to thyrotropin-
releasing hormone following diethylstilbestrol

pretreatment..ICHJJIEndocrinol.Metab 36:1069-1073, 1973.

50.

51.

52.

53.

54.

55.

54
Brown MJ, Hedge GA. Multiple effects of glucocorticoids
on TSH secretion in unanesthetized rats. Endocrinology

92:1305-1311, 1973.

Nicoloff JT, Fisher DA, Appleman MD. The role of
glucocorticoids in the regulation of thyroid function in

man. J Clin Invest 49:1922-1929, 1970.

Gershengorn MC. Thyroid hormone regulation of thyrotropin
production and interaction with thyrotropin releasing
hormone in thyrotropic cells in culture. In: Oppenheimer
JH, Samuels HH.(eds) Molecular Basis of Thyroid Hormone

Action. New York, Academic Press, 1983, pp 387-412.

Hinkle PM, Goh KBC. Regulation of thyrotropin-releasing
hormone receptors and responses by L-triiodothyronine in
dispersed rat pituitary cell cultures. Endocrinology

110:1725-1731, 1982.

Larsen PR. Thyroid-pituitary interaction. N Engl J Med

306:23-32, 1982.

Amir SM, Goldfine ID, Ingbar SH. Properties of the
interaction between bovine thyrotropin and thyroid plasma

membranes. J Biol Chem 251:4693-4699, 1976.

56.

57.

58.

59.

60.

61.

55
Kotani M, Kariya T, Field JB. Studies of thyroid-
stimulating hormone binding to bovine thyroid plasma

membranes. Metabolism 24:959-971, 1975.

Rodbell M. The role of hormone receptors and GTP-
regulatory proteins in membrane transduction. Nature

284:17-22, 1980.

Tao M. Mechanism of activation by adenosine 3'5'-cyclic
monophosphate of a protein phosphokinase from rabbit

reticulocytes. Proc Natl Acad Sci USA 67:408-414, 1970.

Lecocq RE, Lamy F, Dumon JE. Pattern of protein
phosphorylation in intact stimulated cells: Thyrotropin

and dog thyroid. Eur J Biochem 102:147-152, 1979.

Halmi NS, Granner DK, Doughman DJ, et al. Biphasic effect
of TSH on thyroid iodide concentration in rats.

Endocrinology 67:70-81, 1960.

Van Sande J, Grenier G, Willems C, et al. Inhibition by
iodide of the activation of the thyroid 3',5'-cyclic AMP

system. Endocrinology 96:781-786, 1975.

62.

63.

64.

65.

66.

67.

56
Ingbar SH. Effects of iodine: Autoregulation of the
thyroid. In: Werner SC, Ingbar SH. (eds) The Thyroid (Ed

4). New York, Harper & Row, 1978, p 206.

Farer LS, Robbins J, Blumberg BS, et al. Thyroxine-serum
protein complexes in various animals. Endocrinology

70:686-696, 1962.

Murata Y, Sarne DH, Horwitz AL, et al. Characterization
of thyroxine-binding globulin secreted by a human
hepatoma cell line. J Clin Endocrinol Metab 60:472-478,

1985.

Irvine CHG. Four compartment model of thyroxine
metabolism. In: Harland WA, Orr.JS. (eds) Thyroid.Hormone

Metabolism. London, Academic Press, 1975, p263.

Crantz FR, Silva JE, Larsen PR. An analysis of the
sources and quantity of 3,5,3'-triiodothyronine
specifically bound to nuclear receptors in rat cerebral

cortex and cerebellum. Endocrinology 110:367-375, 1982.

Oppenheimer JH, Schwartz HL. Stereospecific transport of
triiodothyronine from plasma to cytosol and from cytosol
to nucleus in rat liver, kidney, brain, and heart. J Clin

Invest 75:147-154, 1985.

68.

69.

70.

71.

72.

57
DeGroot LJ, Strausser J. Binding of T3 in rat liver

nuclei. Endocrinology 95:74-83, 1974.

Latham KR, Ring JC, Baxter JD. Soubilized nuclear
"receptors" for thyroid hormones: Physical
characteristics and binding properties , evidence for

multiple forms. J Biol Chem 251:7388-7397, 1976.

Rastogi RB, Singhal RL. Influence of neonatal and adult
hyperthyroidism on behavior and biosynthetic capacity for
norepinephrine, dopamine, and 5-hydroxytryptamine in rat

brain. J Pharm Exper Ther 198:609-618, 1976.

Nakamura M, Hamada S, Imura M. Sequential changes in rat
liver nuclear triiodothyronine receptors and
mitochondrial a-glycerophosphate dehydrogenase activity
after administration of triiodothyronine. Biochem J

182:377-382, 1979.

Fishman M, Huang YP, Tergis DC, et al. Relation of
triiodothyronine and reverse triiodothyronine
administration in rats to hepatic L-triiodothyronine
aminotransferase activity. Endocrinology 100:1055-1059,

1977.

73.

74.

75.

76.

77.

78.

58
Kaneko JJ. Clinical applications of the thyroidal 1‘31

uptake tet in the dog. JAVMA 135:516-520, 1959.

Bullock L. Protein bound iodine determination as a
diagnostic aid for canine hypothyroidism. JAVMA 156:892-

899, 1970.

Thomson RAE, Michaelson SM. A source of false iodine-131
uptake and protein bound iodine values in dogs. AJVR

28:1623-1625, 1967.

Panciera DL, MacEwen EG, Atkins CE, et al. Thyroid
function tests in euthyroid dogs treated with L-

thyroxine. Am J Vet Res 1990 51:22-25, 1990.

Chopra IJ, Solomon DH, Chu-Teso GN, et al. An inhibitor
of the binding of thyroid hormones to serum proteins is
present in extra thyroidal tissues. Science 215:407-409,

1982.

Carter WJ, Shakir KM, Hodges S, et al. Effect of thyroid
hormones on metabolic adaptation to fasting. Metabolism

24:1177-1183, 1975.

79.

80.

81.

82.

83.

59
Kemppainen RJ, Thompson FN, Lorenz MD et al. Effects of
prednisone on thyroid and gonadal function in dogs. J

Endocrinol 96:293-302, 1983.

Remers TJ, Mummery LK, ETCann, et al. Effects of
reproductive state on concentrations of thyroxine,
3,5,3"-triiodothyronine and cortisol in serum of dogs.

Biol Reprod 31:148-154, 1984.

Van der Walt JA, Van der Walt LA, LeRoux PH. Functional
endocrine modification of the thyroid following
ovariectomy in the canine. J S Afr Vet Assn 54:225-227,

1983.

Woltz HH, Thompdon GN, Kemppainen RJ. Effect of
prednisone on thyroid gland morphology and
triiodothyronine concentrations in the dog. AJVR 44:2000-

2003, 1983.

Premachandra PM, Lang 8. Circulating reverse
triiodothyronine (FTQ in the dog. Lif Sci 20:1449-1454,

1977.

84.

85.

86.

87.

88.

89.

60
Nelson RW, Ihle SL, Feldman EC. Serum free thyroxine
concentrations in healthy dogs, dogs with hypothyroidism,
and euthyroid dogs with concurrent illness. JAVMA

198:1401-1407, 1991.

Hoge WR, Lund JE, Blakemore JC. Response to thyrotropin
as a diagnostic aid for canine hypothyroidism. JAAHA

10:167-170, 1974.

Belshaw BE. Thyroid diseases. In: Ettinger SJ. (ed)
Textbook of Veterinary Internal Medicine. lst Ed. Vol.2.

Philadelphia, Saunders, 1983, p 1592.

Pinchera A, Fenzi G, Chiovato L, et al. Autoimmune
hypothyroidism. In: Drexhage HA. (ed) The Thyroid and
Autoimmunity. Amsterdam, Elsevier Science Pub. BV (Biomed

Div), 1986, pp 73-80.

Gosselin SJ, Capen CC, Martin CL. Histologic and
ultrastructural evaluation of thyroid lesions associated

with hypothyroidism in dogs. Vet Pathol 18:299-309, 1981.

Beisel KW, Rose NR. Genetics of the autoimmune
endocrinopathies: animal models. In: Davies TF.(ed)
Autoimmune Endocrine Disease. New York, John Wiley &

Sons, Inc., 1983, pp 41-58.

90.

91.

92.

93.

94.

95.

61
McLachlan S, Jansson R, Whitehead P, et al.
Hypothyroidism and autoimmune thyroid destruction: a role
for' thyroid. autoantibodies? In: Drexage .HA.(ed) The
Thyrofld and Autoimmunity. Amsterdam, Elsevier Science

Pub. BV (Biomed Div), 1986, pp 99-105.

Volpe' R. The role of autoimmunity in hypoendocrine and

hyperendocrine function. Ann Int Med 87:86-99, 1977.

Schleusener H, Bogner U. Cytotoxic mechanism in
autoimmune thyroiditis. In: Walfish PG, Wall JR, Volpe'
R. (eds) Autoimmunity and the Thyroid. Toronto, Academic

Press, 1985, pp 95-107.

Van Herle AJ, Uller RP, Matthews NL, et al.
Radioimmunoassay for measurement of thyroglobulin in

human serum. J Clin Invest 52:1320-1327, 1973.

Verschueren CP, Selman PJ, de Vijlder JJ, et al.
Characterization of and radioimmunoassay for canine

thyroglobulin. Dom An Endocrinol 8:509-519, 1991.

Marriotti S, Pinchera A, Vitti P, et al. Comparison of
radioassay and haemagglutination methods for anti-thyroid

microsomal antibodies. ClinIbanmmunol 34:118-125, 1978.

96.

97.

98.

953.

lCHD.

62
Nye L, Ponles de Carvalho LC, Roitt IM. Restrictions in
the response to autologous thyroglobulin in the human.

Clin Exp Immunol 41:252-263, 1980.

Chan CTJ, Byfield PGH, Himsworth RL, et al. Human
autoantibodies to thyroglobulin are directed towards a
restricted number of human specific epitopes. Clin Exp

Immunol 70:516-523, 1987.

McLachlan Sm, Stephenson A, Devey M, et al. Functional
affinity and patterns of inheritance of microsomal and
thyroglobulin autoantibodies of different IgG subclasses.

J Endocrinol 117 (Supplement 267):abstract, 1988.

Fukuma N, McLachlan SM, Petersen VB, et al. Human
thyroglobulin autoantibodies of subclasses 1962 and IgG4
bind to different epitopes on thyroglobulin. Immunology

67:129-131, 1989.

Weetman AP. IgA class and subclass thyroid auto-
antibodies in Graves' disease and Hashimoto's
thyroiditis. Int Archs Allergy Appl Immun 83:432-435,

1987.

101.

102

1103.

1134.

1055.

63
Roitt IM, Pujol-Borrell R, Hanafusa TL et al.
Asialoagalacto-thyroglobulin binds to the surface of
human thyroid cells at a site distinct from the
microsomal autoantigen. Clin Exp Immunol 56:129-134,

1984.

Pearce CJ, Byfield PGH, Edmonds.CJ, et al..Autoantibodies
to thyroglobulin cross-reacting with iodothyronines. Clin

Endocrinol (Oxf) 15:1-10, 1981.

Hehrmann R, Hoffken B, Bon Zur Muhlen A, et al. Anti-
thyroid hormone autoantibodies under experimental and

clinical conditions. Horm Metab Res 9:326-332, 1977.

Copping S, Byfield PGH. The role of thyroid hormone
autoantibodies in serum transport. Acta Endocrinol

(Copenh) 121:551-559, 1989.

Thacker ET, Refsal KR, Bull RW. Prevalence of
autoantibodies to thyroglobulin, thyroxine, or
triiodothyronine and relationship of autoantibodies and
serum concentrations of iodothyronines in dogs. AJVR

53:449-453, 1992.

106.

107.

108.

109.

110.

64
Nieschlag E, Herrmann J, Usadel KH, et al. Thyroid
hypertrophy and hyperfunction caused by active
immunization with thyroid hormones. J. Endocrinol 57:555-

556, 1973.

Czarnocka B, Ruf J, Ferrand M, et al. Purification of the
human thyroid peroxidase and its identification as the
microsomal antigen involved in autoimmune thyroid

diseases. FEBS Lett 190:147-152, 1985.

Khoury EL, Hammond L, Bottazzo GT, et al. Presence of the
organ-specfic "microsomal" autoantigen on the surface of
human thyroid cells in culture: It's involvement in
complement-mediated cytotoxicity. Clin Exp Immunol

45:316-328, 1981.

Chiovato L, Vitti P, Cuchi P, et al. The expression of
the microsomal/peroxidase autoantigen in human thyroid
cells is thyrotrophin dependent. Clin Exp Immunol 76:47-

53, 1989.

Kohno Y, Hiyama Y, Shimojyo N, et al. Autoantibodies to
thyroid peroxidase in patients with chronic thyroiditis;
Effect of antibody binding on enzyme activities. Clin Exp

Immunol 65:534-541, 1986.

111.

112.

113.

114.

115.

65
Okamoto Y, Hamada N, Saito H, et al. Thyroid peroxidase
activity-inhibiting immunoglobulins in patients with
autoimmune 'thyroid. disease. .I (nix: Endocrinol. Metab

68:730-734, 1989.

Kohno Y, Hiyama Y, Shimojo N, et al. Autoantibodies to
thyroid peroxidase in patients with chronic thyroiditis:
effect of antibody binding on enzyme activities. Clin Exp

Immunol 65:534-541, 1986.

Doble ND, Banga JP, Pope R, et al. Autoantibodies to the
thyroid ndorosomal/thyroid peroxidase antigen are
polyclonal and directed to several distinct antigenic

sites. Immunol 64:23-29, 1988.

Bogner U, Schleusener H, Wall JR. Antibody-dependent cell
mediated cytotoxicity against human thyroid cells in
Hashimoto's thyroiditis but not Graves' disease. J Clin

Endocrinol Metab 59:734-738, 1986.

Calder EA, Irvine WJ. Cell-mediated immunity and immune
complexes in thyroid disease. J Clin Endocrinol Metab

4:287-291, 1975.

116.

117.

118.

119.

120.

66
Davies TF. Cocultures of human thyroid monolayer cells
and autologous T cells: Impact of HLA class II antigen

expression. J Clin Endocrinol Metab 61:418-422, 1985.

Lloyd RV, Johnson TL, Blaivas M, et al. Detection of HLA-
DR antigens in paraffin embedded thyroid epithelial cells
with a monoclonal antibody. Am J Pathol, 120:106-111,

1985.

Todd I, Pujol-Borrell R, Londei M, et al. Inappropriate
HLA class II expression on epithelial cells:
Consolidation and progress. In: Drexhage HA.(ed) The
Thyroid and Autoimmunity. Amsterdam, Elsevier Science

Pub. BV (Biomed Div), 1986, pp 127-138.

Todd I, Pujol-Borrell R, Hammond LJ, et al. Interferon
gamma induces HLA-DR expression by thyroid epithelium.

Clin Exp Immunol 61:265-273, 1985.

Bottazzo GF, Pujol-Borrell R, Hanafusa T, et al. Role of
aberrant HLA-DR expression and antigen presentation in
induction of endocrine autoimmunity. Lancet 2:1115-1119,

1983.

121.

122.

123.

124.

125.

67
Aguayo J, Iitaka M, Row W, et al. Studies of HLA-DR
expression on cultured human thyrocytes; Effect of
antithyroid. drugs and other agents on interferon-y-
induced HLA-DR expression. J Clin Endocrinol Metab

66:903-908, 1988.

Iwatani Y, Gerstein HC, Iitaka M, et al. Thyrocyte HLA-DR
expression and interferon-7 production in‘autoimmune
thyroid disease. J Clin Endocrinol Metab 63:695-708,

1986.

Aoki N, DeGroot LJ. Lymphocyte blastogenic response to
human thyroglobulin in Graves' disease, Hashimoto's
thyroiditis and :metastatic thyroid cancer. Clin Exp

Immunol 38:523-530, 1979.

Okita N, Row' VV, Volpe R. Suppressor T-lymphocytes
deficiency in Graves' disease and Hashimoto's

thyroiditis. J Clin Endocrinol Metab 52:528-533, 1981.

Del Prete GF, Maggi E, Mariotti S, et al. Cytolytic T
lymphocytes with natural killer activity' in ‘thyroid
infiltrate of patients with Hashimoto's thyroiditis:
Analysis at clonal level. J Clin Endocrinol Metab 62:52-

57, 1986.

126.

127.

128.

129.

130.

131.

68
McLachlan SM, Wee SL, McGregor AM, et al. In vitro
studies on the control of thyroid autoantibody synthesis.

J Clin Lab Immunol 3:15-19, 1980.

Beall GN , Kruger SR . Production of human
antithyroglobulin in vitro: II. Regulation by T cells.

Clin Immunol Immunopathol 16:498-503, 1980.

Beall GN , Kruger SR . Production of human
antithyroglobulin in vitro: III The role of
immunolglobulins bearing cells. Clin Exp Immunol 45:56-

63, 1981.

Weetman AP, McGregor AM, Lazarus JH, et al. Thyroid
antibodies are produced by thyroid-derived lymphocytes.

Clin Exp Immunol 48:196-200, 1982.

Milne KA, Hayes HM Jr. Epidemiologic features of canine

hypothyroidism. Cornell Vet 71:3-14, 1981.

Gosselin SJ, Capen CC, Martin SL, et al. Autoimmune
lymphocytic thyroiditis in dogs. Vet Immunol Immunopath

3: 185-201, 1982 .

132.

133.

134.

135.

136.

137.

138.

69
Gosselin SJ, Capen CC, Martin SL. Histologic and
ultrastructural evaluation of thyroid lesions associated

with hypothyroidism in dogs. Vet Pathol 18:299-309, 1981.

Lucke VM, Gaskell CJ, Wolton PR. Thyroid pathology in

canine hypothyroidism. J Comp Pathol 93:415-421, 1983.

Conaway DH, Padgett GA, Bunton TE, et al. Clinical and
histological features of primary, progressive, familial
thyroiditis in a colocy of Borzoi dogs. Vet Pathol

22:439-446, 1985.

Conaway DH, Padgett GA, Nachreiner RF. The familial
occurrence of lymphocytic thyroiditis in Borzoi dogs. Am

J Med Gen, 22:409-414, 1985.

Tucker WE. Thyroiditis in a group of laboratory dogs. Am

J Clin Pathol 92:681-690, 1978.

Musser EM, Graham WR. Familial occurrence of thyroiditis

in purebred beagles. Lab Anim Care 18:58-68, 1968.

Haines DM, Lording PM, Penhale WJ. Survey of
thyroglobulin autoantibodies in dogs. Am J ‘Vet Res

45:1493-1497, 1984.

139.

140.

141.

142.

143.

144.

70
Beale KM, Halliwell Res, Chen CL. Prevalence of
antithyroglobulin antibodies detected by enzyme-linked
immunosorbent assay of canine serum. J Am Vet Med Assoc

196:745-748, 1990.

Bierwaltes WH, Nishiyama RH. Dog thyroiditis: occurance
and similarities to Hashimoto's struma. Endocrinology

83:501-508, 1968.

Gosselin SJ, Capen CC, Martin SL, et al. Biochemical and
immunological investigation on hypothyroidism in dogs.

Can J Comp Med 44:148-168, 1980.

Haines DM, Lording PM, Tenhale WJ. The detection of
canine autoantibodies to thyroid antigens by enzyme-
linked immunosorbent assay, hemagglutination and indirect

immunofluroescence. Can J Comp Med 48:262-267, 1984.

Mizjewski GJ, Baron J, Poissant G. Immunolgic
investigations of naturally occurring canine thyroiditis.

J Immunol 107:1152-1160, 1971.

Vollset I, Larsen HJ, Knaevelsrud T. Passive
hemagglutination for detection of autoantibodies against

thyroglobulin in dogs. Acta Vet Scand 28:117-119, 1987.

145.

146.

147.

148 .

149.

150.

71
Chastain CB, Young DW, Kemppainen RJ. Anti-
triiodothyronine antibodies associated with
hypothyroidism and lymphocytic thyroiditis in a dog.

JAVMA 194:531-534, 1989.

Young DW, Kemppainen RJ, Sartin JL. Characterization of
canine ‘triiodothronine (TQ autoantibodies and their
effect on total'P31n canine serum. Proc Soc Exp Biol Med

188:219-228, 1988.

Young DW, Sartin JL, Kemmpainen RJ. Abnormal canine
triiodothyronine-binding factor characterized as a
possible triiodothyronine autoantibody. Am J Vet Res

46:1346-1350, 1985.

Volpe R. Autoimmune thyroid disease: a perspective. Mol

Biol Med 3:25-51, 1986.

Premachandra BN, Blumenthal HT. Abnormal binding' of
thyroid hormone in sera from patients with Hashimoto's

disease. J Clin Endocrinol Metab 27:931-936, 1967.

Ikekubo K, Konishi J, Nakajima K, et al. A case report of
Hashimoto's disease tans: anti-thyroxine autoantibody.

Foia Endocrinol Jpn 52:1020-1032, 1976.

72
151. Balfour BM, Doniach D, Roitt IM, et al. Fluorescent
antibody studies in human thyroiditis: autoantibodies to
an antigen of the thyroid colloid destinct from

thyroglobulin. Br J Exp Pathol XLII:307-315, 1961.

73

Reprinted lrom the AMERICAN JOLRNAL OF "v’El tRlNARY RESEARCH. Vol 53. No 4 Pages 449-453
‘Amencan Verena»,- Meoxal Association. 1992 All Right! Reserved.

Prevalence of autoantibodies to thyroglobulin, thyroxine,
or triiodothyronine and relationship of autoantibodies
and serum concentrations of iodothyronines in dogs

Eileen L. Thacker, DWI; Kent R. Refsal, DVM, PhD; Robert W. Bull, DVM

SUMMARY

Assays were developed to detect and measure auto-
antibodies (AA) to thyroglobulin (Tg) and to the thyroid
hormones, thyroxine ('13) and triiodothyronine (T3). An
ELISA to detect AA to Tg was developed, using purified
canine Tg as the antigen and goat anti-canine IgG con-
jugated with alkaline phosphatase as the second anti-
body. A highly charged agarose electrophoresis assay was
used for determination of AA to '1‘, and T3. Sera from dogs
(I: - 119) with clinical signs consistent with hypothy-
roidism were tested for AA to 'I‘g, '13, and T3. Autoanti-
bodies to at least 1 of the 3 thyroid antigens were detected
in 58 of the 119 (48.7%) sera tested. Autoantibodies to Tg
were detected more frequently in samples with low serum
concentrations of thyroid hormones than in samples with
normal concentrations. The presence of AA to 13, T3, or
both was not significantly associated with low thyroid
hormone concentrations, but this lack of association may
have been attributable to binding of AA in the measure-
ment of thyroid hormones by radioimmunoassay.

 

Autoimmune thyroiditis in dogs is an endocrine disease
characterized by lymphocytic infiltration of the thyroid
gland.1 Although the pathogenesis is unknown, cellular
and humoral immune mechanisms appear to be
involved.” Cell-mediated immune mechanisms found in
human beings with autoimmune lymphocytic thyroiditis
include the actions of cytotoxic T cells or the cytotoxic
effects of lymphokines released by activated T cells re-
sponding to thyroid antigens.“5 The role of humoral im-
munity is suggested by the detection of circulating
antibodies to various thyroid antigens. Additionally, im-
mune complexes of thyroglobulin and antithyroglobulin
antibodies have been found along the thyroid follicular
basement membranes of human beings with autoimmune
lymphocytic thyroiditisf‘fl In dogs and human beings.
clinical signs of hypothyroidism develop as the thyroid
gland is destroyed, either by lymphocytic invasion of the
thyroid follicular epithelium or by atrophy of thyroid fol-
licles.

Hypothyroidism in dogs is routinely diagnosed by mea-

Received for publication Sept 12. 1990.

From the Departments of Pathology l'fhackcr. Hull) and Small Animal Clinical
Sciences «Refsall. Michigan State University, East Lansing. Ml 45521

Am J Vet Res, Vol 53, No. 4, April l99'2

suring serum concentrations of thyroid hormones, with
or without prior administration of thyroid-stimulating
hormone (TSH).9-‘° However, concentrations of thyroid
hormones frequently are not affected until the later stages
of thyroid disease, and. therefore, are of limited value in
the early diagnosis of lymphocytic thyroiditis. Addition-
ally, other physiologic factors. such as systemic disease
and drug treatment, may lower hormone concentrations
in dogs without primary thyroid disease.

Autoantibodies (AA) directed against the following thy.
roid antigens have been found to be important in the
diagnosis of ‘lymphocytic thyroiditis in human beings:
thyroglobulin (Tg); a microsomal antigen identified as
thyroid peroxidase; thyroid hormones, thyroxine (’13) and
triiodothyronine (T3); and a colloid antigen.”"" Studies
have identified AA to Tg, T“ '13, microsomal antigen. and
colloid antigen in dogs; however, the association between
various AA and clinical disease varied among studies.“‘"
Improving the assays used in the detection of these AA may
aid in early diagnosis of autoimmune thyroid disease in
dogs. The objectives of the study reported here were to de-
velop sensitive and specific assays to detect AA to Tg, T,,
and T, in dogs. to assess the prevalence of these AA in dogs
having clinical signs consistent with hypothyroidism, and
to determine the relationship between AA and serum con.
centrations of thyroid hormones.

Materials and Methods

Samples—Negative reference sera usd in the ELISA for
detection of AA to Tg were obtained from 13 dogs with
normal serum concentrations of thyroid hormones. as de-
termined by radioimmunoassay (RIA); dogs were euthan-
atized, and microscopic evidence of lymphocytic thyroiditis
was not found on histologic examination of thyroid glands.
Sera also were obtained from 2 dogs given repeated in-
jections of purified canine Tg in adjuvant‘; these samples
were designated as positive reference sera in the ELISA
for Tg AA. In the agarose gel electrophoresis assay for
detecting T, and '1‘, AA. serum from a dog with high '1‘,
concentrations and low 13 concentrations on RIA, consis-
tent with AA to T3 and '13, was designated as positive
reference serum. Autoantibodies to T, and T3 interfere
with measurement of hormone concentrations by RIA by
binding to the radioactive ligand)”:

Sera designated as test samples (n = 119) in this study

' RlBl. Hamilton. Mont.

1.
I-

74

were originally submitted for determination of thyroid
hormone concentrations and were obtained from dogs with
reported clinical signs of alopecia, lethargy. and obesity.
Dogs with pruritus, or dogs that were given exogenous
thyroid hormones or glucocorticoids. were excluded.

The ELISA methodology-Purified canine Tg,” diluted
in phosphate-buffered saline solution (PBSS. pH 7.4) was
used as the antigen in the ELISA for detection of AA to Tg.
Fifty microliters of this solution (10 ug/ml) was added to
polystyrene microtitration trays‘ and incubated for 16
hours at 4 C. After the antigen was fixed with 50 ul of
0.5% glutaraldehyde. the wells were emptied and washed
3 times with PBSS containing 4% bovine serum albumin
(BSA) and 0.05% Tween 20 (BSA-Tween in 9855). A titra-
tion curve of serum dilutions from 1:20 to 1:1,280 was
constructed from triplicate testing of the positive and
negative reference sera. Test sera and positive and neg-
ative reference sera were diluted 1:50, 1:100, and 1:200
in 0.5% BSA in P885. and 50 ul of each were added to wells
and incubated at 25 C for 2 hours. All samples were run
in duplicate. Fifty microliters of 0.11% diethanolamine so-
lution (pH 9.8) also was added to each well, and incubated
at 25 C for 30 minutes. The wells were emptied and washed
3 times with BSA-Tween in 9358. To each well, 50 ul of
goat anti-canine IgG conjugated with alkaline phospha-
tase,‘ diluted 1:100 in 0.5% BSA in 9333 was added, and
wells were incubated at 37 C for 1 hour. The wells were
washed twice with BSA-Tween in 9838 and then washed
with double-distilled water. To each well, 50 ul of the
enzyme substrate, p-nitrophenyl phosphate (1 mg/ml) in
1.0M diethanolamine buffer solution (pH 9.0). was added,
and wells were incubated for 20 minutes at 37 C. The
enzyme reaction was stopped by the addition of 50 ul of
1.0N NaOH, and the absorbance was read on a spectro-
photometer at 405 nm.

For 16 hours at 4 C, 500 pl of positive and negative
reference sera were incubated with 50 ul of Tg, T., or T,
in increasing concentrations (0 to 2.36 myml, 0.25 to 20
ng/ml, and 5 to 100 nyml, respectively) to determine
specificity of the ELISA.

Agarose gel electrophoresis—Autoantibodies to T. and
T, were assayed, using a highly charged agarose‘ gel slab
electrophoresis technique modified from Brown et al.”
Sera (50 1.1.1) were incubated for 16 hours with 150 uCi
ml-labeled T. and T, (750 to 1,250 uCi/ug)‘ and were
electrophoresed for 3 hours. Autoradiographs of the gels
were made to identify appropriate portions of the gels to
be excised. and the radioactivity was counted by a gamma
counter." To confirm IgG migration toward the cathode,
10 ug of purified canine IgG‘ also was electrophoresed.

To determine the effect of endogenous T. and T, on the
binding of AA and antigens, thyroid hormones were re-
moved from serum by use of a modified technique of Salt-
ata et al." To remove endogenous T. and T,, 100 pl of
positive and negative reference sera each were incubated

‘ Provided by Dr. Torn Brown. Wayne State University. Detroit, Mich.
' Corning Glass Works lnc. Corning, NY.

‘ Kirkegaard & Perry Laboratories inc, Geithersburg. Md.

' Model EL 307. Bio-tel: Laboratories. Burlington. Vt.

' Agarose High-L1,. Bio-Rad Laboratories. Richmond. Calif.

' KEN Research Products, Dupont Co. Boston. Mose.

" Model 1190. Tm Analytic. Elk Grove Village, 111.

' Cappel Laboratories. Melvern. Pa.

450

for 16 hours with 100 pl of 0.05M glycine-HCl buffer so-
lution with 2.5% dextran and 0.25% charcoal (pH 2.2).

The effect of thyroxine-binding globulin (me) on the
binding of AA to T. and T, and 12"’I-radiolabeled T. or T,
was assessed by adding 8-anilino-l-naphthalene-sulfonic
acid (ANS, 90 mg/ml) to the positive and negative refer-
ence sera. Bound iodothyronines are displaced from serum-
binding proteins including TBG, thus allowing radiola-
beled hormones to bind to the AA.

Measurement of thyroxine and triiodothyronine-Sewn:
concentrations of total T. were measured by RIA with a
solid-phase assay modified for use in dogs. Serum T, con-
centrations were determined by a charcoal-separation RIA
technique validated for use in dogs.“

Statistics—Chi-square analysis was used to assess sig-
nificant association between serum iodothyronine con-
centrations and detection of AA. A P-value < 0.05 indicated
statistical significance.

Results

Evaluation of ELISA technique—A dilution of 1: 100 pro-
vided acceptable separation of positive and negative ref-
erence sera (Fig 1). Mean absorbance of positive reference
sera at a 1:100 dilution was 0.299 (SD, 0.096), whereas
the negative sera had a mean absorbance of 0.011 (SD,
0.009). Absorbance values greater than 0.030, or more
than 2 SD greater than the mean absorbance for the neg.
ative reference sera at a dilution of 1:100, were consid-
ered positive for AA to Tg. The intra-assay coefficient of
variation (CV) for a pooled sample of negative reference
sera was 12.2%, and the interassay CV for the same sam-
ple was 8.6% (n =- 4 assays).

Specificity of the ELISA for AA to Tg was confirmed by
decrease in absorbance as the positive reference serum
was incubated with increasing amounts of Tg (Fig 2). Ab-
sorbance was not changed when the positive reference
serum was incubated with increasing concentrations of
T, and T..

 

absorbance 1405 mu
0 5 1
o 4 J

034

32‘

 

o fi fi if

600 800
Serum Dilution

 

 

 

Figure 1 -The eflect of dilution (1:20 to 121.280) on absorbance in positive
(l-I) and negative (O—O) reference serum samples in the ELISA lor
thyroglobulin (1’ 9). Each pomt represents a mean value 01 8 assays. Serum
Ollu‘ilons of 1:100 (serum dilution - 100 on plot) were adequate to dlflef‘
eniiate sera with autoantibodies (M) lrom negative sera.

Am J Vet Res, Vol 53, No. 4, April 1992

 

 

 

 

Aggavsafice (.405 ”ml
’3 4 1
O
O 3 ‘1
O 2 d
0 1 ‘
o -A_+ A L ‘
O 50 100 150 200
. Tg Concentration loo/WI

 

 

Figure Z-Eliect of Increasing concentrations of T9 an absorbance in pos-
itive (I-I) and negative (O—O) relerence serum samples in the eusA
for AA to Tg. Each point is the mean of triplicate determinations. Spealiciry
oi the ELISA is demonstrated by the decrease in absorbance with the ad-
dition of increasing amounts of Tg.

w. ‘ -

.0-0— .

 

‘s ‘ ' .

b l l l l l
Figure S-Auioradlographs of agarose gels lrorn sera incubated with '"l-
Iabeled therxine (7.; a) and mHabeled triiodothyronine (13; 0). Origin is
large arrow: the small arrows indicate sera with evidence oi AA. In Figure
3a. lanes 1 and 9 have 79.4 and 30.1%. respectively. of '“l-labeled T.
binding in the immunoglobulin region. In Figure 38. values 101' "’l-labeled
1', binding ranged irom 28.9 to 86.1% Ior lanes 1 to 4 and 9.

Evaluation of agarose gel electrophoresis technique—
Migration of IgG toward the cathode was confirmed by
electrophoresis of purified canine IgG and canine sera in
the highly charged agarose. Purified canine IgG migrated
only toward the cathode. whereas positive and negative
reference sera migrated toward the anode and the cath~
ode, depending on the polarity of serum proteins.

'Am J Vet Res, Vol 53, No. 4, April 1992

75

Negative reference serum samples had a mean of 1.67%
(SD, 1.28) of radiolabeled '1‘. binding in the immunoglob-
ulin region. and 0.94% (SD, 1.11) of radiolabeled T, in the
immunoglobulin region. The mean value of radiolabeled
T. in the immunoglobulin region for the positive refer-
ence sera was 74.335% (SD, 1.63); for T,, the mean value
was 19.22% (SD, 2.47). The intra-assay CV for negative
reference sera was 8.52% for T. and 6.32% for T,. Values
> 4%, which was approximately 2 SD greater than the
mean percentage for negative reference sera, were con-
sidered positive for AA to T. and T, (Fig 3).

Endogenous T. and T, had no apparent effect on de-
tection of AA to T. or T,, because removal of the hormones
with charcoal dextran solution from reference sera had
no effect on the percentages of radiolabeled hormones mi-
grating to the immunoglobulin region. Addition of ANS to
these sera resulted in displacement of radiolabeled thy-
roid hormones from no to albumin; however, effects on
the binding of hormones in the immunoglobulin region
were not found.

Prevalence of AA to Tg, T4. and T3—Of the 119 test
samples, 58 (48.7%) had AA to at least 1 of the 3 antigens
(Table 1). Autoantibodies to Tg, alone or with AA to T.,
T,, or both, were found in 50 samples (42%). Autoanti-
bodies to T. or T, were found in 31 (26.1%) and 39 (32.8%)
of the samples, respectively. Of the 58 samples that had
positive results, 19 (32.7%) had AA to all 3 antigens. Sera
with AA to Tg and T, were found most frequently (25.9%
of samples with positive results) whereas sera with AA to
only '1‘, were found least frequently (1.7%).

Association of serum hormone concentrations and AA to
Tg, T., and T,-Concentrations of T. in the sera tested
ranged from 4.7 to 150 ngml (normal, 17 to 40 ngml).
whereas T, values were 0 to 1.65 ng/ml (normal, 0.75 to 2
ng/ml). Serum concentrations of T. or T, were within the
normal range in 67 (56.3%) and 40 (33.6%) of the test sera,
respectively. Six of the 119 samples had abnormally high
or low thyroid hormone concentrations, consistent with in-
terference with the RIA from AA to thyroid hormones.

Autoantibodies to Tg, alone or with AA to T., T,, or
both, were detected more frequently (P < 0.05) in samples
with low serum concentrations of one or both hormones
(Fig 4; Table 1). When AA to only thyroid hormones were

Table 1—Oistnbution of AA to thyroid antigens in serum
samples Irom 119 dogs with clinical signs oi hypothy-
roidism. Fifty-eight (48.7%) of the samples tested had
AA to Tg, T., or T,.

 

 

 

Positive Positive
No. of results results
positive 1% of all 1% of samples
Antigen results samples) havmg AA)
Tg 11 9.2‘ 19.0
Tg and T. 5 4.2' 8 6
Tg and T, 15 12.6' 25.9
T. J 2.5 5.2
1', i o a 1.7
T. and T, d 3.4 6.9
All 19 15 9' 32.3
‘ Significantly (P < 0 05) associated with low serum
T. or T, concentrations.
AA - autoantibodies; Tg - thyroglobulin. T. - thy-
roune: T, - triiodothyronine.

 

 

 

451

76

 

 

N
O

 

I DAA to T9 IAAto r4 EAAtora

 

d
9
1

to r~

Number of samples with AA

    

 

 

 

 

 

 

' ,
{A

 

 

 

 

 

0 6 . .
Borderline 1’4 Normal to
Cone. Cone.
l0- 1 5 nglmll ”5.20 nglml) l) 20 nglrnll
40

 

[EAAtng .AAto‘N .MtoTSJ

 

Number of samples with AA

 

 

 

 

 

 

 

Normal 1'3
Cone.
l)! w

Iorderlne 1’3
Cone.
(0.19.1.0 nglmn

 

 

 

(0.0.75 rig/mil

 

Figure A-Distributlon or AA to Tg, T.. and T, in samples with low to normal
serum 1'. (top) and T, (bottom) concentrations. Autoantibodies to Tg, with
or without AA to T., T,, or both. were more lrequently detected (P < 0.05)
in samples with low serum 1’. or 1’, concentrations.

evident. significant association was not found between
serum iodothyronine concentrations and detection of AA.

Discussion

Autoantibodies to at least 1 of 3 thyroid antigens were
detected in 48.7% of the sera tested in this study. Auto-
antibodies to Tg were detected in 42% of 119 samples;
similar percentages were found in previous studies of AA
to Tg in dogs.“-”'“

Autoantibodies to T., T,, or both, with or without AA
to Tg, were found in 39.5% of the samples. The agarose
gel electrophoresis technique provided a sensitive and re-

452

producible method for measuring AA to T, and T.. There
was a wide range of binding in the immunoglobulin re-
gion in samples that had AA. whereas in samples with
negative results, the amount of radioactive binding was
low. The use of 4% binding of the radiolabeled hormone
as the cut-off point may be conservative, especially for AA
to T,. because the binding of T, for negative reference
sera was < 1%. In an earlier study, AA to T, were found
in only 0.2% of canine sera. but AA detection was on the
basis of marked alterations in the results of RIA of thyroid
hormones, not on the direct demonstration of AA to T,.2°
In several studies of human beings with autoimmune
lymphocytic thyroiditis, the prevalence of AA to T. and
T, was found to be dependent on method of detection.
ranging from 40% (6 of 15) detected by electrophoresis
techniques to 2.6% (1 of 38) from a modified RIA
technique?“

Because T. and T, are haptens. and alone are unable
to induce antibody production, Tg has been suggested as
the autoantigen for formation of AA to T. and T,.“ Thy-
roglobulin contains large quantities of T. and T, that could
provide antigenic sites for development of AA of T. and
T,. In this study, 7 samples had AA to T., T,, or both. but
not to Tg, suggesting that Tg may not be the only im-
munogen for the production of AA to T. and T,. Similarly,
human beings with AA to T., T,, or both. but not to Tg,
have been identified, suggesting that a serum protein,
other than Tg, might bind with the hormones in certain
pathologic conditions and result in production of AA that
bind to thyroid hormones?”

The electrophoretic heterogeneity of AA to T, and T.
may be attributable to variation in the population of AA.
The variation in percentage and migration pattern would
suggest that the antibodies are polyclonal, and not mon~
oclonal, which may be important in the study of the path-
ogenesis of lymphocytic thyroiditis.” Production of
polyclonal AA is suggestive of lymphocytes responding to
multiple autoantigens within the thyroid, whereas pro-
duction of monoclonal AA is suggestive of either a B-cell
malignancy or an easily induced immune response.

Approximately half (56.3% for T.. 33.3% for T,) of the
serum samples tested in this study had normal thyroid
hormone concentrations by RIA. A significant association
was found between serum thyroid hormone concentra-
tions and the frequency and type of AA. Detection of AA
to Tg, alone or with AA to T., T,, or bath, was significantly
associated with low thyroid hormone concentrations. De-
tection of AA to one or both thyroid hormones without AA
to Tg was not associated with thyroid hormone concen-
trations; however, discrepancy between measured serum
thyroid hormone concentrations and actual concentra-
tions must be considered when the sample contains AA to
T. and T,. In some serum samples with measured T. con-
centrations in the normal range. actual values may have
been low, because of AA and RIA antibodies competitively
binding to T. in serum.

Previous studies have shown correlation between AA
to thyroid antigens and lymphocytic thyroiditis in
dogs.“-“~"-” We suggest that assays that detect these
AA, combined with measurement of thyroid hormone con-
centrations, may be useful in the diagnosis of thyroid dis-
ease in dogs.

Am J Vet Res, Vol 53, No. 4, April 1992

77

References

l. Pinchera A. Fenzi G. Chiovato L. et al. Autoimmune hypothy~
roidism. In: 0?!th HA. ed. The thyroid and autoimmunity. Amster~
dam: Elsevier Science Publications. 1986;73—80.

2. McLachlan SM. Jansson R, Whitehead P. et al. Hypothyroidism
and autoimmune thyroid destruction: a role for thyroid autoantibodies?
In: Drexhage HA ed. The thyroid and autoimmunity. Amsterdam: EI-
sevier Science Publications. 1986;99-105.

3. Volpe It. The role of autoimmunity in hypoendocrine and hyper-
endeocrine function. Ann Intern Med 1977;87:86-99.

4. Schleusener I‘i. Bogner U. Cytotoxic mechanism in autoimmune
thyroiditis. In: Walfish PG, Wall JR. Volpe R. eds. Autoimmunity and
the thyroid. Toronto: Academic Press Inc. 1985;95—105.

5. Kidd A. Okita N. Row W, et a1. Immunologic aspects of Graves
and Hashimoto’s disease. Metabolism 1980;29:80-99.

6. Kalderon AB. Bogaars HA. Immune complex deposits in Graves
disease and Hashimoto's thyroiditis. Am J Med 1977;63:729—734.

7. Pinchera A. Fenzi G, Bartalena 1., et al. Antigen-antibody sys-
tems involved in thyroid autoimmunity. In: Pinchera A. Doniach D,
Fenzi G. et al. eds. Autoimmune aspects of endocrine disease. Proceed-
ings of the Serene Symposia. London: Academic Press Inc. 1980;57:72.

8. Gosselin SJ. Capen CC, Martin SL. Histologic and ultrastructural
evaluation of thyroid lesions associated with hypothyroidism in dogs.
Vet Pathol 1981;18:299-309.

9. Belshaw B, Rynberk A. In: Kirk R. ed. Current veterinary therapy.
VII. Philadelphia: WB Saunders, 1979;994—998.

10. Kallfels P. Observations on thyroid gland function in dogs: re-
sponse to thyrotropin and thyroidectomy and detennination of thyrox-
ine secretion rate. Am J Vet Res 1973;34:535—538.

11. Voller A. Bidwell DE, Burek CL. An enzyme-linked immunosor-
bent assay (ELISA) for antibodies to thyroglobulin. Proc Soc Exp Biol
Med 1980;63:402-405.

12. Marriotti S. Anelli J. Rut R. et a1. Comparison of serum thyroid
microsomal and thyroid peroxidase autoantibodies in thyroid disease.
J Clin Endocrinol Metab 1987;27:987-993.

13. Staeheli V, Valletton MB. Burger A. Detection of human anti-
thyroxine and anti-triiodothyronine antibodies in dill'erent thyroid con-
ditions. J Clin Endocrinol Metab 1975;11:669—675.

14. Gosselin SJ. Capen CC, Martin SL. Autoimmune lymphocytic
thyroiditis in dogs. Vet Immunol Immunopathol 1982;3:185—20 1.

15. ’Nclter WE. Thyroiditis in a group of laboratory dogs. Am J Clin
Pathol 1962;38:70—74.

Am J Vet Res, Vol 53, No. 4, April 1992

16. Chastain CB. Young DW, Kemppainen RJ. Anti-triiodothyronine
antibodies associated with hypothyroidism and lymphocytic thyroiditis
in a dog. J Am Vet Med Assoc 1989;194:531-534.

17. Rajatanavin R. Fang S. Pine 8. et al. Thyroid hormone antibodies
and Hashimoto's thyroiditis in mongrel dogs. Endocrinol 1989;124:2535-
2540.

18. Panciera DL, MacEwen BG. Atkins CE, et al. Thyroid function
tests in euthyroid dogs treated with L-thyroxine. Am J Vet Res
1990;51:22-25.

19. Saints S. Nakamura S. Miura K. Autoantibodies against thyroid
homones or iodothyronine. Ann Intern Med 1985;103:579—589.

20. Young DW. Sartin JL. Kemppainen RJ. Abnormal canine triio-
dothyronine-binding factor characterized as a possible triiodothyronine
autoantibody. Am J Vet Res 1985;46:1346-1350.

21. I-Iaines DM, Lording PM. Penhale WJ. The detection of canine
autoantibodies to thyroid antigens by enzymeolinked immunosorbent
assay, hemagglutination. and indirect immunofluorescence. Can J Camp
Med 1984;48:262-267.

22. Young DW, Kemppainen RJ. Sartin JL. Characterisation of ca-
nine triiodothyronine (T3) autoantibodies and their effect on total T3
in canine serum. Proc Soc Exp Biol Med 1988;188:219-228.

23. Brown TR. Bagchi N. Sundiclt R. Analysis of triiodothyronine
and thyroxineobinding autoantibodies in chickens susceptible to au-
toimmune thyroiditis. J Immunol 1985;134:3845-3848.

24. Premachandra BN, Blumenthal HT. Abnormal binding of thyroid
humane in sera from patients with Hashimoto's disease. J Clin En-
docrinol Metab 1967;27:931-936.

25. Ikekubo K. Konishi J, Nakajima K. et al. A case report of Hash-
imoto's disease with anti-thyroxine autoantibody. Folio Endocrinol Jpn
1976;52:1020—1032.

26. Pearce C. Byfield P, Edmonds C, et al. Autoantibodies to thyrog-
lobulin cross reacting with iodothyronines. Clin Endocrinol (Oxf)
1981; 15: 1-10.

27. Hehrmann ll Hoffken B. Ben Zur Muhlen A, et al. Anti-thyroid
hormone autoantibodies under experimental and clinical conditions. Harm
Metab Res 1977; 9: 326-332.

28. Willtin'l‘, SwansonbeckJ. I-IayeaP, etaLApassivehaemagglu-
tination (TRC) inhibitor in thyrotoaic serum. Clin Endocrinol (Oxf)
1979110: 507417.

29. Beale KM. Halliwell RB. Chen CL. Prevalence of antithyroglob-
ulin antibodies detected by e-Iinked immunosorbent assay of ca-
nine sertim. J Am Vet Med Assoc 1990;196:745—748.

453

78
GOLDEN RETRIEVER STUDY
SUMMARY

Hypothyroidism is a common endocrine disorder in dogs.
Lymphocytic thyroiditis is the most common cause of canine
hypothyroidism. The presence of autoantibodies to the thyroid
antigens; Tg,T,H and‘13have been shown to be associated with
lymphocytic thyroiditis in dogs.l While the presence of
thyroid autoantibodies have been shown to associated with
lymphocytic thyroiditis, nothing is known as to how early
these autoantibodies occur in dogs and how they relate to the
development of thyroid disease.2 A familial tendency has been
determined in Great Danes, beagles, and Borzois.‘4 Conway, et
a1 (1985) found evidence of a autosomal recessive genetic
pattern in Borzois.5

The. objective, of 'this study' was to study' a normal
population of dogs to determine the incidence of thyroid
autoantibodies according to age and clinical signs. Because
Golden Retrievers are a popular breed with a fairly high
incidence of thyroid disease, we selected this breed for our
study.6 The first population of dogs from which serum was
collected was from clinically normal show dogs. A second
population of related dogs traced from a Golden Retriever in
the first population which had autoantibodies to Tg, I}, and
13 was investigated to determine if the presence of thyroid

autoantibodies might have a familial basis.

79
MATERIALS AND METHODS

Populations and serum samples Negative reference serum used
in the ELISA and the agarose gel electrophoresis was obtained
from 13 dogs with normal serum thyroid hormone concentrations,
no evidence of autoantibodies to Tg, T,, or T3, and upon
euthanasia had no histologic evidence of thyroid disease.

Test sera (n=135) were obtained from Golden Retrievers at
the Fort Detroit Golden Retriever Specialty Dog Show. All
owners of the dogs answered questionnaires about possible
clinical signs of hypothyroidism and any prior history of
thyroid disease or diagnostic work. Any dogs which had been
administered thyroid medication were eliminated from the study
as were dogs on glucocorticoids. Sera for the second
population (n=42) of dogs was from relatives of one dog in the
original population that contained autoantibodies to Tg, T“
and T3. All dogs were included in this study, only the

primary dog was on thyroid replacement therapy.

Autoantibody and Thyroid Hormone Assays
The assay useds on all sera obtained from the Golden

Retrievers was previously described in chapter 3.

Statistics Chi-square analysis was used to assess
relationships between the ages of the dogs and the presence of
autoantibodies . Fishers ' exact test was used to evaluate

significant association between the sexes of the dogs and the

80
presence of autoantibodies. A P-value <0.05 indicated

statistical significance.

RESULTS

The results of the first population of Golden Retrievers
from the dog show are shown in Table 2. An overall percentage
of 16.3% of the dogs had evidence of autoantibodies to Tg, T4,
or T,. Autoantibodies to Tg alone were present most
frequently, making up 40.0% of the positive samples.
Autoantibodies to only ‘3, were present in 27.3% of the
positive samples with autoantibodies to'n,being seen with the
least frequency. Twenty-two (22.7%) percent of the sera had
autoantibodies to more than 1 antigen.

Of the dogs in the first population, only 2 dogs had
clinical signs suggestive of hypothyroidism as defined by mild
lethargy and slight obesity. The ages of the dogs tested
ranged from 5.5 months to 11 years. The ages of dogs with
thyroid autoantibodies ranged from 7.5 months to 8.5 years
with a mean of 2.6 i 2.3 years. Autoantibodies were found to
be present most frequently (P<0.05) in dogs one year of age or
younger. There was no difference in the frequency of the
autoantibodies based on sex.

In the second population, 14.3% of the dogs had evidence
of autoantibodies to Tg, TH cu'Tg. Again, autoantibodies to
T9 were present in the highest percentage with 50% of the dogs

with autoantibodies being positive. However, in this

81
population, autoantibodies to T3 were less common, with the
remaining’ dogs having autoantibodies to '1, and ‘Dr The
incidence of autoantibodies in this second population of dogs
is summarized in Table 3.

In the second population of dogs, the ages ranged from 9
weeks to 8 years of age with the ages of the dogs with
autoantibodies ranging from 9 months to 6 years with a mean
age of 2.3 i- 2.0 years. No differences between the 2
populations could be determined statistically.

The thyroid hormone concentrations were within the normal
range for both populations of dogs. Four dogs in the first
study and two dogs in the second study had borderline low
hormone concentrations, however these dogs did not have

thyroid autoantibodies.

DISCUSSION

The purpose of this study was twofold. The first
objective was to establish a baseline for the prevalence of
thyroid autoantibodies in a normal population of dogs from a
breed predisposed to developing hypothyroidism. The second
objective was to perform a preliminary investigation into the
prevalence of thyroid autoantibodies in a group of related
dogs. It is suprising that autoantibodies to Tg, Tg, and T3
are found so frequently in dogs with no history of thyroid
disease and normal serum thyroid concentrations. This study

suggests that the presence of autoantibodies occurs earlier in

82

thyroid disease before enough of the thyroid gland is
destroyed to result in clinical and serological evidence of
hypothyroidism. It. would 1x2 anticipated. that. the
autoantibodies would be present in the serum well before the
destruction of the thyroid would affect serum thyroid hormone
concentrations. Alternatively, thyroid autoantibodies may not
be related to the development of impaired thyroid function.

Thyroid autoantibodies were detected in 12 (36.4%) dogs
prior to the first year of age. The presence of
autoantibodies in dogs 1 year of age or younger were found to
be significant. when compared. to the general population.
However, to understand the biological significance of these
autoantibodies in relation to thyroid disease it would be
necessary to follow these dogs for several years, to determine
progressive changes in circulating thyroid hormone
concentrations and changing titers to thyroid autoantibodies.

By working with these populations of dogs in a long term
study, the pathogenesis and heritability of autoimmune thyroid
ciisease in the dog could be further studied. The role of the
DTHC in the dog, dietary iodine, environmental influences and
c>ther genetic markers could be studied to develop a better

Landerstanding of this complex autoimmune endocrine disease.

83

TABLE 2.

Distribution of autoantibodies to thyroid antigens in serum
samples from 135 Gold Retrievers. Twenty-two (16.3%) of the
samples had autoantibodies to Tg, Tg, or T3.

 

No. of Percent of Percent of
Antigen, Pos. results Pos. results .ALlsamples
T9 9 40.9 6
Tg & T3 1 4.5 0.7
All 3 4 18.2 2.9
T, 2 9.1 1.5
T3 6 27.3 4 4

TABLE 3.

Distribution of autoantibodies to thyroid antigens in serum
samples from 42 related Gold Retrievers. Six (14.3%) of the
samples had autoantibodies to Tg, I}, or T,.

No. of Percent of Percent of
Anti gen Pos. results Pos. results All samples
“Pg 3 50 7.1
Tg & T3 1 16.7 2.4
T, 1 16.7 2.4

T, a T, 1 16.7 2.4

84

REFERENCES
Haines DM, Lording PM, Tenhale WJ. The detection of
canine autoantibodies to thyroid antigens by enzyme-
linked immunosorbent assay, hemagglutination and indirect

immunofluroescence. Can J Comp Med 48:262-267, 1984.

Chastain CB, Young DW, Kemppainen RJ. Anti-
triiodothyronine antibodies associated with
hypothyroidism and lymphocytic thyroiditis in a dog .

JAVMA 194:531-534, 1989.

Conaway DH, Padgett GA, Bunton TE, et al. Clinical and
histological features of primary, progressive, familial
thyroiditis in a colocy of Borzoi dogs. Vet Pathol

22:439-446, 1985.

Tucker WE. Thyroiditis in a group of laboratory dogs. Am

J Clin Pathol 92:681-690, 1978.

Conaway DH, Padgett GA, Nachreiner RF. The familial
occurrence of lymphocytic thyroiditis in Borzoi dogs. Am

J Med Gen, 22:409-414, 1985.

Cutaneous Endocrinology. In: Mueller GW, Kirk RW, Scott
DW. (eds). Small Animal Dermatology. Philadelphia, W. B.

Saunders, 1989, p 642.

85

ISOLATION OF THYROID PEROXIDASE AND INVESTIGATION OF

THYROID PEROXIDASE AUTOANTIBODIES IN THE DOG.

Eileen L. Thacker, DVM; John R. Davis;

Kent R. Refsal, DVM,PhD.;Robert W. Bull, DVM

From the Departments of Pathology (Thacker and Bull), Medicine
(Davis), Small Animal Clinical Sciences (Refsal), Michigan

State University, East Lansing, MI 48824.

86
SUMMARY

Thyroid peroxidase (TPO) was purified from canine thyroid
tissue using a modification of the procedure for purifying
porcine TPO. The enzyme was solubilized from the membrane
using a deoxycholate-trypsin solution followed by ammonium
sulfate precipitation and DEAE sephadex chromatography. .A
monoclonal antibody to canine thyroglobulin (Tg) coupled to a
immunoaffinity column was used to eliminate the contaminating
Tg from the TPO preparation. SDS-polyacrylamide gel
electrophoresis (SDS-PAGE) under non-reducing conditions
identified protein bands at 107 kDa and 47 kDa. SDS-PAGE
under reducing conditions eliminated the 107 kDa band which
was replaced with components at 60, 45 and 39 kDa. Using the
TPO preparation as an antigen in a Western immunoblot assay,
50 canine sera were screened for evidence of autoantibodies to
TPO. The sera used in this study were from dogs with clinical
signs of hypothyroidism and autoantibodies to thyroglobulin,
1;, or 13. No autoantibodies to TPO were found in any of the
sera. Assays were also performed using purified porcine and
human TPO and no evidence of cross-reactivity was identified.
The absence of autoantibodies to TPO in the dog suggests a
different pathogenesis for canine autoimmune thyroid disease

than that hypothesized for human lymphocytic thyroiditis.

87

INTRODUCTION

Thyroid peroxidase (TPO) is a membrane-bound,
glycosylated, hemoprotein enzyme that plays an important role
in thyroid hormone biosynthesis by catalyzing both the
iodination of thyroglobulin (Tg) and the coupling of
iodotyrosyl residues in Tg to form thyroxine 020 and
triiodothyronine (T3).l Purification procedures have been
reported for porcinez, bovine’, and human TPO.4

Sera from human patients with lymphocytic thyroiditis
(Hashimoto's disease) usually contain antibodies against a
thyroid microsomal antigen. It has been established that TPO
is very closely related to, if not identical with, the thyroid
microsomal antigen that elicits the production of the serum
microsomal autoantibodies.4

Lymphocytic thyroiditis in dogs is comparable both
histologically'and.serologically'toJHashimoto's thyroiditis in
human beings.5 Several studies in dogs have suggested the
presence of microsomal autoantibodies in canine thyroid
disease.“°7 In this study, we developed a method for the
purification of canine TPO based on the trypsin-detergent
solubilization procedure used to purify porcine TPO.8

Using the purified canine TPO and Western blot analysis,
we studied canine sera for the presence of autoantibodies to
canine TPO and its ability to cross-react with human or

porcine TPO. To eliminate contamination with Tg, we passed

88
the sample through an immunoaffinity column to which a murine

monoclonal antibody to canine Tg was adhered.

MATERIALS AND METHODS

Isolation and purification of thyroid antigens

Preparation of T9 Tg was prepared from freshly harvested,
flash frozen canine thyroid glands obtained frommdogS'utilized
in a student surgery laboratory. Due to the labile nature of
TPO, all procedures were performed on ice or in a cold room
unless otherwise noted” 'The thyroids were homogenized in 0.05
M‘Tris, 0.4mM KI, pH 7.0, followed by centrifugation at 48,000
G for 1 h. The supernate was used for the Tg preparation, the
pellet was further processed for isolation of canine TPO. The
supernate containing the Tg was centrifuged at 100,000 G for
1 h and the supernate saved. The Tg was precipitated from the
supernate with 1.52-1.76 M ammonium sulfate followed by
centrifugation at 20,000 G for 40 min. The pellet was
resuspended in phosphate buffered saline (PBS) followed by
centrifugation at 100,000 G for 40 min. The supernate was
dialyzed for 16 hours in PBS at 4 C and the solution loaded on
a Sephycryl S300 column (Pharmacia Fine Chemicals, Piscataway,
NJ) and eluted with PBS. The first large peak measured at A

280 was collected and contained the purified Tg.

Isolation and purification of thyroid peroxidase The pellet

remaining after the first 48,000 G centrifugation was

89
homogenized with 0.5mM KI in PBS, pH 7.0. Deoxycholate (1%)
was added and the suspension was stirred 12 h at 4 C. The
solution was treated with 0.013% trypsin, 0.15 M KCl, 0.3%
Triton X-100 and stirred at 25 C for 30 min followed by
centrifugation at 100,00 G for 60 min. The pellet was
discarded and 50% ammonium sulfate was added to precipitate
the protein. Following a 17,000 G centrifugation for 40 min,
the soft pellet floating on the top of the tube was retained
and re-suspended in a 0.15M KCL Tris HCl, 0.1mM KI solution.
This was than centrifuged at 100,000 G for 1 h and the
supernate retained. The supernate was dialyzed in 0.025 M
Tris-HCl, 0.4 mM KI, pH 7.0 for 16 h. This was then eluted on
a DEAE sephadex A-50 column ixxaa step-wise gradient using
0.05, 0.2, and 0.25M KCl. The TPO enzyme activity was
collected in a broad peak at 0.2 M KCL Tris HCL of the
gradient. The peak fractions were pooled and concentrated in
an Amicon ultrafiltration cell. The material from the column
was than dialyzed against PBS and loaded on a immuno-affinity
column with the Tg murine monoclonal antibody and eluted with
PBS, 0.5mM KI, pH 7.0 (see below). The TPO activity was in
the fractions in the first peak associated with the flow
through of the immunoaffinity column. Protein concentrations
were determined by the method of Lowry et al. using bovine
serum albumin (BSA) as a standard.9 The bound thyroglobulin
was than eluted from the immunoaffinity column with PBS, pH

2.9.

90
other antigens iPurified.human and porcine TPO were generously
supplied by Dr. A Taurog, Department of Pharmacology,

University of Texas Health Science Center, Dallas, Texas.

Peroxidase Assay During the purification procedure, the TPO
enzyme activity was determined by a modified iodide oxidation
assay as described by Nakashima, et a1 (1978).‘0 Briefly, the
assay was performed as follows: 2.0 ml of 0.012 M KI in PBS,
pH 7.0 were placed in 2 cuvettes. To these were added 50 pl
of 5% glucose and 50 p1 of sample. The volume in each cuvette
was adjusted to 2.1 ml with PBS, pH 7.0 and the cuvettes
placed in a Perkin Elmer Lambda 2 double beam
spectrophotometer with attached recorder (Okidata, Microline
320). The reaction was initiated in the sample cuvette at
room temperature by the addition of 10 pl of 500 pg/ml glucose
oxidase and AAu,wes followed for 5 min on the recorder. The
AAm/min was determined from the linear portion of the chart
recording. A straight line was drawn through the points,
including the origin, and AAm/min per pl of sample was
calculated from the slope. A unit of activity is defined as
AA353/min = 1.0.

In addition, the samples was assayed for the ability of
the TPO to catalyze the oxidation of guaiacol to
tetraguaiacol.” This assay was also performed in the double
beam spectrophotometer. Both the sample and the reference

cuvette contained 2.0 ml of 66mM guaiacol in PBS, pH 7.0 with

91
50 pl of sample. The reaction was initiated by the addition of
10 pl of 66 mM Hg% and AAm,after 60 sec was determined from
the chart recording. A unit of activity is defined as AAno==

1.0 in 60 sec.

Serum Samples Negative reference sera used in the immunoblots
were obtained from 13 dogs with no evidence of autoantibodies
to Tg, T3, or T3, and normal serum concentrations of thyroid
hormones, as determined by radioimmunoassay (RIA); the dogs
were euthanatized, and no evidence of lymphocytic thyroiditis
was found on microscopic examination of thyroid glands. The
fifty canine sera screened in the Western blot technique were
obtained from Dr..Jean Dodds, Profession Animal Laboratory and
from the Endocrine Laboratory at the Animal Health Diagnostic
Laboratory, Michigan State University. All canine sera had
autoantibodies to at least one of the three thyroid
autoantibodies; Tg, T5, or T3. The human serum samples were
obtained from 2 Michigan State University Clinical Center
patients with Hashimoto's thyroiditis based on commonly
accepted clinical and laboratory criteria. One patient had a
high microsomal antibody titer and no evidence of Tg
autoantibodies while the other sample had high titers to both

Tg and the microsomal antigen.

Thyroglobulin monoclonal hybridoma production BALB/c mice

were immunized ip with 0.2 mg purified canine Tg in RIBI

92

adjuvant (RIBI, Hamilton, MT) at 2 week intervals for 8 weeks.
The mice were then bled and the serum assayed for anti-Tg
antibodies. The mice received a final immunization 3 days
prior to cell fusion. Splenic lymphocytes from the immunized
mice were fused with P3-X63-Ag8 (CRL 1580) murine myeloma
cells using Hyclone Laboratories modification of Fazekas De
St. Groth protocol.”” Hybridoma supernates were tested for
anti-Tg antibodies using the ELISA assay beginning 14 days
post-fusion. The isotype of the positive monoclonal
antibodies (mAb), as determined by the Sigma ImmunoType Kit
monoclonal antibody test, was IgGZB.

Ascites fluid containing mAb to Tg was obtained from the
peritoneal cavity of mice that had been inoculated ip with
hybridoma cells. The mAb was purified from the ascites fluid
using a protein A-Sepharose CL-4B column with PBS at pH 2.9 to
elute off the mAb.

The Tg mAb were concentrated in an Amicon ultrafiltration
cell and coupled to Affi-Gel Hz hydrazide gel (Bio-Rad,
Richmond, CA) according to manufacturers instructions.
Briefly, 0.96 ml of mAb (A280 of .983), 96 pl Na perodate (25
mg in 1.2 ml), and 96 pl 0.5 M Na acetate, pH 5.5 were
combined and rotated for 1 h at 25 C. The solution was
desalted over a Sephadex G 25 column equilibrated with 50 mM
Na acetate, 50 mM NaCl, pH 5.5 and fractions 4-9 from a steep
peak as detected at A280 were retained. The fractions were

concentrated using a Amicon ultrafiltration cell and added to

93
Affi-Gel Hz hydrazide gel which had been washed with a
coupling buffer consisting of 50 mM Na acetate and 50 mM NaCl,
pH 5.5, 5 times. The Tg mAb and Affi-Gel Hz were rotated 12
hours at 4 C prior to packing the column. This column was

then used to bind and remove the Tg from the TPO preparation.

Enzyme-linked immunosorbent assay for anti-Tg The ELISA was
a modification of a procedure described previously.”

Disposable flat-bottomed polystyrene plates (Corning Glass
Works, Corning, NY) were coated with 50 p1 of canine Tg (10
pg/ml) in PBS overnight at 4 C. Alkaline phosphatase-
conjugated goat antimouse IgG, heavy and light chain, (Sigma)
was used as the second antibody. Serum from mice prior to
immunization and media from wells without growing hybridomas
were used as negative controls. Serum from the mice prior to
fusion was used as the positive control. Supernates were
considered positive if the absorbance was 2 standard

deviations above the average of the negative control.

Gel Electrophoresis and Western Blotting Polyacrylamide
electrophoresis (PAGE) under reducing and non-reducing
conditions was carried out using the procedure described by
Laemmli, using 7.5% cross-linkedgel.‘4 After electrophoresis,
the gel was either stained with Coomassie brilliant blue R-
250, or the gel contents were transferred to a nitrocellulose

(NC) membrane. The protein bands were electrophoretically

94
transferred onto NC in 25mM Tris and 200mM glycine buffer,
containing 20% methanol in a Bio-Rad Trans Blot Cell. After
transfer, NC sheets were incubated overnight in PBS, pH 7.2
containing 5% BSA to block remaining active sites on the NC
sheets. The NC sheets were washed twice with PBS with 5% BSA
and incubated for 2 h at room temperature with serum sample
diluted 1:100 in PBS containing .5% BSA and washed with PBS
with 5% BSA. Three different antibody sources were used: 50
canine sera samples positive for autoantibodies to Tgy 1;, or
$5; rabbit anti-porcine TPO antibodies; and human serum from
patients with autoantibodies to TPO as described previously.
Bound antibody in canine sera was visualized by incubation
with horse radish conjugated goat anti-dog IgG (Kirkegaard &
Perry; Gaithersburgy MD) .reacted.‘with. 4-chloro-1-naphthol
(4CIN)and th. Antibody from the porcine sample was
visualized with horseradish peroxidase conjugated rabbit anti-
porcine IgG (Sigma) reacted the same as with the dog HRP-IgG.
Human antibodies were visualized by incubation for 1 h with
[?”I]protein A and exposed to Kodak XAR-5 film (Eastman Kodak,

Rochester, NY).

RESULTS
Figure 1 shows results of SDS-polyacrylamide gel
electrophoresis under non-reducing conditions (A) and under
reducing conditions (B). Only protein staining was used as

enzyme activity is lost with treatment with SDS.

95

Under non-reducing conditions, two bands were observed
corresponding to approximately 107 kDa and 47 kDa as
determined from simultaneously added markers. The appearance
of the bands at 60, 45, and 39 kDa under reducing conditions
and the disappearance of the 107 kDa protein suggests that the
107 kDa component is composed of several fragments linked by
one or more disulfide bridges.

The enzyme activity of the purified canine TPO is
presented in Table 4. A 33 fold purificationof the, as
measured by the iodide assay, was attained from the original
crude homogenate to the final product following chromatography
with the Tg mAb. The trypsin solubilized peroxidase activity
was separated by DEAE-sephadex chromatography into fractions
with parallel activities toward guaiacol and iodide. After
chromatography on DEAE-sephadex, a reduced guaiacol to iodide
activity ratio was observed. The higher recovery of iodide
activity may be due to the inactivation of some guaiacol
active centers or the unmasking of iodide binding sites on the
TPO.

The use of the affinity-column eliminated the
contaminating Tg without decreasing the specific activity of
the TPOn The bands which retained antigenicity toward the Tg-
mAb were no longer present when immunoblotting was performed
on the gels. The homogeneity of the Tg-mAb was shown by the
fact that all of the Tg-mAb was complexed with Tg when the Tg-

mAb was mixed with an excess amount of Tg.

96

Canine sera with autoantibodies to various thyroid
antigens including Tg,fm orfm,were immunoblotted against the
purified TPO. No evidence of autoantibodies to TPO was
observed in the 50 canine sera screened. Additionally, all
evidence of the antigenicity of Tg in Tg positive sera was
eliminated indicating' the success of the affinity-column
purification step. There was no evidence of cross-reactivity
between the purified canine TPO and either anti—porcine or
anti-human TPO antibodies. Additionally, when canine serum
was immunoblotted against either porcine or human TPO, there
was no evidence of antibodies or cross-reactivity with these
perparations of TPO. The porcine and human TPO did react
positively in the Western blots using the respective species
specific positive serum.

DISCUSSION

Thyroid peroxidase was purified to near homogeneity
following solubilization with deoxycholate and trypsin using
a combination of differential centrifugation, DEAE-sephadex
chromatography and an immunoaffinity column coupled with a mAb
to canine Tg. The method used to purify canine TPO was
similar to the procedure used by Taurog, et al (1990) to
purify porcine TPO.8 ‘The resulting TPO preparation showed a
high specific activity for iodide and guaiacol oxidation.
Tryptic digestion of thyroid subcellular components releases
a diverse number of peroxidase components depending on the

conditions of the digestion. Factors affecting these can

97
include quality and purity of the trypsin, pH, temperature,
length of digestion, and the detergent and solvent treatment
of the particles.”

The molecular weight of the canine TPO determined by SDS-
polyacrylamide gel electrophoresis under non-reducing
conditions was 107 kDa with an additional band at 47 kDa.
Under reducing conditions three bands were present at 60, 45,
and 41 kDa. These results suggest that TPO is a molecule
composed of several peptide chains of unequal size. Since
dissociation of these chains in SDS occurs after reduction of
the protein, we conclude that the chains are covalently cross-
linked by disulfide bonds. These results are similar to the
molecular weight and structural findings for human and.porcine
TPO. Kotani, et al (1986) showed human TPO to be present at
100 and 107 kDa region and porcine TPO to be approximately 100
kDa on immunoblots.16 Taurog, et al (1990) however found TPO
at 90 and 60 with SDS-PAGE under reducing conditions and at
93, 36 and 24 KDa under reducing conditions.l7 However, there
is enough variation in the molecular weights to suggest
possible differences in structure which is confirmed by the
lack of cross-reactivity between canine TPO and either human
or porcine TPO. In addition, the use of trypsin to solubilize
the thyroid membranes may have eliminated or changed the
various components of the canine TPO molecule. It has been
documented that trypsin-detergent solubilization procedure

cleaves sites at the amino and carboxyl ends of TPO and an

98

internal site of both human and porcine TPO.8J7 Thus the use
of trypsin for the solubilization may have affected epitopes
within the canine TPO molecule. However, previous findings
have shown that autoantibodies to TPO are heterogenous in
nature so it is unlikely that all antigenicity of the molecule
was destroyed in the purification procedures.’8

The immunoaffinity chromatography method used in this
study for the purification of‘TPO eliminated the contaminating'
Tg. Since it has been previously shown that dogs have
autoantibodies.to Tg, the elimination of this antigen improved
our ability to investigate possible autoantibodies to canine
TP0.7'l3 The relationship of thyroid autoantibodies to Tg, Tm,
or T3 in canine serum to canine lymphocytic thyroiditis has
been previously investigated.” These studies have indicated
that there is a high incidence of autoimmune thyroiditis in
dogs that have evidence of these autoantibodies. In the
present study we used canine sera with these autoantibodies to
investigate autoantibodies to canine TPO. Previous studies
had indicated that autoantibodies to a microsomal antigen may
be present in dogs, however no confirming evidence was found

6” Previous studies which had identified

in this study.
microsomal autoantibodies in the dog were based on a
complement fixation assay using a crude thyroid extract or
indirect immunofluorescence . The presence of microsomal

autoantibodies was based on the theory that Tg autoantibodies

do not fix complement so positive results were indicative of

99
microsomal autoantibodies which have been shown to fix
complement in human sera.19 The indirect immunofluorescence
methods used were discarded as they were found to be
technically unreliable and inconsistent. This study used
purified TPO as an antigen and sensitive immunoblotting
procedures to search for autoantibodies to TPO in the dog; No
evidence of autoantibodies to TPO were found. The lack of
autoantibodies to TPO is similar to results seen in the obese
strain. of chicken and rats with lymphocytic thyroiditis
(personal communication). The absence of autoantibodies to
TPO in the dog suggests that canine lymphocytic thyroiditis
has a different pathogenesis than found in human thyroid

disease.

100

TABLE 4.

Activity of canine TPO at various stages of purification.

Purification Total Proteina
(mg)

1. Ammonium sulfate
precipitation 57.11

2. DEAE sephadex
chromatography 4.20

3. Immunoaffinity
chromatographyb 0.78

a Based on Lowry protein determination
b Canine Tg mAb coupled to Affi-Gel Hz

Specific Activity
Guaiacol I3
Oxidation Formation

0.61 0.83
2 15 1.32
19.80 4.35

101

FIGURE 5.

SDS-polyacrylamide gel electrophoresis of purified canine TPO
stained with Coomassie blue. Lane (a) gel was
electrophoresised under non-reducing conditions and lane (b)
was electrophoresised under reducing conditions.
Approximately 50 pg of protein was applied to gel.

102
REFERENCES
Taurog A. Thyroid peroxidase and thyroxine biosynthesis.

Recent Prog Horm Res 26:189-247, 1970b.

Rawitch AB, Taurog A, Chernoff SB, et al. Hog thyroid
peroxidase: Physical, chemical and catalytic properties
of the highly purified enzyme. Arch Biochem Biophys

194:244-257, 1979.

Alexander NM. Purification of bovine thyroid peroxidase.

Endocrinology 100:1610-1620, 1977.

Czarnocka B, Ruf J, Ferrand M, et al. Purification of the
human thyroid peroxidase and its identification as the
microsomal antigen involved in autoimmune thyroid

diseases. FEBS Lett 190:147-152, 1985.

Rajatanavin R, Fang SL, Pino S, et al. Thyroid hormone
antibodies and Hashimoto's thyroiditis in mongrel dogs.

Endocrinology 124:2535-2540, 1989.

Mizjewski GJ, Baron J, Poissant G. Immunolgic
investigations of naturally occurring canine thyroiditis.

J Immunol 107:1152-1160, 1971.

10.

11.

12.

103
Haines DM, Lording PM, Penhale WJ. The detection of
canine autoantibodies to thyroid antigens by enzyme-
linked immunosorbent assay, hemagglutination and indirect

immunofluorescence. Can J Comp Med 48:262-267, 1984.

Taurog A, Dorris ML, Yokoyama N, et al. Purification and
characterization of a large, tryptic fragment of human
thyroid peroxidase with high catalytic activity. Arch

Biochem Biophys 278:333-341, 1990.

Lowry OH, Rosbrough NJ, Farr AL, et al. Protein
measurement with the Folin phenol reagent. J Biol Chem

193:265-275, 1951.

Nakashima T, Taurog A. Improved assay procedures for
thyroid peroxidase; application to normal and adenomatous

human thyroid tissue. Clin Chem Acts 83:129-140, 1978.

Hosoya T, Morrison M. The isolation and purification of

thyroid peroxidase. J Biol Chem 242:2828-2836, 1967,

Fazekas De St. Groth S, Scheidegger D. Production of
monoclonal antibodies: Strategy and tactics. J Immunol

Methods 35:1-21, 1980.

13.

14.

15.

16.

17.

104
HyClone Laboratories. Cell fusion with PEG: is it

reproducible? Art to Science 2:1-5, 1983.

Thacker EL, Refsal KR, Bull RW. Prevalence of
autoantibodies to thyroglobulin, thyroxine, or
triiodothyronine and relationship of autoantibodies and
serum concentrations of iodothyronines in dogs. Am J Vet

Res 53:449-453.

Laemmli UK. Cleavage of structural proteins during
assembly of the head of bacteriophage‘np Nature 227:680-

685, 1970.

Kotani T, Umeki K, Matsunaga S, et al. Detection of
autoantibodies to thyroid peroxidase in autoimmune
thyroid diseases by micro-ELISA and immunoblotting. J Cl

Endocrinol Metab 62:928-933, 1986.

Yokoyama N, Taurog A. Porcine thyroid peroxidase:
Relationship between the native enzyme and an active,
highly purified tryptic fragment. Mol Endocrinol 2:838-

844, 1988.

18.

19.

20.

105
Hamada N, Naeduck N, Portmann L, et al. Antibodies
against denatured and reduced thyroid microsomal antigen
in autoimmune thyroid disease. J Cl Endocrinol & Metab

64:230-238, 1987.

Khoury EL, Hammond L, Bottazzo GF, et al. Presence of the
organ-specific 'microsomal' autoantigen on the surface of
human 'thyroid cells in culture: its involvement in
complement-mediated cytotoxicity. Clin Exp Immunol

198:316-328, 1981.

Chastain CB, Young DW, Kemppainen RJ. Anti-
triiodothyronine antibodies associated with
hypothyroidism and lymphocytic thyroiditis in a dog.

JAVMA 194:531-534, 1989.

106
STUDIES OF AUTOANTIBODIES TO CANINE TPO BY
ENZYME-LINKED IMMUNOSORBENT ASSAY

AND INHIBITION OF ENZYME ACTIVITY

SUMMARY

Serum autoantibodies to the microsomal antigen which has
been identified as thyroid peroxidase (TPO) are common in
human patients with autoimmune thyroid diseases.
Autoantibodies to TPO are important in the diagnosis of
autoimmune thyroid disease in human beings as greater than 90%
of these patients have autoantibodies to this thyroid antigen.

Two studies in dogs found 'microsomal' autoantibodies in
dogs 'with thyroid disease.L2 These studies used either
complement fixation and an ELISA using a crude thyroid
preparation or indirect immunofluorescence on frozen and fixed
canine thyroid glands. Because of the nature of these assays,
we decided to develop a sensitive ELISA for the detection of
autoantibodies to TPO in the dog.

Studies of TPO autoantibodies in human beings indicate
the presence of TPO autoantibodies in the serum inhibits the
enzyme activity of TPO.3 Using purified canine TPO, canine
sera were tested for TPO autoantibodies by assessing their
ability to inhibit the enzyme activity of TPO.

The objective of this study was to develop assays to
detect autoantibodies to TPO in dog. Hypothyroidism is a

commonly diagnosed endocrine disease in dogs. However, the

107
accurate diagnosis of hypothyroidism is often difficult due to
the many factors which affect thyroid hormone concentrations.
The development of assays to detect and measure autoantibodies
to 'TPO ‘will potentially aid veterinary practitioners in

diagnosing canine autoimmune thyroiditis.

MATERIAL AND METHODS
Serum Samples Negative control serum was obtained from dogs
which had normal serum thyroid hormone concentrations as
diagnosed by RIA, no detectable autoantibodies to TG, T3, or
'L,and upon euthanasia their thyroid glands had no microscopic
evidence of thyroid disease. The canine sera screened for
evidence of autoantibodies to TPO was from dogs with
autoantibodies to TG, T4, or T3 as documented by assays

performed in our laboratory.

Preparation of canine TPO The purification of TPO used in
this project was described in chapter V. Two preparations of
TPO were used in the ELISA. The first preparation consisted
of TPO which had been isolated by differential centrifugation
and DEAR-sephadex chromatography, but had not been eluted
through the immunoaffinity column bound with the Tg mAb. The
second preparation consisted of‘TPO from the first preparation
which had been eluted through the Tg immunoaffinity column
prior to using in the ELISA” The inhibition studies used only

TPO which had not been run through the immunoaffinity column

108
as Tg autoantibodies should not interfere with the TPO enzyme

activity.

Enzyme-linked immunosorbent assay (ELISA)

The ELISA was a modification of the assay used for the
measurement of Tg autoantibodies as described previously.
Briefly, 100 pl of antigen solution was added to the wells of
a microtiter plate at the following concentrations: TPO 20
pg/ml diluted in 50 mM Carbonate buffer, pH 8.5 and Tg, 10
pg/ml in PBS, pH 7.4 and incubated overnight at 4 C. Canine
serum was diluted 1:100 with PBS with 0.5% BSA and incubated
for 2 h at 25 C. Alkaline phosphatase-conjugated goat anti-
canine immunoglobulin G (heavy and light chain) (Kirkegaard &
Perry, Gaithersburg, MD) was added following washing the
plates with 5% BSA in PBS, pH 7.4 and incubated for 1 h at 25
C. The secondary antibodies were reacted with the enzyme
substrate, p-nitrophenyl phosphate (1 mg/ml) and the

absorbance was read on a spectrophotometer at 405 nm.

Inhibition of TPO catalytic activity by patient serum

Purified canine TPO with a specific activity of 21.43 U/mg
protein as determined by the iodide oxidation assay described
earlier was used in the assay. Fifty pl of TPO was incubated
with 50 pl of patient sera or negative control sera and
incubated 1 h at 25 C. For the measurement of the catalytic

activity, 2.0 ml PBS with 0.012 M KI pH 7.0 and 50 pl 5%

109

glucose to 2 cuvettes, 50 pl of the incubated sample was added
to 1 of the cuvettes. The reaction was initiated by the
addition of 10 pl of 500 pg/ml glucose oxidase and read in a
Perkin Elmer Lambda 2 double beam spectrophotometer connected
to. a Okidata Microline 320 reader. The AA353/min. was
determined from the linear portion of the chart recording. A
straight line was drawn through the points, including the
origin, and AA353/min per pl of sample was calculated from the
slope. A unit of activity is defined as AA353/min = 1.0.

In one experiment, the effect of addition of varying
amounts of serum to the iodide oxidation assay was determined.
Incubation and assay were performed as described above. For
each volume of patient serum added, results were compared with
a similarly treated control serum.

To confirm that autoantibodies to Tg‘were not interfering
with the catalytic activity of the TPO, serum was incubated

with excess Tg to absorb out all Tg autoantibodies.

SDS-polyacrylamide electrophoresis and immunoblotting This
procedure was explained in the chapter on the purification of
TPO. Purified canine TPO, Tg and trypsinized Tg were
electrophoresed on SDS-polyacrylamide gels and immunoblotted
as previously described. The proteins were transferred to a
nitrocellulose sheet.which was incubated.with negative control
canine sera, serum from 2 dogs with documented autoantibodies

to Tg, and the Tg mAb described in chapter V.

110
RESULTS

Binding of the canine serum to the canine TPO antigen
preparation occurred in 5 of the 39 canine serum samples
screened in the ELISA. However, when purified canine Tg was
incubated with the serum samples to absorb out all
autoantibodies to canine Tg, all samples became negative. To
confirm the presence of Tg contamination in the initial
purified canine TPO, the TPO was immunoblotted along with
purified canine Tg and trypsinized canine Tg. The
immunoblotting confirmed the contamination of the TPO‘with.Tg.
After purifying the TPO further using immunochromatography
with the mAb to canine Tg, all evidence of the contaminating
Tg disappeared on SDS-polyacrylamide gel electrophoresis and
immunoblotting. When this TPO was used as the antigen in the
ELISA, no positive results were evident.

Because inhibition of the catalytic activity of TPO had
been demonstrated with serum from human patients with
autoimmune thyroid disease, we repeated the assay using the
iodide oxidation assay and canine serum. Consistent
inhibition of TPO activity by patient sera in relation to

negative control sera did not occur.

DISCUSSION
No autoantibodies to canine TPO were detected using an
ELISA or an assay for the inhibition of enzyme catalytic

activity. Initially, in the ELISA using the first TPO

111

preparation, 5 serum samples appeared to have evidence of TPO
autoantibodies. IHowever, the positive results were the result
of a small amount of Tg which was contaminating the TPO
sample. Because it has been documented dogs with thyroid.
disease have autoantibodies to Tg, it was not surprising that
dogs with very high titers to the Tg appeared positive in this
ELISA“. To confirm that the positive results seen in our
assay were to Tg, the autoantibodies were absorbed with excess
Tg. 'The serum samples than became negative in the ELISA“ The
presence of Tg in the TPO preparation was confirmed using SDS-
polyacrylamide electrophoresis and immunoblotting. Because
the addition of a large amount of Tg might have an effect
other aspects of the ELISA it was necessary to eliminate the
contaminating Tg.

To eliminate the contaminating Tg, the TPO preparation
was eluted through an immunoaffinity column that contained Tg
mAb coupled to the gel. The resulting preparation had no
evidence of Tg as shown by repeating the SDS-polyacrylamide
gel electrophoresis and immunoblotting. When this purified
TPO was used as the antigen in the ELISA, no sera samples were
positive. ‘To investigate whether canine autoantibodies to‘TPO
may inhibit the catalytic activity of the enzyme, purified TPO
was incubated with serum. Because the presence of Tg
autoantibodies should not interfere with the activity of the
TPO, we used the initial TPO preparation. No sera

consistently inhibited the activity of the TPO as measured by

112

iodide oxidation” To determine that Tg autoantibodies did not
play a role in this lack of inhibition, the Tg autoantibodies
were again absorbed out with Tg. No change in results was
seen following the depletion of the Tg autoantibodies. We
would not expect to see any inhibition of TPO activity by
canine serum since no evidence of TPO autoantibodies were
found using the more sensitive ELISA.

The previous study that reported the presence of
microsomal autoantibodies in 1 dog was an ELISA used a very
crude thyroid extract as antigen.2 It is our contention that
this dog reacted to the Tg which was present in the antigen.
Thyroglobulin autoantibodies can be present in very high titer
(>24,000) and thus it is difficult to absorb out all the Tg
autoantibodies using excess Tg without diluting out the serum
sample. On the basis of these results, we concluded that the
incidence of autoantibodies to TPO was either below the
sensitivity of these assays, or very low or dogs do not

produce autoantibodies to TPO.

REFERENCES
1. Mizjewski <3], Baron .I, Poissant (3. Immunologic
investigations of naturally occurring canine thyroiditis.

J Immunol 107:1152-1160, 1971.

113
Haines DM, Lording PM, Penhale WJ. The detection of
canine autoantibodies to thyroid antigens by enzyme-
linked immunosorbent assay, hemagglutination and indirect

immunofluorescence. Can J Comp Med 48:262-267, 1984.

114
SUMMARY AND CONCLUSIONS

The objective of this study was to develop assays to
detect autoantibodies to the thyroid antigen, Tg, T“ 13, and
TPO in the dog. An ELISA was developed to detect and measure
autoantibodies to canine Tg. An agarose gel electrophoresis
assay was used for the measurement of autoantibodies to canine
1; and'nr using these assays, a study was performed on serum
from dogs (n=119) with clinical symptoms of hypothyroidism
including alopecia, lethargy and obesity. Autoantibodies to
Tg, T4, or T3 were found in 48.7 percent of the samples
measured. Autoantibodies to Tg were found most frequently
(42%) and were statistically related to low serum.Tgihormone
concentrations. In addition, autoantibodies to T4 or T3 were
found in 26.1% and 32.8% of the samples respectively. Of the
positive samples 32.7% had autoantibodies to all 3 antigens.

Using the assays to detect serum Tg, IQ, and T3
autoantibodies, a study was done on 2 populations of Golden
Retrievers. The first population (n=135) was from clinically
normal dogs. In this population, autoantibodies to one of the
3 antigens were found in 16.3% of the dog. A significantly
higher percentage of thyroid autoantibodies were present in
dogs less than 1 year of age. The serum thyroid hormone
concentrations of these dogs were normal and only 2 dogs had
slight clinical signs of lethargy or obesity. The percentage
and the presence of these autoantibodies in young dogs

indicates these assays may be helpful in early diagnosis of

115
autoimmune thyroid disease in dogs when coupled with resting
serum T4 or T3 concentrations.

The second population of Golden Retrievers consisting of
dogs related to a dog from the first population which had
autoantibodies to Tg, T4, and T3 was examined to determine if
there was a familial basis to thyroid autoantibody
development. In this population, 14.3% of the dogs had
autoantibodies to the various thyroid antigens. Again, 66.6%
of the dogs with autoantibodies were less than 1 year of age.
The fact that the percentages of autoantibodies in this
population was similar to the.general population suggests that
while heritability may be a factor, it cannot be considered
the sole initiating factor in the development of thyroid
autoantibodies. Further studies should be done to confirm
this observation.

Canine TPO was isolated and purified using a modification
of a technique developed for porcine TPO using differential
centrifugation and DEAE-sephadex chromatography. An ELISA
using semi-purified canine TPO was used to screen sera for the
presence of autoantibodies. Of the 39 canine serum assayed 5
appeared to be positive. Upon further investigation this
preparation was found to be contaminated with Tg' which
interfered with the testing for TPO autoantibodies as the Tg
autoantibodies present in the serum resulted false positive
results. To eliminate the contaminating Tg from the TPO, a

monoclonal antibody (mAb) to canine Tg was made. The Tg mAb

116
was coupled to an immunoaffinity column and the canine TPO was
eluted through the column removing the Tg.

When all Tg was removed from the TPO preparation, as
confirmed by SDS-polyacrylamide gel electrophoresis and
immunoblotting with canine sera containing autoantibodies to
TG, there was no indication of autoantibodies to TPO in the
ELISA. To confirm this assumption, a sensitive Western blot
assay was developed. Canine serum with autoantibodies to
other thyroid antigens including Tg,'n,and/or'L,were screened
using the Western blot assay for evidence of autoantibodies to
canine TPO. No evidence of autoantibodies were found in any
of the samples. Additionally, an assay to investigate the
inhibition of TPO's catalytic activity by TPO autoantibodies
using the iodide oxidation assay was inconsistent and showed
no reliable data to support the existence of TPO
autoantibodies.

On the basis of our study of thyroid antigens and
autoantibodies, it appears that Tg plays a major role in
canine autoimmune thyroid disease. In the study of Tg, T3,
and T3 autoantibodies, it was apparent that the presence of
autoantibodies to Tg are significant in relation to clinical
signs of hypothyroidism and decreased serum concentrations of
T4. Additionally, autoantibodies to T4 and T3 may be a subset
of Tg autoantibodies. Thyroid hormones are haptens and thus
alone cannot induce antibody formation. Thyroglobulin

contains many T4anuifm molecules to which autoantibodies can

117

be produced. However, autoantibodies to'l(and'E,occasionally
were present without Tg autoantibodies suggesting that the
thyroid hormones may occasionally bind to other serum.proteins
to induce antibody production. However, when thyroid hormone
autoantibodies were present without Tg autoantibodies, they
were not statistically significant in relation to serum T4
concentrations. This suggests that when they are present
alone, it may be more indicative of changed serum binding than
thyroid disease. However this should be further investigated.

Canine: TPO ‘was isolated and 'used in ‘this study ‘to
determine:if dogs had autoantibodies to this antigen. lBecause
of the frequency of autoantibodies to Tg, it was necessary to
eliminate this antigen as a source of contamination of our
TPO. Because the quantity of TPO was small, we used a Western
blots as opposed to the less sensitive ELISA procedure to
examine sera for TPO autoantibodies. When all Tg was
eliminated from.the TPO, no autoantibodies to TPO were present
in the samples examined. In addition, TPO and serum from
normal or potentially hypothyroid dogs did not cross-react
with either TPO or anti-TPO serum from pigs or human beings.
Thyroid peroxidase is an important antigen in the pathogenesis
of human thyroid disease. It is felt that the combined
expression of‘TPO and the inappropriate expression of class II
MHC antigens on thyroid epithelial cells play a role in the
initiation of autoimmune thyroid diseases in human beings.

This study, by failing to detect autoantibodies to ‘TPO

118
indicates a different pathogenesis and different initiating
factors may be present in canine thyroid disease.

Because the accurate diagnosis of hypothyroidism in dogs
is an important concern to pet owners and dog breeders alike,
it is imperative that we learn more about this disorder. The
thyroid is one of the most important endocrine organs in the
body and its function is affected by many outside influences.
Influences which alter the thyroid hormone concentrations also
make the diagnosis of hypothyroidism difficult. Additionally,
it is a disorder that tends to affect dogs in their middle
years, so many reproduce prior to the discovery of autoimmune
thyroid disease. Because the genetic heritability of
autoimmune lymphocytic thyroiditis has been‘documented.in some
breeds of dog, an accurate early diagnosis would be useful to
selectively breed dogs without incipient thyroid disease.

Continued investigation of Tg is necessary to determine
its exact role in the initiation of autoimmune thyroid
disease. Some aspects to investigate are:

1. Continue measuring the thyroid hormone concentrations

and. the thyroid autoantibodies in the young Golden

Retrievers to determine whether the presence of thyroid

autoantibodies in young, normal dogs is a diagnostic tool

for determining dogs at risk for autoimmune thyroiditis

and clinical hypothyroidism.

119
2. The role lymphocytes play in the destruction of the
thyroid gland. It needs to be determined whether thyroid
parenchymal destruction in canine lymphocytic thyroiditis

is primarily B or T lymphocyte mediated disorder.

3. If a change in the structure of Tg plays a role in the
induction of thyroid disease. In the obese strain of
chickens, which are susceptible to autoimmune lymphocytic
thyroiditis, increased iodination plays a major role in
increased Tg antigenicity. Studies could be done using
the mAb to Tg developed in this project to investigate

the structure and location of the Tg in the thyroid.

4. Investigate the Tg gene to see if there is a variation

in the sequence in normal and affected dogs.

5. Family studies, including investigation of the MHC in
dogs and the possible role it may play in the

pathogenesis of thyroid disease.

5. Investigate whether dogs have other thyroid
autoantibodies, e.g. thyroid enzyme inhibiting

autoantibodies.

At this time, the initiating event which triggers the

beginning of the destruction of the thyroid by lymphocytes is

120
unknown. Although there is a genetic predisposition, we may
find that there are other environmental and dietary factors

play a major role in its pathogenesis.

"741141114111444111