Protein expression of insuline-like
growth factor I and II, growth hormone
and growth hormone receptor in canine
cortisol-producing adrenocortical
tumors
Name:
Aafke Wilms
Student number: 3072592
Supervisors:
Dr. Sara Galac (supervisor, endocrinology)
Dr. Jan Mol (supervisor, laboratory)
Adri Slob (daily supervisor, laboratory)
Noortje van der Helm (daily supervisor, laboratory)
Contents
ABSTRACT .................................................................................................................................................... 3
INTRODUCTION............................................................................................................................................ 4
MATERIALS AND METHODS ..................................................................................................................... 6
3.1 TISSUE SAMPLES ......................................................................................................................................... 6
3.2 IMMUNOHISTOCHEMISTRY ........................................................................................................................... 6
3.3 SCORING METHOD OF IMMUNOHISTOCHEMISTRY ....................................................................................... 7
3.4 STATISTICAL METHODS ................................................................................................................................ 8
RESULTS ....................................................................................................................................................... 9
4.1 LOCATION OF STAINING ............................................................................................................................... 9
4.2 PROPORTION AND INTENSITY OF STAINING ................................................................................................. 9
DISCUSSION ............................................................................................................................................... 14
REFERENCES ............................................................................................................................................. 15
2
Abstract
Cushing’s syndrome is an important endocrinological disorder in dogs. In 15-20% of
the cases it is caused by excessive secretion of glucocorticoids by an adrenocortical
tumor (AT) that is benign or malignant. It has become clear that the growth hormoneinsulin-like growth factor pathway plays a role in the pathogenesis of ATs. In this
research we compared the expression of growth hormone (GH), the growth hormone
receptor (GHR), insulin-like growth factor I receptor (IGF-IR) and insulin-like growth
factor II receptor (IGF-IIR) in normal adrenal tissue, cortisol-producing adrenocortical
adenomas (ACAs) and carcinomas (ACCs) of dogs by immonohistochemistry (IHC).
A significant higher intensity of staining of GHR was demonstrated in the ACAs as
compared to the normal adrenal glands. However, the intensity of staining in ACAs
was also stronger than in the ACCs (although not significant). This makes the role of
GHR in tumorigenesis less likely. There were no significant differences between the
intensity or proportion of staining of GH, IGF-IR and IGF-IIR between the normal
adrenal glands and the ATs. Therefore, there also seems to be no roles for GH, IGFIR and IGF-IIR in tumorigenesis in adrenal tissue.
3
Introduction
Cushing’s syndrome or hypercortisolism is an endocrinological disorder in dogs and
occurs with a high incidence. In most of the cases (80-85%), hypercortisolism is
caused by hypersecretion of adrenocorticotropic hormone (ACTH) by a corticotroph
adenoma of the pituitary gland. This is called ACTH- or pituitary-dependent
hypercortisolism (PDH). The remaining cases are ACTH-independent. They are
caused by excessive secretion of glucocorticoids by an adrenocortical tumor (AT)
that is benign or malignant [Galac et al. 2010].
In the research that has been done the last years, it became clear that there are
several signalling pathways that are important in the pathogenesis of ATs. This
project will focus on the growth hormone-insulin-like growth factor pathway.
Growth hormone (GH) is produced in the anterior lobe of the pituitary gland, but also
in extrapituitary sites where it acts on the autocrine and paracrine level. GH mediates
its effect through binding with the growth hormone receptor (GHR). It has a direct
effect on somatic growth regulation, mostly by stimulating the hepatic insulin-like
growth factor I (IGF-I) secretion [Perry et al. 2006].
One of the extra-pituitary sites where GH is produced is canine mammary tissue. The
GH production occurs in normal, hyperplastic and neoplastic mammary tissue. The
production is stimulated by progestins and it could be mediating the progestinstimulated development of mammary tumors in dogs [Mol et al. 1995].
In human ATs and normal adrenal glands high concentrations of the progesterone
receptor (PR) have been demonstrated [De Cremoux et al. 2008]. So it could be
possible that progestins also in the adrenal gland stimulate GH production which in
turn could stimulate cell-proliferation and hyperplasia.
The insulin-like growth factor (IGF) system plays a central role in the mechanism of
transformation and tumourigenesis [Weber et al. 2000]. The two structurally related
polypeptide ligands, IGF-I and insulin-like growth factor II (IGF-II), their receptors and
binding proteins are synthesized en secreted by the adrenal glands of various
species, but the situation in dogs is not known yet [Fottner et al. 2004].
IGF-I and IGF-II act at the endocrine and para/autocrine level with multiple activities
[Boulle et al. 1998, Fottner et al. 1998]. Their main growth- and differentiationpromoting effects are mediated through binding to the IGF-I receptor (IGF-IR). The
IGF-II receptor (IGF-IIR) only binds IGF-II. It has been implicated as a tumor
suppressor gene because the clearance and inactivation of IGF-II is mediated by
IGF-IIR [Fottner et al. 2004].
The last decade various studies that used transcriptome analysis to study the gene
profile expression of human ATs have been performed [Fottner et al. 2004]. Among
all studied genes, the IGF-II turned out to be the most overexpressed gene in
adrenocortical carcinomas (ACC), in comparison with adrenocortical adenomas (ACA)
and is judged as pathognomonic for ACCs [Ragazzon et al. 2011].
Based on the studies in ATs in men, the influence of IGF-I on the adrenocortical
tumorigenesis is little. In none of the studies the overexpression of IGF-I in ATs could
be demonstrated [Fottner et al. 2004]. On contrary, a strong overexpression of IGFIR has been described in ACCs, but not in ACAs [Fottner et al. 2004]. These elevated
levels of IGF-IR in combination with a high concentration of IGF-II may be
4
representing an autocrine stimulatory loop by which it contributes to adrenocortical
tumorigenesis [Fottner et al. 2004].
As mentioned before, it seems that IGF-IIR clears and inactivates IGF-II, thereby
playing an important role in regulating cell growth. In human ACCs there is a frequent
loss of heterozygosity (LOH) at the IGF-IIR locus. This may contribute to an
increased IGF-II bioavailability. In ACAs LOH was detected in only 9%, compared to
58% in ACCs, which suggests that the changes in the IGF-IIR gene locus are a late
step in adrenocortical tumorigenesis [Fottner et al. 2004].
The goal of this project is to compare the expression of GH, GHR, IGF-IR and IGFIIR in normal adrenal tissue, cortisol-producing ACAs and cortisol-producing ACCs by
immunohistochemistry (IHC) in dogs. The outcomes may contribute to a better
understanding of the pathogenesis of ATs in dogs and to the differentiation between
malignant and benign AT’s.
5
Materials and methods
3.1 Tissue samples
The adrenal gland tissues used in this study were collected in previous years by the
Department of Clinical Sciences of Companion Animals at the Faculty of Veterinary
Medicine of the University of Utrecht. The tissues from normal adrenal glands were
obtained from dogs that were euthanized for other reasons than adrenocortical. The
tissues from ATs were from dogs with ACTH-independent hypercortisolism after
adrenalectomy. In this study, 9 ACCs, 9 ACAs and 7 normal adrenal glands were
included. Parts were fixed in 4% methanol buffered formaldehyde. The tissues were
embedded in paraffin, after 24 hours (48 hours at the most) of fixation. They were cut
into sections of 4-μm and mounted on Starfrost® adhesive microscope slides (KnittelGläser, Braunschweig, Germany). From every tissue there was one section that was
stained with hematoxylin and eosin for hisological evaluation. The remaining sections
were used for immunohistochemistry (IHC).
3.2 Immunohistochemistry
Before all the sections were stained, the protocols for immunohistochemical staining
were optimised for the different antibodies. The optimised protocols were as follows:
Insulin-like growth factor I receptor (IGF-IR)
Immunohistochemical staining was performed on paraffin-embedded adrenal tissue
sections. The sections were deparaffinized in xylene (2x5 min) and in a descending
serie of ethanol they were rehydrated (3 min 96%, 3 min 80%, 2 min 70%, 2 min
60%). The sections were washed in a 10% tris buffered saline solution (TBS), with
pH 7,6 (2x5 min), and subsequently in TBS + 0,025% Triton X-100 (TBST) (2x5 min).
A treatment with 10% normal goat serum in 1% bovine serum albumine (BSA) in
TBST was performed to block non-specific antibody binding (30 min). The sections
were incubated overnight at 4 ºC with a rabbit polyclonal IGF-IR antibody (PAB18560
Abnova), diluted 1:200 in 1% BSA in TBST. Following a wash step in TBST (2x5 min),
incubation with 0,35% H2O2 in TBS took place to block endogenous peroxidase (15
min). Again, the slides were washed in TBST (2x5 min). And then they were
incubated with evision anti-rabbit HRP (Dako, Carpinteria, CA, USA), the second
anitbody (30 min). After washing with TBS (2x5 min), the sections were incubated
with DAB-solution (Dako) for 3 minutes and then washed in Milli-Q (5 min).
Hematoxylin was applied for 5 seconds to counterstain, followed by washing in
tapwater (10 min). After dehydration by an ascending ethanol serie (2 min 60%, 2
min 70%, 3 min 80%, 2x3 min 96%) an xylene (2x5 min), the slides were mounted
with Vectamount-permanent mounting medium.
Insulin-like growth factor II receptor (IGF-IIR)
The same method as for IGF-IR, was used for immunohistochemistry for IGF-IIR.
The only difference was the duration of incubation with DAB-solution (Dako), which
was 2 minutes with IGF-IIR. The antibody used was a polyclonal rabbit IGF-IIR
antibody (PAB18393 Abnova), diluted 1:200 in 1% BSA in TBST.
6
Growth hormone receptor (GHR)
Again, the same method as previously described was used for the
immunohistochemical staining. The incubation with DAB-solution was 10 minutes. A
mouse monoclonal anti-GHR antibody (ab11380 Abcam) was used as the primary
antibody. The secondary antibody used was envision anti-mouse (Dako).
Growth hormone (GH)
As the primary antibody a rabbit polyclonal anti-GH antibody (7801H) was used, with
envision anti-rabbit (Dako) as the second antibody. The sections were incubated with
DAB-solution for 3 minutes. The method for staining was equal to those previously
described.
3.3 Scoring method of immunohistochemistry
The sections were assessed on the location of the staining in the cells, the proportion
of positive stained cells and the intensity of staining. For the normal adrenal tissues,
the glomerulosa and fasciculata were separately assessed.
The staining could be located in the cytoplasm, the cell membrane, the nucleus
and/or the nuclear membrane. The gradation used for the proportion of positive cells
throughout the slide was in the range of 0-4; no staining ,0; 1-25% cells positive, 1;
26-50% cells positive, 2; 51-75% cells positive, 3; 76-100% cells positive, 4. These
numbers were replaced by the class middle (1  12,5%, 2  37,5%, 3  62,5%, 4
 87,5%) for the statistical analyses.
Only adrenal cells were taken into account, all other cells (like connective tissue and
endothelium) were left outside. The intensity of staining was graded 0-3; negative, 0;
weak, 1; moderate, 2; strong, 3. This was based on the mean intensity when there
was heterogeneous staining.
The immunohistochemical staining of the sections were scored by three observers
who did not know to which patients the sections belonged and which antibody (IGFIR, IGF-IIR, GHR or GH) had been used. All of the observers got slides with the
examples of weak, moderate and strong staining (figures 1-3).
Figure 1: Weak staining
Figure 2: Moderate staining
7
Figure 3: Strong staining
3.4 Statistical methods
The difference in expression of IGF-IR, IGF-IIR, GHR and GH among cortisolproducing ACAs, cortisol-producing ACCs and healthy adrenal glands was analyzed
with a one way ANOVA with Bonferroni adjustment. A Q-Q plot was performed to
determine if the data were normally distributed. In all statistical comparisons, p<0,05
was accepted as a significant difference. All analyses were executed using SPSS
16.0 software.
8
Results
4.1 Location of staining
The IHC demonstrated the following location of the GH, GHR, IGF-IR and IGF-IIR in
the normal adrenal glands and ATs (table 1).
Table 1: Location of staining
Cytoplasm
Nucleus
Cell membrane
+
+
+
+
+/+/+
+
+/-
GH
GHR
IGF-IR
IGF-IIR
Nuclear
membrane
+/+/-
4.2 Proportion and intensity of staining
The IHC staining of the GHR demonstrated significantly intenser staining in the ACAs
when compared to normal adrenal glands (table 7). The proportion of staining of
staining with antibodies against GHR and the proportion and the intensity of staining
with antibodies against GH, IGF-IR and IGH-IIR between ACA, ACC and normal
adrenals, did not differ significantly (tables 6, 8-13).
Table 2: Proportion and intensity of staining - GH
N
Mean
Std. Error of
Median
Mean
ACC
Std.
Deviation
Intensity
7
1,76
,17
1,67
,46
Proportion
7
12,50
2,41
12,50
6,36
Intensity
9
1,70
,13
1,67
,39
Proportion
9
17,59
4,27
20,83
12,80
Normal adrenal: zona
Intensity
7
1,81
,49
1,00
1,288
glomerulosa
Proportion
7
17,07
11,50
,00
30,42
Normal adrenal: zona
Intensity
7
1,48
,29
1,00
,77
fasciculata
Proportion
7
6,55
3,26
,00
8,62
Normal adrenal cortex
Intensity
7
1,64
,38
1,17
1,02
Proportion
7
11,81
7,17
2,09
18,97
ACA
9
Table 3: Proportion and intensity of staining – GHR
N
Mean
Std. Error of
Median
Mean
ACC
Std.
Deviation
Intensity
7
2,43
,19
2,33
,50
Proportion
7
44,05
5,74
45,83
15,19
Intensity
9
2,89
,14
2,67
,41
Proportion
9
63,43
7,01
70,83
21,02
Normal adrenal: zona
Intensity
7
2,10
,16
2,00
,42
glomerulosa
Proportion
7
57,73
9,55
62,50
25,28
Normal adrenal: zona
Intensity
7
1,57
,158
1,33
,42
fasciculata
Proportion
7
33,33
7,66
29,17
20,27
Normal adrenal cortex
Intensity
7
7
1,84
,15
1,67
,40
45,54
8,11
45,83
21,46
ACA
Proportion
Table 4: Proportion and intensity of staining – IGF-IR
N
Mean
Std. Error of
Median
Mean
ACC
Std.
Deviation
Intensity
7
3,33
,19
3,33
,51
Proportion
7
62,50
7,50
70,83
19,84
Intensity
9
3,33
,26
3,67
,78
Proportion
9
64,35
8,87
70,83
26,61
Normal adrenal: zona
Intensity
7
3,72
,18
4,00
,49
glomerulosa
Proportion
7
80,36
7,14
87,50
18,90
Normal adrenal: zona
Intensity
7
2,95
,17
3,00
,45
fasciculata
Proportion
7
45,83
7,27
45,83
19,24
Normal adrenal cortex
Intensity
7
7
3,34
,17
3,50
,46
63,10
6,77
66,67
17,91
ACA
Proportion
Table 5: Proportion and intensity of staining – IGF-IIR
N
Mean
Std. Error of
Median
Mean
ACC
Std.
Deviation
Intensity
7
3,33
,21
3,00
,54
Proportion
7
60,12
3,50
62,50
9,27
Intensity
9
2,81
,28
2,67
,83
Proportion
9
50,00
3,54
54,17
10,62
Normal adrenal: zona
Intensity
7
3,09
,38
3,33
,99
glomerulosa
Proportion
7
57,74
11,62
62,50
30,75
Normal adrenal: zona
Intensity
7
2,81
,37
3,33
1,00
fasciculata
Proportion
7
33,93
9,11
25,00
24,11
Normal adrenal cortex
Intensity
7
7
2,95
,37
3,33
,98
45,83
10,10
41,67
26,71
ACA
Proportion
10
Table 6: ANOVA analysis of GH - Intensity of staining
(I)
(J)
Group
Group
s
s
1
2
.02
.12
1.00
-.29
.33
3
.08
.13
1.00
-.25
.41
1
-.02
.12
1.00
-.33
.29
3
.06
.12
1.00
-.25
.37
1
-.08
.13
1.00
-.41
.25
2
-.06
.12
1.00
-.37
.25
2
3
Mean
Std. Error
Sig.
95% Confidence Interval
Difference (I-J)
Lower Bound
Upper Bound
Group 1 = ACCs, group 2 = ACAs, group 3 = normal adrenal glands
Table 7: ANOVA analysis of GHR – Intensity of staining
(I)
(J)
Group
Group
s
s
1
2
-.46
.22
.14
-1.03
.11
3
.59
.23
.06
-.02
1.20
1
.46
.22
.14
-.11
1.03
3
1.05
*
.22
.00
.48
1.62
1
-.59
.23
.06
-1.20
.02
2
-1.05*
.22
.00
-1.62
-.48
2
3
Mean
Std. Error
Sig.
95% Confidence Interval
Difference (I-J)
Lower Bound
Upper Bound
Group 1 = ACCs, group 2 = ACAs, group 3 = normal adrenal glands
*. The mean difference is significant at the 0.05 level.
Table 8: ANOVA analysis of IGF-IR - Intensity of staining
(I)
(J)
Group
Group
s
s
1
2
-.00
.31
1.00
-.82
.82
3
-.00
.33
1.00
-.87
.86
1
.00
.31
1.00
-.82
.82
3
-.00
.31
1.00
-.82
.81
1
.00
.33
1.00
-.86
.87
2
.00
.31
1.00
-.81
.82
2
3
Mean
Std. Error
Sig.
Difference (I-J)
95% Confidence Interval
Lower Bound
Upper Bound
Group 1 = ACCs, group 2 = ACAs, group 3 = normal adrenal glands
11
Table 9: ANOVA analysis of IGF-IIR - Intensity of staining
(I)
(J)
Group
Group
s
s
1
2
.52
.41
.65
-.55
1.59
3
.38
.43
1.00
-.75
1.52
1
-.52
.41
.65
-1.59
.55
3
-.14
.41
1.00
-1.20
.93
1
-.38
.43
1.00
-1.52
.75
2
.14
.41
1.00
-.93
1.20
2
3
Mean
Std. Error
Sig.
Difference (I-J)
95% Confidence Interval
Lower Bound
Upper Bound
Group 1 = ACCs, group 2 = ACAs, group 3 = normal adrenal glands
Table 10: ANOVA analysis of GH – Proportion of staining
(I)
(J)
Group
Group
s
s
1
2
-5.08
6.86
1.00
-23.00
12.85
3
.68
7.28
1.00
-18.33
19.69
1
5.08
6.86
1.00
-12.85
23.00
3
5.76
6.86
1.00
-12.16
23.69
1
-.68
7.28
1.00
-19.69
18.33
2
-5.76
6.86
1.00
-23.69
12.16
2
3
Mean
Std. Error
Sig.
Difference (I-J)
95% Confidence Interval
Lower Bound
Upper Bound
Group 1 = ACCs, group 2 = ACAs, group 3 = normal adrenal glands
Table 11: ANOVA analysis of GHR - Proportion of staining
(I)
(J)
Group
Group
s
s
1
2
-19.38
9.88
.19
-45.18
6.43
3
-1.49
10.48
1.00
-28.86
25.88
1
19.38
9.88
.19
-6.43
45.18
3
17.89
9.88
.26
-7.91
43.69
1
1.49
10.48
1.00
-25.88
28.86
2
-17.89
9.88
.26
-43.69
7.91
2
3
Mean
Std. Error
Sig.
Difference (I-J)
95% Confidence Interval
Lower Bound
Upper Bound
Group 1 = ACCs, group 2 = ACAs, group 3 = normal adrenal glands
12
Table 12: ANOVA analysis of IGF-IR - Proportion of staining
(I)
(J)
Group
Group
s
s
1
2
.018
.82
1.00
-2.12
2.16
3
-.04
.87
1.00
-2.31
2.23
1
-.018
.82
1.00
-2.16
2.12
3
-.06
.82
1.00
-2.20
2.09
1
.04
.87
1.00
-2.23
2.31
2
.06
.82
1.00
-2.09
2.20
2
3
Mean
Std. Error
Sig.
Difference (I-J)
95% Confidence Interval
Lower Bound
Upper Bound
Group 1 = ACCs, group 2 = ACAs, group 3 = normal adrenal glands
Table 13: ANOVA analysis of IGF-IIR - Proportion of staining
(I)
(J)
Group
Group
s
s
1
2
10.12
8.51
.75
-12.11
32.34
3
14.28
9.02
.39
-9.29
37.86
1
-10.12
8.51
.75
-32.34
12.11
3
4.17
8.51
1.00
-18.06
26.39
1
-14.28
9.02
.39
-37.86
9.29
2
-4.17
8.51
1.00
-26.39
18.06
2
3
Mean
Std. Error
Sig.
Difference (I-J)
95% Confidence Interval
Lower Bound
Upper Bound
Group 1 = ACCs, group 2 = ACAs, group 3 = normal adrenal glands
13
Discussion
A significant higher intensity of the staining of GHR was demonstrated in the ACAs
as compared to the normal adrenal glands. In addition, the intensity of staining in
ACAs is also higher than in the ACCs (although this is not significant). The proportion
of staining was not significantly different. These results make the role of GHR in
tumorigenesis less likely.
There were no significant differences between the intensity or proportion of staining
of GH, IGF-IR and IGF-IIR between the normal adrenal glands and the ATs.
Therefore, there also seems to be no roles for GH, IGF-IR and IGF-IIR in
tumorigenesis in adrenal tissue.
Based on these results, the use of GH-inhibitors or the use of GHR-, IGF-IR-, or IGFIIR-blockers do not seem useful in the treatment of dogs with cortisol-producing
adrenocortical tumors.
Unfortunately, there were some shortcomings in this research. The most important
one being the immunohistochemical staining protocol, which may not have been
specific enough. The staining was very different among the sections in one group.
Further optimising of the protocols may be needed.
Another shortcoming is the subjective way of the assessment of the sections. They
were assessed by three observers. There were some directives, so the grading of the
sections would be as objective as possible. However, after the completed forms were
evaluated, it showed that there were many differences in grading apparent among
the observers (see tables 2-5). In a future research it would be advisable to train the
observers in the assessement of sections and give more directives of how the
sections should be graded. When there are still great differences between the scoring
of some sections among the observers, it would be advisable to take another look at
those sections with the observers together, to come to a consensus.
14
References
Beuschlein F, Reincke M. 2006 Adrenocortical Tumorigenesis. Annals of the New York
Academy of Sciences 1088:319-334
Boulle N, Logié A, Gicquel C, Perin L, Le Bouc Y. 1998 Increased levels of insulin-like
growth factor II (IGF-II) and IGF-binding protein-2 are associated with malignancy in
sporadic adrenocortical tumors. Journal of Clinical Endocrinology and Metabolism 83:17131720
De Cremoux P, Rosenberg D, Goussard J, Brémont-Weil C, Tissier F, Tran-Perennou C,
Groussin L, Bertagna X, Bertherat J, Raffin-Sanson ML. 2008 Expression of progesterone
and estradiol receptors in normal adrenal cortex, adrenocortical tumors, and primary
pigmented nodular adrenocortical disease. Endocrine-Related Cancer 15:465-474
Fottner C, Engelhardt D, Weber MM. 1998 Regulation of steroidogenesis by insulin-like
growth factors (IGFs) in adult human adrenocortical cells: IGF-I and, more potently, IGF-II
preferentially enhance androgen biosynthesis through interaction with the IGF-I receptor and
IGF-binding proteins. Journal of Endocrinology 158:409-417
Fottner Ch, Hoeflich A, Wolf E, Weber MM. 2004 Role of the insulin-like growth factor
system in adrenocortical growth control and carcinogenesis. Hormone and Metabolic
Research 36:397-405
Galac S, Reusch CE, Kooistra HS, Rijnberk A. 2010 Adrenals. In: Rijnberk A, Kooistra HS,
eds. Clinical endocrinology of dogs and cats, 2nd edition. Hannover, Germany: Schlütersche
93-154
Mol JA, Van Garderen E, Selman PJ, Wolfswinkel J, Rijnberk A, Rutteman GR. 1995
Growth hormone mRNA in mammary gland tumors of dogs and cats. The Journal of Clinical
Investigation 95:2028-2034
Perry JK, Starling Emerald B, Mertani HC, Lobie PE. 2006 The oncogenic potential of
growth hormone. Growth Hormone & IGFResearch 16:277-289
Ragazzon B, Assié G, Bertherat J. 2011 Transcriptome analysis of adrenocortical cancers:
from molecular classification to the identification of new treatments. Endocrine-Related
Cancer 18:R15-R27
Weber MM, Fottner C, Wolf E. 2000 The role of the insulin-like growth factor system in
adrenocortical tumourigenesis. European Journal of Clinical Investigation 30:69-75
15