ELSEVIER
Domestic Animal Endocrinology
journal homepage: www.domesticanimalendo.com
DOMESTIC ANIMAL ENDOCRINOLOGY
Expression of angiogenesis-related genes in canine cortisol-secreting adrenocortical tumors
M.M.J. Kool, S. Galac*, H.S. Kooistra, J.A. Mol
Department of Clinical Sciences of Companion Animals, Faculty of Veterinary Medicine, Utrecht University, Yalelaan 108, 3584 CM, Utrecht, The Netherlands
ARTICLE INFO
Article history: Received 26 June 2013
Received in revised form 31 October 2013 Accepted 5 November 2013
Keywords:
Cushing’s syndrome Hypercortisolism Angiopoietin VEGF
Dog
ABSTRACT
The aim of this study was to evaluate the expression of angiogenesis-related genes in canine cortisol-secreting adrenocortical tumors (ATs). Quantitative RT-PCR analysis revealed mRNA encoding for vascular endothelial growth factor, vascular endothelial growth factor receptors 1 and 2, angiopoietin 1 and 2 (ANGPT1 and ANGPT2), the splice variant ANGPT2443, the ANGPT-receptor Tie2, and basic fibroblast growth factor in 38 canine cortisol-secreting ATs (26 carcinomas and 12 adenomas) and 15 normal adrenals. The relative expression of both ANGPT2 and ANGPT2443 was higher in adenomas (P = 0.020 for ANGPT2 and P = 0.002 for ANGPT2443) and carcinomas (P = 0.003 for ANGPT2 and P < 0.001 for ANGPT2443) compared with normal adrenals, and this enhanced expression was also detected with Western blot analysis. Immunohistochemistry indicated expression of ANGPT2 protein in AT cells and in vascular endothelial cells of carcinomas, whereas Tie2 was mainly present in the tumor vascular endothelial cells. The ANGPT2-to-ANGTPT1 ratio, a marker for a proangiogenic state, was higher in both adenomas (P = 0.020) and carci- nomas (P = 0.043). With the use of the human H295R cortisol-producing adrenocortical carcinoma cell line, we were able to demonstrate that the ANGPT2 expression was stim- ulated by cyclic adenosine monophosphate and progesterone but not by cortisol. In conclusion, canine cortisol-secreting ATs have enhanced ANGPT2 expression with a concomitant shift toward a proangiogenic state. On the basis of this information, treatment modalities may be developed that interfere with ANGPT2 expression, including inhibition of the cyclic adenosine monophosphate/protein kinase A pathway, or of the effect of ANGPT2, by using specific ANGPT2 inhibitors.
@ 2014 Elsevier Inc. All rights reserved.
1. Introduction
Adrenocorticotropin (ACTH)-independent hyper- cortisolism in dogs as a result of autonomous glucocorticoid production by an adrenocortical tumor (AT) accounts for approximately 15% of cases of spontaneous canine hyper- cortisolism [1]. The treatment of choice is adrenalectomy, because the successful complete removal of the affected adrenal gland will eliminate the clinical signs related to glucocorticoid excess without the need for lifelong
medication. However, invasive growth in the surrounding tissues or metastasis or both may preclude complete removal. Factors involved in tumor growth and metastasis of canine cortisol-secreting ATs are largely unknown.
In general, angiogenesis, the process of new blood vessel formation from existing vasculature, is an important factor in tumor development and metastasis. By means of intratumoral angiogenic feedback loops, tumors may acti- vate angiogenesis and provide themselves with the nutri- ents and oxygen necessary to grow beyond a certain size [2]. In human cancer research, the use of antiangiogenic drugs is one of the most rapidly emerging therapeutic strategies [3]. The aim of the present study was to evaluate the expression of angiogenesis-related genes in canine
* Corresponding author. Tel .: +31 30 253 9683, fax: +31 30 251 8126. E-mail address: S.Galac@uu.nl (S. Galac).
cortisol-secreting ATs. Knowledge on the role of angio- genesis in AT development may help in the development of new treatment modalities.
In the regulation of tumor angiogenesis and the devel- opment of antiangiogenic drugs, basic fibroblast growth factor (bFGF), the vascular endothelial growth factor (VEGF) family, and the angiopoietin (ANGPT) family play a pivotal role [2,4]. Although inhibition of bFGF is still in the early stages of development, its role as a potent mitogen for fetal and adult adrenocortical cells [5,6] and its increased expression in human adrenal medullary tumors [7] make bFGF a target of specific interest in adrenal gland pathology.
Drugs that target VEGF have shown potential in various clinical and preclinical trials, inhibiting angiogenesis and tumor growth [2,8]. Furthermore, high VEGF expression has been associated with increased microvascular density in human adrenocortical carcinoma [9].
Angiopoietin signaling has likewise been implicated as a significant factor in the pathogenesis of human ATs [10]. Especially, the ratio between ANGPT1 and ANGPT2 is considered an important indicator of activation of the angiogenic switch in tumors [11]. Selective ANGPT2 inhibi- tion, either combined with VEGF inhibition or by itself, has shown promise in slowing tumor angiogenesis and tumor growth in different tumor types [12,13]. The expression of ANGPT2 in the human adrenal gland is thought to be regu- lated by ACTH-cyclic adenosine monophosphate (cAMP)- steroidogenic factor 1 signaling [14,15]. Recently, we demonstrated that a large proportion of canine cortisol- secreting ATs harbors an activating mutation in guanine nucleotide binding protein, alpha stimulating, the gene responsible for cAMP production on ACTH-stimulation [16]. Data about the relation between cAMP signaling and angiogenesis in canine ATs are lacking.
The aim of the study was to evaluate the expression of bFGF, VEGF, VEGF receptors 1 and 2, ANGPT1 and ANGPT2, the splice variant ANGPT2443, and the ANGPT-receptor Tie2 in canine cortisol-secreting carcinomas and adenomas, compared with that in normal adrenals. In addition, we investigated whether the expression of genes of interest was influenced by cAMP or the adrenocortical hormones cortisol and progesterone in vitro.
2. Materials and methods
2.1. Patient material
In this study 38 canine cortisol-producing ATs and 15 normal adrenal glands were used. Adrenal glands from healthy dogs were available as archived tissue for com- parison with AT tissue obtained from patients. After sur- gical removal of an AT in the patients or resection of a normal adrenal gland in the healthy dogs, the tissue was stored on ice and inspected, and material was saved for quantitative RT-PCR (qPCR) analysis and histopathology. The fragments for RNA isolation were cut and snap-frozen in liquid nitrogen within 10 min. They were kept at -80℃ until further use. The remaining part of the AT tissue was immersed in formalin for fixation and embedded in paraffin after 24 to 48 h. The tumor group consisted of all histologically confirmed ATs derived from patients referred
to the Department of Clinical Sciences of Companion Animals of the Faculty of Veterinary Medicine in Utrecht between 2001 and 2009 with clinical signs of hyper- cortisolism. The diagnosis of ACTH-independent Cushing’s syndrome because of an AT was based on elevated urinary corticoid-to-creatinine ratios, which were not suppressible with high doses of dexamethasone; suppressed or even undetectable plasma ACTH concentrations [1]; and demonstration of an AT by ultrasonography or computed tomography [17]. All dogs underwent unilateral adrenal- ectomy. The dogs’ ages at the time of surgery ranged from 6 to 14 y (mean, 9 y). Twelve dogs were mongrels, and the other dogs were of 10 different breeds. Eighteen dogs were male (8 castrated) and 20 female (15 neutered). The ages of the control dogs ranged from 2 to 5 y. Five of the control dogs were male and 10 were female; all control dogs were intact. Permission to use the AT tissue for this study was obtained from all patient owners, and the study was approved by the Ethical Committee of Utrecht University.
2.2. Histopathology
Histopathologic evaluation was performed on formalin- fixed and paraffin-embedded tissue slides of all samples and used to confirm the diagnosis and to classify the tu- mors. All histologic evaluations were performed by a single pathologist. Classification was performed on the basis of the criteria described by Labelle et al [18]. Classification as a carcinoma was based on histologic evidence of vascular invasion, peripheral fibrosis, capsular invasion, trabecular growth, hemorrhage, necrosis, and single-cell necrosis. Typical histologic characteristics of adenomas were he- matopoiesis, fibrin thrombi, and cytoplasmic vacuolization. On the basis of these criteria, the tumor group consisted of 12 adenomas and 26 carcinomas.
2.3. Total RNA extraction and reverse transcription
Total RNA for qPCR analysis was isolated from tissue and cell culture samples with the use of the RNeasy mini kit (Quiagen, Hilden, Germany), according to manufacturer’s protocols. An optional DNAse step was performed to avoid DNA contamination. RNA concentrations were measured on the NanoDrop ND-1000 (NanoDrop Technologies, Wil- mington, DE, USA). Synthesis of cDNA was performed with the iScript cDNA synthesis kit (Bio-Rad, Hercules, CA, USA), according to manufacturer’s protocols. For all samples, 1 cDNA reaction was performed without reverse tran- scriptase, to check for contamination with genomic DNA.
2.4. Quantitative RT-PCR
Quantitative RT-PCR primers for all target genes were designed with DNA-star primer select version 8.1, Oligo- explorer version 1.1.0, or Perl-primer version 1.1.14 ac- cording to the parameters in the Bio-Rad iCycler manual and were ordered from Eurogentec (Maastricht, The Netherlands). Primers for distinguishing both ANGPT2 var- iants were designed to anneal to areas of the transcript unique to each isoform. For each primer pair a qPCR tem- perature gradient was performed, to determine the optimal
annealing temperature. Formation of the proper PCR products was confirmed by a sequencing reaction, using the ABI3130XL Genetic analyzer (AB Applied Biosystems, Carlsbad, CA, USA) according to the manufacturer’s protocol.
A pool of all cDNA samples was used to create a 4-fold dilution series for use as a standard. The rest of the cDNA was diluted 4 times with milliQ water, to achieve a working stock. On all of the canine samples, mRNA expression levels for the following target genes were measured: VEGF, VEGFR1, VEGFR2, ANGPT1, ANGPT2 (both full-length [FL] and the splice variant ANGPT2443), Tie2, and bFGF. To correct for differences in cDNA concentration, ribosomal protein S5 (RPS5), RPS19, and hypoxanthine-guanine phosphor- ibosyltransferase were used as reference genes [19]. On the human H295R samples mRNA expression levels were determined for ANGPT2. Tata-binding protein gene and RPS19 were used as reference genes. Analysis of the relative expression levels of the reference genes showed no sig- nificant differences between groups, and their expression was shown to be stable with the use of GeNorm software [20], justifying their use as reference genes.
All qPCR reactions were performed on a MyIQ single- color real-time PCR detection system (Bio-Rad). Detection was performed with SYBRgreen supermix (Bio-Rad), and data were analyzed with iQ5 software (Bio-Rad). The raw data were used to calculate the reaction efficiency. Reaction efficiencies between 90% and 110% were accepted. Calcula- tion of normalized relative expression levels for each of the target genes was performed with the 44-Ct method [21].
Human and canine PCR primers and their characteristics are listed in Table 1.
2.5. Western blot analysis
To evaluate whether the differences on mRNA level for ANGPT2 were also reflected at the protein level, Western blot analysis for ANGPT2 was performed on all tissue samples. Protein was isolated from the frozen tissue samples with the use of radioimmunoprecipitation assay buffer. Protein con- centrations were measured with the DC protein assay (500- 0116; Bio-Rad). Gel electrophoresis was performed on 7.5% polyacrylamide gels together with a dual-color Precision plus Protein Standard (Bio-Rad) for molecular weight determi- nation. After gel electrophoresis, proteins were transferred onto Hybond electrochemiluminescence (ECL) nitrocellulose membrane (Amersham, GE Healthcare, Diegem, Belgium). Nonspecific binding was blocked with 4% ECL in a buffer of 50mM Tris (pH 7.6), 150mM NaCl, and 0.1% Tween. Blots were incubated overnight at 4℃ with the use of a goat polyclonal anti-ANGPT2 antibody known to cross-react with canine ANGPT2 (C-19, sc-7015; Santa Cruz Biotechnologies, Santa Cruz, CA, USA), in a dilution of 1:200. Blots were washed and incubated for 1 h at room temperature with a secondary antibody (donkey anti-goat horseradish peroxidase [HRP]- conjugated IgG, sc-2020; Santa Cruz Biotechnologies) at a dilution of 1:20,000. Proteins were visualized with the ECL advanced Western blotting detection kit (Amersham RPN2135; GE Healthcare) and measurement of chem- iluminescence (ChemiDoc XRS; Bio-Rad). As a loading control
| PCR primers (position) | Sequence (5'-3') | Annealing temperature, ℃ | Product length, bp | |
|---|---|---|---|---|
| cf_ANGPT1 | Fw | AAT AAT ATG CCA GAA CCC AAA AAG | 62 | 162 |
| (926-1087) | Rv | CCC CAG CCA ATA TTC ACC AGA G | ||
| cf_ANGPT2 | Fw | ACA GCA TCG GGA GAA GGC AGT ATC | 54 | 553 (ANGPT2 FL) |
| (106-658) | Rv | TCT TCT TTT ATT GAC CGT AGT TGA | 398 (ANGPT2443) | |
| cf_ANGPT2-FL | Fw | AGA ACC AGA CTG CCG TGA T | 65 | 110 |
| (376-485) | Rv | TGT TGT CTG ATT TAA TAC TTG TGC | ||
| cf_ANGPT2443 | Fw | TAC GCA GTG GCT AAT TAA GGT ATT | 64.5 | 226 |
| (293-518) | Rv | CTG GAG CTG ATC TTT CTC TTC TTT | ||
| cf_Tie2 | Fw | CAG CTT ACC AGG TGG ACA TTT TTG | 58 | 104 |
| (25-138) | Rv | GTC CGC TGG TGC TTG AGA TTT AG | ||
| cf_VEGF | Fw | CTT TCT GCT CTC CTG GGT GC | 58 | 102 |
| (6-107) | Rv | GGT TTG TGC TCT CCT CCT GC | ||
| cf_VEGFR1 | Fw | GGC TCA GGC AAA CCA CAC | 63 | 190 |
| (189-378) | Rv | CCG GCA GGG GAT GAC GAT | ||
| cf_VEGFR2 | Fw | GGA AGA GGA AGT GTG TGA CCC C | 64 | 181 |
| (3606-3785) | Rv | GAC CAT ACC ACT GTC CGT CTG G | ||
| cf_bFGF | Fw | TTC TTC CTG CGG ATC CA | 61 | 76 |
| (2251-2326) | Rv | GTT GCA ATT TGA CGT GGG | ||
| hs_ANGPT2 | Fw | AAC ATC CCA GTC CAC CTG AG | 60 | 102 |
| (1820-2021) | Rv | GGT CTT GCT TTG GTC CGT TA | ||
| hs_TBP | Fw | TGC ACA GGA GCC AAG AGT GAA | 63.5 | 132 |
| (940-1071) | Rv | CAC ATC ACA GCT CCC CAC CA | ||
| hs_RPS19 | Fw | CCT TCC TCA AAA AGT CTG GG | 61 | 95 |
| (431-525) | Rv | GTT CTC ATC GTA GGG AGC AAG |
Abbreviations: ANGPT, angiopoietin; bFGF, basic fibroblast growth factor; cf, Canis familiaris; FL, full length variant; Fw, forward; hs, Homo sapiens; RPS19, ribosomal protein S19; Rv, reverse; TBP, Tata binding protein; VEGF, vascular endothelial growth factor; VEGFR, vascular endothelial growth factor receptor. All positions are based on the mRNA sequence, as published on the National Center for Biotechnology Information GenBank database. Accession numbers used: canine ANGPT1, NM_001005754; ANGPT2, NM_001048126.1; Tie2, AF282848; VEGF, NM_001003175.2; VEGFR1, XM_534520; VEGFR2, NM_001048024.1; bFGF, XM_533298.1; human ANGPT2, NM_001147.2; TBP, NM_003194.4; RPS19, NM_001022.3.
all blots were stripped and incubated with a-tubulin anti- body (TUB-1A2, ab11325; AbCam, Cambridge, UK) in a 1:2,000 dilution. An HRP-conjugated anti-mouse IgG (R&D Systems, Minneapolis, MN, USA) was used as secondary antibody in a dilution of 1:20,000. Protein expression was quantified with QuantityOne software (Bio-Rad) and normalized to the intensity of a-tubulin staining. The speci- ficity of the ANGPT2 staining was confirmed by pre- incubation of the primary antibody with a blocking peptide for ANGPT2 (sc-7015P; Santa Cruz Biotechnologies) for 2 h at room temperature in a 1:40 dilution.
2.6. Immunohistochemistry
Tissue localization of ANGPT2 and its receptor Tie2 was evaluated in a representative subset of tumors and normal adrenals. For this purpose, immunohistochemical staining was performed on 21 formalin-fixed, paraffin-embedded tissue slides (10 carcinomas, 7 adenomas, and 4 normal ad- renals) for ANGPT2 and on 9 tissue slides (3 carcinomas, 3 adenomas, and 3 normal adrenals) for Tie2. Tissue slides were rehydrated in a series of xylene-alcohol baths. Antigen retrieval for ANGPT2 staining was performed with 10mM sodium citrate buffer, pH 6, for 20 min at 95℃. For Tie2 staining, no antigen retrieval was used. To block endogenous peroxidase activity, slides were incubated with peroxidase block (S2001; Dako, Glostrup Denmark) for 5 min (ANGPT2) or 20 min (Tie2). Nonspecific binding sites were blocked for 20 min, using 10% normal goat serum in PBS (ANGPT2) or 5% BSA in PBS (Tie2). A rabbit anti-human anti-ANGPT2 anti- body (Ab6583; AbCam) was used, in a 1:250 dilution in 10% normal goat serum in PBS. For Tie2, a rabbit anti-human antibody (sc-9026; Santa Cruz Biotechnologies) was used in a 1:200 dilution in 1% BSA in PBS. Slides were incubated with the primary antibody for 120 min at room temperature (ANGPT2) or overnight at 4℃ (Tie2). Subsequently, all slides were incubated with anti-rabbit HRP-conjugated secondary antibody (Dako K4003) for 30 min. Antibody detection was performed with Dako K3468 HRP substrate. All slides were incubated with diaminobenzidine for 4 min (ANGPT2) or 1.5 min (Tie2) and subsequently counterstained with hema- toxylin. The specificity of the reaction for ANGPT2 was confirmed by using blocking peptide (angiopoietin 2 pep- tide, ab98299; AbCam) in a concentration of 100 µg/mL. Preincubation of the antibody with this blocking peptide abolished all staining. For Tie2, the specificity of the reaction was verified by omission of the primary antibody as well as by using normal rabbit serum as first antibody. As a positive control, human placenta and canine ovarian tissue slides were used. All slides were analyzed in a descriptive manner, to determine the presence of positive staining, localization within the tissue, and cellular localization.
2.7. Cell culture
To investigate whether ANGPT expression was influ- enced by the cAMP/protein kinase A (PKA) pathway, in vitro experiments were performed with the human H295R adrenocortical carcinoma cell line. To exclude that the in- crease in ANGPT2 was due to an increase in cAMP-induced cortisol or progesterone secretion, the effect of both
dexamethasone and progesterone on ANGPT2 mRNA expression was investigated. The H295R cells were cultured according to a protocol described previously [22], adapted for use with commercially available Nu-serum. Growth medium consisted of Dulbecco modified Eagle medium/F12 (1:1) (GibcoBRL, Breda, The Netherlands), supplemented with 1% (vol/vol) insulin, transferrin, and selenium sup- plement (GibcoBRL); 1.25% (wt/vol) BSA (Sigma-Aldrich, St. Louis, MO, USA), 1% (wt/vol) penicillin/streptomycin, and 2.5% (vol/vol) Nu-serum (BD Biosciences, Franklin Lakes, NJ, USA). For all experiments, cells were plated onto 24-wells Primaria culture plates at a density of 5 x 105 cells per well and allowed to attach for 24 h. Next, cells were incu- bated with 8-bromo-cAMP sodium salt (Sigma-Aldrich; final concentrations in the medium: 10-3-10-5M), pro- gesterone (Sigma-Aldrich; 2 × 10-8M), or dexamethasone (Sigma-Aldrich; 10-7M) in serum-free medium, using a physiological range of concentrations. After 15 min Nu- serum was added to a concentration of 2.5%. After incuba- tion for 2 to 48 h the cells were prepared for RNA isolation, and medium was collected for cortisol measurement. All stimulations and controls were performed in replicates of 4, and for each time point an unstimulated control was included. Cortisol concentrations in the culture media were measured by means of RIA, as described previously [23].
2.8. Statistical analyses
Statistical analyses were performed with SPSS16 (IBM, Armonk, NY, USA). Individual relative mRNA expression levels for all qPCR experiments were calculated with the 44-Ct method [21]. The relative expression levels as obtained by qPCR, the normalized protein concentrations as obtained by Western blot analysis, and the cortisol concentrations from cell culture stimulation experiments were analyzed with 1-way ANOVA and Scheffe post hoc test or Student t test, to establish the presence of differ- ences between the groups and its statistical significance. For Western blot analysis and qPCR experiments on canine tissue, the relative expression levels in normal adrenals were compared with those in adenomas and carcinomas and between adenomas and carcinomas. For all H295R stimulation experiments, relative expression levels and cortisol concentrations of treated sample groups were compared with the values in the untreated control group at the same time point. Correlations between the relative expression levels of ANGPT2-FL and ANGPT2443 were calculated with Pearson correlation coefficient. For all sta- tistical tests, a P value < 0.05 was considered significant.
3. Results
3.1. Canine cortisol-secreting ATs overexpress ANGPT2 and have an increased ANGPT2-to-ANGPT1 ratio
Analysis of the relative mRNA expression levels of the target genes showed no significant differences in the expression levels of ANGPT1, Tie2, VEGF, VEGFR2, and bFGF. A significantly lower expression of VEGFR1 was found in ad- enomas than in normal adrenals (0.34-fold; P = 0.006).
With the use of variant-specific primers, the pres- ence of the ANGPT2 splice variant ANGPT2443 Was found for the first time in canine ATs. The relative expression levels of ANGPT2-FL were significantly higher in the tumor groups than in normal adrenals: 2.73-fold higher (P = 0.020) in adenomas and 4.56-fold higher (P = 0.003) in carcinomas. The increase in rela- tive expression of ANGPT2443 was 3.18-fold (P = 0.002) in adenomas and 6.60-fold (P < 0.001) in carcinomas. For ANGPT2443, the difference in relative expression between adenomas and carcinomas was also significant (2.07-fold; P = 0.036).
The ANGPT2-to-ANGPT1 ratio was calculated for ade- nomas, carcinomas, and normal adrenals, and the ratio was
significantly higher in both the adenomas (3.96-fold; P = 0.020) and carcinomas (7.82-fold; P = 0.043) than in normal adrenals (Fig. 1A, B). Levels of ANGPT2-FL and ANGPT2443 correlated significantly in both normal adrenals and ATs (r = 0.96; P < 0.001).
The presence and expression of ANGPT2 protein in canine ATs and normal adrenals was studied by Western blot analysis. In most of the samples 2 bands were detected at 61 and 68 kDa, corresponding to the expected weights of ANGPT2443 and ANGPT2-FL, respectively. Both ANGPT2-FL and ANGPT2443 showed a significantly more intense staining (P = 0.04) in the ATs than in normal adrenal tissue (Fig. 1C, D). No significant difference between adenomas and carcinomas could be detected.
A
B
*
10
Adenomas
12
*
Carcinomas
*
8
10
*
6
*
Fold change
Ang2/Ang1 ratio
8
*
4
*
*
6
2
*
T
4
-2
-4
*
2
-6
ANGPT2-FL
ANGPT2-443
0
ANGPT1
ANGPT2
Tie2
VEGF
VEGFR1
VEGFR2
bFGF
N
A
C
C
D
14
Normal adrenals
Adenomas
*
ANGPT2
N
A
C
C
12
Carcinomas
*
68 kDa
61 kDa
Relative expression
10
a-Tubulin
8
55 kDa
6
4
2
*
*
0
T
I
ANGPT2-FL
ANGPT2-443
Twelve patients were available for follow-up: 8 of them developed symptoms and signs of hypercortisolism within 2 y after surgery, and the recurrence of hypercortisolism was confirmed by endocrine testing. Four of them were in remission for at least 2 y after adrenalectomy. The relative mRNA expression levels of ANGPT2-FL and ANGPT2443 compared with normal adrenals were higher in the group with recurrence than in patients without recurrence of hypercortisolism (6.90-fold vs 2.57-fold for ANGPT2-FL and 10.11-fold vs 3.37-fold for ANGPT2443). However, these dif- ferences did not reach a significant level.
3.2. Immunohistochemical staining for ANGPT2 is present in both tumor cells and vascular endothelial cells in canine ATs and virtually absent in normal adrenals
Immunohistochemical staining of 17 canine cortisol- secreting ATs (7 adenomas and 10 carcinomas) and 4 normal adrenals showed a striking difference in ANGPT2 staining between normal adrenals and ATs. In normal
adrenal cortices, no staining was detected in the zona reticularis and zona fasciculata. The zona glomerulosa exhibited very low-to-absent staining (Fig. 2C). In all ATs a weak-to-moderate cytoplasmic granular staining was pre- sent in most of the tumor cells, whereas islands of moderately to strongly stained tumor cells were scattered throughout the tumor slides (Fig. 2D, E). In addition, in part of the ATs a strong cytoplasmic granular staining was seen in the vascular endothelial lining (Fig. 2E). The endothelial staining was detected in only 1 of the 7 adenomas and in 6 of the 10 carcinomas.
Immunohistochemical staining for the Tie2 receptor, the presence of which is needed to enable an effect of ANGPT2, was performed in 6 cortisol-secreting ATs (3 adenomas and 3 carcinomas) and 3 normal adrenals. Very low-to-negative staining was seen in the normal adrenal cortex and in most AT cells (Fig. 3A, B). In contrast, the tumor vasculature showed a clear cytoplasmic gran- ular staining pattern of the vascular endothelial cells which could not be found in normal adrenal cortex
A
B
C
D
200 um
E
F
100 ml
A
B
500 pin
100 um
C
D
100 Am-
(Fig. 3D). In addition, in some ATs distinct regions of Tie2- positive tumor cells were detected, with a cytoplasmic granular staining pattern (Fig. 3D).
3.3. Cyclic AMP stimulates ANGPT2 mRNA expression in H295R cells
Because cortisol secretion in adrenocortical cells is cAMP dependent, stimulation of H295R cells with the use of 8-bromo-cAMP resulted in a dose-dependent increase of cortisol in the medium. Interestingly, the highest 8-bromo- CAMP concentration also clearly stimulated the mRNA expression of ANGPT2 during the first 8 h of incubation (3.25-fold; P = 0.026), whereas expression returned to baseline values after 48 h (Fig. 4). Dexamethasone did not influence the expression of ANGPT2, but progesterone led to a transient induction of ANGPT2 expression (1.78-fold; P = 0.037) (Fig. 5).
4. Discussion
Angiogenesis is recognized as an important factor in tumor development and metastasis. In the present study we investigated the hypothesis that the highly vascularized, cortisol-secreting ATs have a higher expression of certain angiogenesis-related genes than normal adrenals. The re- sults of the present study show a markedly increased rela- tive expression of ANGPT2, at both the mRNA and protein levels, in canine cortisol-secreting adrenocortical adenomas and carcinomas compared with normal adrenals. The concurrently increased ANGPT2-to-ANGPT1 ratio is a reliable indicator for a shift of the angiogenic balance toward a
proangiogenic state [11]. These results are in line with the findings by Giordano et al [10], who reported a 3-fold in- crease of ANGPT2 expression in human adrenocortical car- cinomas with the use of microarray. On the basis of their observations, Giordano et al [10] suggested that ANGPT-
4
CAMP -5
·
CAMP -4
CAMP-3
3
Cortisol production Fold change
2
.
O
o
1
0
2h
8h
24h
48h
A
5
ANGPT2 expression Fold change
4
3
2
1
0
B
4h
8h
48h
5
ANGPT2 expression Fold change
4
3
2
1
0
4h
24h
48h
C
5
ANGPT2 expression Fold change
4
3
2
1
0
4h
24h
48h
induced angiogenesis is an important aspect of human adrenocortical tumorigenesis.
This is the first study to show the presence of the ANGPT2 splice variant ANGPT2443 in adrenocortical tumor tissue. Moreover, the fold changes in the relative mRNA expression of ANGPT2443 even exceeded that of ANGPT2-FL when ATs were compared with normal adrenocortical tissue. Although no data have been published on the presence and regulation of ANGPT2443 in human adreno- cortical tissue, expression analyses of ANGPT2443 in other human tumor cell lines and primary tumor tissues indicate a role in tumorigenesis for this splice variant [24]. The increased relative expression of ANGPT2443 mRNA in ATs, as well as its higher expression in malignant tumors, suggest a tumor-specific role for this variant in the development of canine ATs.
The higher expression of ANGPT-FL and ANGPT2443 and the higher ANGPT2-to-ANGPT1 ratio in carcinomas than in adenomas suggest involvement of the ANGPT family in adrenocortical carcinogenesis. This is supported by the higher ANGPT2-FL and ANGPT2443 expression levels in the group of dogs showing recurrence of hypercortisolism within a 2-y-period, although this difference did not reach significance, possibly because of low numbers. The higher expression of ANGPT2-FL and ANGPT2443 also suggest that the ANGPT family may be an interesting target for medical intervention.
In the present study we used immunohistochemistry to investigate the localization of the ANGPT2 protein. Studies that investigated ANGPT2 localization in various tumors in humans have documented the expression in either tumor cells, endothelial cells, or both, depending on the tumor type [25-27]. In the human fetal adrenal gland, ANGPT2 expression was shown to be present in the cortical cells of the definitive zone [14]. In the present study, ANGPT2 protein expression was found in AT cells but not in normal adrenocortical cells of the zona fasciculata and the zona reticularis, once more suggesting a role for ANGPT2 in AT development. The strong endothelial staining in the vasculature of several canine ATs is likely to represent destabilized blood vessels. Notably, this endothelial stain- ing was present mostly in carcinomas, possibly indicating a role in carcinogenesis.
To exhibit an effect, ANGPT2 has to bind to its receptor Tie2. Immunohistochemical localization and overexpression of Tie2 has been consistently shown in tumor vascular endothelial cells, but in some tumor types a positive staining is also present in the tumor cells, the significance of which is yet unknown [28]. Because the results of the present study suggest that ANGPT2 plays a role in AT development, the presence of the Tie2 receptor in AT had to be investigated. We used immunohistochemistry for this purpose. We detected a strong positive staining for Tie2 in the endothelial lining of the tumor vasculature. This vascular staining pattern might indicate the presence of a paracrine loop in which the ANGPT2 produced by tumor cells or vascular endothelial cells
conditions and for the same period of time. Significant changes (P < 0.05) are marked with an asterisk. ANGPT2, angiopoietin 2; cAMP, cyclic adeno- sine monophosphate.
or both binds to Tie2 receptors in the endothelial lining, thus stimulating angiogenesis. In some ATs regions of Tie2- positive tumor cells were detected, possibly indicating the presence of an autocrine ANGPT2-Tie2 loop in these neoplastic cells.
The function of ANGPT2 not only depends on the pres- ence of the Tie2 receptor but also on the presence of VEGF [29]. In this study, expression analysis found no differences in expression of VEGF and only a limited decrease in expression of VEGFR1 levels in adenomas. Nevertheless, in both human [9,30] and canine ATs abundant expression of VEGF appears to be present, albeit not different between ATs and normal adrenocortical tissue, which is necessary for an angiogenesis-stimulating effect of ANGPT2.
The expression of ANGPT2 is known to be regulated by a variety of factors, including hypoxia, several growth factors, VEGF, bFGF, and ACTH-cAMP signaling [31-33]. Because VEGF and bFGF did not differ significantly between ATs and normal adrenocortical tissue, and a high level of hypoxia would induce not only ANGPT2 expression but also VEGF expression, our attention was drawn toward the ACTH- cAMP pathway. In the human fetal adrenal, ACTH stimula- tion targets ANGPT expression, preferentially up-regulating ANGPT2 [14]. Signaling occurs by binding of ACTH to its cell surface receptor, which uses cAMP as its primary intracellular messenger. One of the effects of increased intracellular concentrations of cAMP in adrenocortical cells is increased expression of steroidogenic enzymes via pro- tein kinase A-steroidogenic factor 1 signaling [34]. More- over, recently we found that a relatively large proportion of canine ATs carries an activating mutation in guanine nucleotide binding protein, alpha stimulating, the gene responsible for cAMP production on ACTH-stimulation [16]. On the basis of these considerations, we hypothesized that cAMP may be driving the increased ANGPT2 expression in adrenocortical tumor cells. To test whether raised cAMP concentrations would indeed increase ANGPT2 expression, cell culture experiments were set up. The ideal model would of course consist of a canine, cortisol-producing AT cell line. Unfortunately, no such cell line is available. The H295R cell line provides the closest in vitro model system for AT cells, producing both cortisol and progesterone, and is a stable, well characterized cell line [35]. Stimulation experiments that use 8-bromo-cAMP in the H295R cells resulted in a dose-dependent increase of cortisol in the medium and a time-dependent increase of ANGPT2 expression, indicating the involvement of cAMP signaling in ANGPT2 regulation in adrenocortical tumor cells.
Because the human H295R adrenocortical carcinoma cells abundantly express glucocorticoid receptors and pro- gesterone receptors (PRs) [30,36], an increase in ANGPT2 expression because of an increase in cAMP-induced cortisol or progesterone secretion or both had to be investigated. An indirect effect on ANGPT2 expression by elevated cortisol levels could be excluded by showing that administration of the potent glucocorticoid dexamethasone did not result in increased ANGPT2 expression. Stimulation of H295R cells with progesterone, however, resulted in a transient increase in ANGPT2 mRNA. In malignant mammary tumors, well known for their high PR expression, a functional rela- tionship appears to exist between angiogenesis and the
presence and signaling of PRs. PR-positive mammary tu- mors have a significantly higher microvessel density [37], and VEGF expression in cultured mammary tumor cells is up-regulated by ligand binding-induced PR signaling [38]. The progesterone-induced higher expression of ANGPT2 in H295R cells may indicate that also in ATs a mechanism may be present in which PR signaling results in a shift of the angiogenic balance toward a proangiogenic state.
In conclusion, the increased relative expression of both ANGPT2-FL and ANGPT2443, the increased ANGPT2-to- ANGTPT1 ratio, the higher expression of ANGPT2443 in car- cinomas, the protein expression of ANGPT2 in AT cells and in vascular endothelial cells, and the strong Tie2 expression in vascular endothelial cells of carcinomas strongly suggest a role for the ANGPT family in AT development. Induction of ANGPT2 expression by 8-bromo-cAMP and progesterone provides some insight in the regulation of increased ANGPT2 expression in ATs. Further research is needed to determine exactly which role ANGPT2 plays in AT devel- opment, and whether targeting of the ANGPT2-Tie2 pathway by selective antagonists such as ANGPT2 traps or monoclonal antibodies [39,40] may hold promise as an adjunctive therapeutic option in ATs.
Acknowledgments
We thank Dr M.B.M. van Duursen of the Institute for Risk Assessment and Toxicology for kindly providing us with H295R adrenocortical carcinoma cells and Dr S. Klar- enbeek for the histopathologic evaluation of all ATs.
This study was financially supported by a Morris Animal Foundation - Pfizer Animal Health veterinary fellowship for advanced study (grant D09CA-913). The funding sources had no involvement in study design, collection, analysis and interpretation of data, writing of the report and in the decision to submit the article for publication.
M.M.J. Kool conceived and designed the study; acquired, analyzed, and interpreted the data; drafted the manuscript; and gave final approval of the definitive version. S. Galac conceived and designed the study; interpreted the data, critically revised the manuscript, and gave final approval of the definitive version. H.S. Kooistra: conceived and designed the study, critically revised the manuscript, and gave final approval of the definitive version. J.A. Mol conceived and designed the study, interpreted the data, critically revised the manuscript, and gave final approval of the definitive version.
The authors have no conflicts of interest to declare that could be perceived as prejudicing the impartiality of the research reported.
References
[1] Galac S, Reusch CE, Kooistra HS, Rijnberk A. Adrenals. In: Kooistra HS, Rijnberk A, editors. Clinical endocrinology of dogs and cats. Hannover, Germany: Schlütersche; 2010. pp. 93-154.
[2] Makrilia N, Lappa T, Xyla V, Nikolaidis I, Syrigos K. The role of angiogenesis in solid tumours: an overview. Eur J Intern Med 2009; 20:663-71.
[3] Oklu R, Walker TG, Wicky S, Hesketh R. Angiogenesis and current antiangiogenic strategies for the treatment of cancer. J Vasc Interv Radiol 2010;21:1791-805; quiz 1806.
[4] Auguste P, Javerzat S, Bikfalvi A. Regulation of vascular development by fibroblast growth factors. Cell Tissue Res 2003;314:157-66.
[5] Chu Y, Ho WJ, Dunn JC. Basic fibroblast growth factor delivery en- hances adrenal cortical cellular regeneration. Tissue Eng Part A 2009;15:2093-101.
[6] Mesiano S, Mellon SH, Gospodarowicz D, Di Blasio AM, Jaffe RB. Basic fibroblast growth factor expression is regulated by cortico- tropin in the human fetal adrenal: a model for adrenal growth regulation. Proc Natl Acad Sci U S A 1991;88:5428-32.
[7] Statuto M, Ennas MG, Zamboni G, Bonetti F, Pea M, Bernardello F, Pozzi A, Rusnati M, Gualandris A, Presta M. Basic fibroblast growth factor in human pheochromocytoma: a biochemical and immuno- histochemical study. Int J Cancer 1993;53:5-10.
[8] Ellis LM, Hicklin DJ. VEGF-targeted therapy: mechanisms of anti- tumour activity. Nat Rev Cancer 2008;8:579-91.
[9] Turner HE, Harris AL, Melmed S, Wass JA. Angiogenesis in endocrine tumors. Endocr Rev 2003;24:600-32.
[10] Giordano TJ, Thomas DG, Kuick R, Lizyness M, Misek DE, Smith AL, Sanders D, Aljundi RT, Gauger PG, Thompson NW, Taylor JM, Hanash SM. Distinct transcriptional profiles of adrenocortical tu- mors uncovered by DNA microarray analysis. Am J Pathol 2003;162: 521-31.
[11] Tait CR, Jones PF. Angiopoietins in tumours: the angiogenic switch. J Pathol 2004;204:1-10.
[12] Hashizume H, Falcon BL, Kuroda T, Baluk P, Coxon A, Yu D, Bready JV, Oliner JD, McDonald DM. Complementary actions of in- hibitors of angiopoietin-2 and VEGF on tumor angiogenesis and growth. Cancer Res 2010;70:2213-23.
[13] Hu B, Cheng SY. Angiopoietin-2: development of inhibitors for cancer therapy. Curr Oncol Rep 2009;11:111-6.
[14] Ishimoto H, Ginzinger DG, Jaffe RB. Adrenocorticotropin preferen- tially up-regulates angiopoietin 2 in the human fetal adrenal gland: implications for coordinated adrenal organ growth and angiogen- esis. J Clin Endocrinol Metab 2006;91:1909-15.
[15] Ferraz-de-Souza B, Lin L, Shah S, Jina N, Hubank M, Dattani MT, Achermann JC. ChIP-on-chip analysis reveals angiopoietin 2 (Ang2, ANGPT2) as a novel target of steroidogenic factor-1 (SF-1, NR5A1) in the human adrenal gland. FASEB J 2011;25:1166-75.
[16] Kool MM, Galac S, Spandauw CG, Kooistra HS, Mol JA. Activating mutations of GNAS in canine cortisol-secreting adrenocortical tu- mors [e-pub ahead of print September 20, 2013]. J Vet Intern Med doi: 10.1111/jvim.12194.
[17] Galac S, Kool MM, Naan EC, Daminet S, Mol JA, Kooistra HS. Expression of the ACTH receptor, steroidogenic acute regulatory protein, and steroidogenic enzymes in canine cortisol-secreting adrenocortical tumors. Domest Anim Endocrinol 2010;39:259-67.
[18] Labelle P, Kyles AE, Farver TB, De Cock HE. Indicators of malignancy of canine adrenocortical tumors: histopathology and proliferation index. Vet Pathol 2004;41:490-7.
[19] Brinkhof B, Spee B, Rothuizen J, Penning LC. Development and evaluation of canine reference genes for accurate quantification of gene expression. Anal Biochem 2006;356:36-43.
[20] Vandesompele J, De Preter K, Pattyn F, Poppe B, Van Roy N, De Paepe A, Speleman F. Accurate normalization of real-time quanti- tative RT-PCR data by geometric averaging of multiple internal control genes. Genome Biol 2002;3. RESEARCH0034.
[21] Livak KJ, Schmittgen TD. Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) Method. Methods 2001;25:402-8.
[22] Rainey WE, Saner K, Schimmer BP. Adrenocortical cell lines. Mol Cell Endocrinol 2004;228:23-38.
[23] Rijnberk A, van Wees A, Mol JA. Assessment of two tests for the diagnosis of canine hyperadrenocorticism. Vet Rec 1988;122: 178-80.
[24] Kim I, Kim JH, Ryu YS, Jung SH, Nah JJ, Koh GY. Characterization and expression of a novel alternatively spliced human angiopoietin-2. J Biol Chem 2000;275:18550-6.
[25] Emoto M, Charnock-Jones DS, Licence DR, Ishiguro M, Kawai M, Yanaihara A, Saito T, Hachisuga T, Iwasaki H, Kawarabayashi T, Smith SK. Localization of the VEGF and angiopoietin genes in uterine carcinosarcoma. Gynecol Oncol 2004;95:474-82.
[26] Hata K, Udagawa J, Fujiwaki R, Nakayama K, Otani H, Miyazaki K. Expression of angiopoietin-1, angiopoietin-2, and Tie2 genes in normal ovary with corpus luteum and in ovarian cancer. Oncology 2002;62:340-8.
[27] Moon WS, Rhyu KH, Kang MJ, Lee DG, Yu HC, Yeum JH, Koh GY, Tarnawski AS. Overexpression of VEGF and angiopoietin 2: a key to high vascularity of hepatocellular carcinoma? Mod Pathol 2003;16: 552-7.
[28] Martin V, Liu D, Fueyo J, Gomez-Manzano C. Tie2: a journey from normal angiogenesis to cancer and beyond. Histol Histopathol 2008; 23:773-80.
[29] Harfouche R, Hussain SN. Signaling and regulation of endothelial cell survival by angiopoietin-2. Am J Physiol Heart Circ Physiol 2006;291:H1635-45.
[30] de Cremoux P, Rosenberg D, Goussard J, Bremont-Weil C, Tissier F, Tran-Perennou C, Groussin L, Bertagna X, Bertherat J, Raffin- Sanson ML. Expression of progesterone and estradiol receptors in normal adrenal cortex, adrenocortical tumors, and primary pig- mented nodular adrenocortical disease. Endocr Relat Cancer 2008; 15:465-74.
[31] Fujii T, Kuwano H. Regulation of the expression balance of angiopoietin-1 and angiopoietin-2 by Shh and FGF-2. In Vitro Cell Dev Biol Anim 2010;46:487-91.
[32] Hegen A, Koidl S, Weindel K, Marme D, Augustin HG, Fiedler U. Expression of angiopoietin-2 in endothelial cells is controlled by positive and negative regulatory promoter elements. Arterioscler Thromb Vasc Biol 2004;24:1803-9.
[33] Simon MP, Tournaire R, Pouyssegur J. The angiopoietin-2 gene of endothelial cells is up-regulated in hypoxia by a HIF binding site located in its first intron and by the central factors GATA-2 and Ets- 1. J Cell Physiol 2008;217:809-18.
[34] Gallo-Payet N, Payet MD. Mechanism of action of ACTH: beyond CAMP. Microsc Res Tech 2003;61:275-87.
[35] Rainey WE, Bird IM, Mason JI. The NCI-H295 cell line: a pluripotent model for human adrenocortical studies. Mol Cell Endocrinol 1994; 100:45-50.
[36] Tacon LJ, Soon PS, Gill AJ, Chou AS, Clarkson A, Botling J, Stalberg PL, Skogseid BM, Robinson BG, Sidhu SB, Clifton-Bligh RJ. The gluco- corticoid receptor is overexpressed in malignant adrenocortical tumors. J Clin Endocrinol Metab 2009;94:4591-9.
[37] Vamesu S. Angiogenesis and progesterone receptor status in pri- mary breast cancer patients: an analysis of 158 needle core biopsies. Rom J Morphol Embryol 2007;48:267-74.
[38] Hyder SM, Chiappetta C, Stancel GM. Pharmacological and endogenous progestins induce vascular endothelial growth fac- tor expression in human breast cancer cells. Int J Cancer 2001; 92:469-73.
[39] Huang H, Bhat A, Woodnutt G, Lappe R. Targeting the ANGPT-TIE2 pathway in malignancy. Nat Rev Cancer 2010;10:575-85.
[40] Eroglu Z, Stein CA, Pal SK. Targeting angiopoietin-2 signaling in cancer therapy. Expert Opin Investig Drugs 2013;22:813-25.