Spindle Assembly Checkpoint Gene Expression in Childhood Adrenocortical Tumors (ACT): Overexpression of Aurora Kinases A and B Is Associated With a Poor Prognosis
Kleiton Silva Borges, MSc,1 Daniel Antunes Moreno, PhD,1 Carlos Eduardo Martinelli Jr., MD, PHD,2 Sonir Roberto Rauber Antonini, MD, PHD,2 Margaret de Castro, MD, PHD,3 Silvio Tucci Jr., MD, PHD,4 Luciano Neder, MD, PHD,5 Leandra Naira Zambelli Ramalho, MD, PHD,5 Ana Luiza Seidinger, PhD,6 Izilda Cardinalli, MD, PHD,6 Maria José Mastellaro, MD, PHD,6 José Andres Yunes, PhD,6 Silvia Regina Brandalise, MD, PHD,6 Luiz Gonzaga Tone, MD, PHD,1,2 and Carlos Alberto Scrideli, MD, PhD2*
Background. Pediatric adrenocortical tumors (ACT) are rare malignancies and treatment has a small impact on survival in advanced disease and the discovery of potential target genes could be important in new therapeutic approaches. Methods. The mRNA expression levels of spindle checkpoint genes AURKA, AURKB, BUB, and BUBR1 were analyzed in 60 children with ACT by quantitative real time PCR. The anticancer effect of ZM447439, an experimental AURK inhibitor, was analyzed in a primary childhood ACT culture carrying the TP53 p.R337H mutation. Results. A significant association was observed between malignancy as defined by Weiss score ≥3 and higher AURKA (2.0-fold, P=0.01), AURKB (7.0-fold, P=0.007), and BUBR1 (5.8-fold, P=0.007) gene expression, and between unfavorable event (death or relapse) and higher expression
of AURKA (6.0-fold, P=0.034) and AURKB (17-fold, P=0.013). Overexpression of AURKA and AURKB was associated with lower event-free survival in uni- (P < 0.001 and P= 0.006, respectively) and multivariate (P=0.002 and P=0.03, respectively) analysis. Signifi- cant lower Event free survival (EFS) was also observed in patients with moderate/strong immunostaining to AURKA (P=0.012) and AURKB (P=0.045). ZM447439 was able to induce inhibition of proliferation and colony formation in a primary childhood ACT culture carrying the TP53 p.R337H mutation. Conclusion. Our results suggest that AURKA and AURKB overexpression in pediatric ACT may be related to more aggressive disease and the inhibition of these proteins could be an interesting approach for the treatment of these tumors. Pediatr Blood Cancer 2013;60:1809-1816. @ 2013 Wiley Periodicals, Inc.
Key words: adrenocortical tumor; AURK; childhood; prognosis; spindle assembly checkpoint genes
INTRODUCTION
Pediatric adrenocortical tumors (ACT) are very rare malignan- cies corresponding to only 0.2% of pediatric cancers [1-3]. However, in Southern Brazil an inherited germline TP53 mutation (p.R337H) greatly increases the predisposition to the development of childhood ACT, whose incidence is 10-15 higher than the worldwide rate. This heterozygous germline mutation has been found in over 80% of Brazilian patients with childhood ACT [4-6].
Current therapy for pediatric ACT primarily relies on surgical resection of the tumor. Mitotane (a DDT-related compound), with or without DNA-damaging agents, has been used in advanced disease, although with small impact on survival [3]. The overall 5- year disease-free survival is estimated at 50%; however, patients with stage IV disease have less than a 20% chance of long-term survival [5,6]. Thus, new therapeutic approaches are needed to improve the survival of these patients.
Chromosomal instability (CIN) is a hallmark of cancer [7]. Several studies have shown high levels of CIN in adrenocortical carcinoma [8,9], inclusive in childhood patients where chromo- somal imbalances were associated with overall survival and tumor progression [10,11]. Several genes are related with development of CIN, among them are the spindle checkpoint genes [12].
Spindle checkpoint genes are important for chromosome segregation during mitosis. A faithful segregation of the duplicate chromosomes is required to preserve the genome stability and one of the main mechanisms of cell division control is the spindle assembly checkpoint (SAC) [12]. SAC is a cell cycle surveillance system that ensures proper kinetochore-microtubule attachment, being essential to avoid chromosome missegregation [13]. The SAC is composed of several signaling proteins including the Aurora kinase family, budding uninhibited by benzimidazoles 1 (BUB1), BUB2, BUB3, and BUB1 homologue beta (BUB1B, also known as BUBR1), mitotic arrest
deficient protein 1 (MAD1), MAD2, and monopolar spindle 1 (MPS1). Dysfunction of these genes could result in aneuploidy, which is implicated in the development of many human cancers [12,13]. Among the SAC components, BUB1, BUBR1, and Aurora kinases A and B are those most frequently involved in human cancer [14,15].
Although some of these genes have been described as overex- pressed in adult ACT in microarray studies [16,17], the expression profile of SAC-related genes in pediatric ACT and their clinical impact remain undetermined. The aim of the present study was to analyze the expression profile of the genes AURKA, AURKB, BUB1,
1Department of Genetics, Ribeirão Preto Medical School, University of São Paulo, Ribeirão Preto, São Paulo, Brazil; 2Department of Pediatrics, Ribeirão Preto Medical School, University of São Paulo, Ribeirão Preto, São Paulo, Brazil; 3Department of Internal Medicine, Ribeirão Preto Medical School, University of São Paulo, Ribeirão Preto, São Paulo, Brazil; 4Department of Surgery, Ribeirão Preto Medical School, University of São Paulo, Ribeirão Preto, São Paulo, Brazil; 5Department of Pathology, Ribeirão Preto Medical School, University of São Paulo, Ribeirão Preto, São Paulo, Brazil; ‘Depart- ment of Oncology Pediatric, Centro Infantil Boldrini and State University of Campinas, Campinas, São Paulo, Brazil
Grant sponsor: Fundação de Amparo a Pesquisa do Estado de São Paulo (FAPESP); Grant numbers: 2010/7020-9; 2010/08699-5
Additional Supporting Information may be found in the online version of this article at the publisher’s web-site.
Conflict of interest: Nothing to declare.
*Correspondence to: Carlos Alberto Scrideli, Pediatrics Department, Ribeirão Preto Medical School, University of São Paulo, Avenida Bandeirantes 3900, Ribeirão Preto, SP 14049-900, Brazil.
E-mail: scrideli@fmrp.usp.br
Received 11 April 2013; Accepted 24 May 2013
@ 2013 Wiley Periodicals, Inc.
DOI 10.1002/pbc.24653
and BUBR1 in consecutive pediatric ACT and its association with clinical and biological features, as well as the anticancer effect of ZM447439, an experimental AURK inhibitor, on a primary childhood ACT culture carrying the TP53 p.R337H mutation.
SUBJECTS AND METHODS
Subjects
Sixty consecutive pediatric patients with ACT diagnosed between October 1991 and July 2009 were enrolled in the study. Disease stage was classified according to the modified Sandrini classification proposed for childhood ACT [3] and the histological diagnosis was based on the criteria described by Weiss et al. [18]. Eleven non-neoplastic adrenal cortex samples were used as control tissues and were obtained from children who had undergone nephrectomy due to Wilms’ tumor without adrenal invasion. The study was approved by the local Ethics Committees (protocol numbers 9374/2003 and 8380/2010) and a signed statement of informed consent was obtained from the parents of pediatric patients and control subjects.
We evaluated 46 females and 14 males with a mean age at diagnosis of 40.5 months (range 5-187 months). Fifty-seven patients presented hormone-secreting tumors. Thirty-five patients were classified as stage I, 10 as stage II, 8 as stage III, and 7 as stage IV. The TP53 p.R337H mutation was evaluated by direct DNA sequencing and was detected in 52/60 (86.6%) patients with ACT. Forty-six tumors (76.6%) presented Weiss classification scores ≥3, 13 scores <3, and 1 had an undetermined score. After surgery, patients were examined monthly during the first year, at least twice a year for 5 years, and thereafter annually. Follow-up ranged from 8 to 168 months (median: 68.3 months). Prognostic clinical and pathological features are described in Table I.
RNA Isolation and cDNA Synthesis
Frozen tumor tissues and control adrenal cortex samples were microdissected. Total RNA of tumor and control tissue samples was
extracted using TRIzol Reagent (Invitrogen, Karlsruche, Germany) according to the manufacturer’s protocol. cDNA synthesis was performed with the High Capacity cDNA Reverse Transcription kit (Applied Biosystems, Inc., Foster City, CA) and MultiScribe enzyme using 500 ng RNA in a 25 ul reaction.
mRNA Quantification by Real-Time PCR (RQ-PCR)
The relative mRNA expression levels of the genes analyzed were quantified using real-time PCR analysis with the ABI PRISM 7700 Sequence Detection System (Applied Biosystems, Inc.). Amplification of the specific sequences was performed using on demand TaqMan probes: AURKA (Hs00269212_m1), AURKB (Hs00177782_m1), BUB1 (Hs01557701_m1), BUBR1 (Hs00176169_m1), and GAPDH (4326317-E 0905031) (Applied Biosystems, Inc.). All samples were analyzed in triplicate using GAPDH as the reference gene. The 2-AACT equation was applied to calculate the relative expression of ACT samples [19]. The median gene expression values of all non-neoplastic adrenal tissues were used as reference and defined as 1.
Immunohistochemistry (IHC) Analysis
Immunohistochemistry studies were performed in a subset of 48 tumor tissue samples from the 60 patients. The sections were incubated with a monoclonal primary antibody specific for AURKA and AURKB (Millipore, Temecula, CA, 1:100 dilution) according to the manufacturer’s instructions, followed by signal detection with the avidin-biotin system (Vectastain Elite ABC Kit, Universal; Vector Laboratories, Inc., Burlingame, CA). Immunohistochemical results were evaluated for intensity and staining frequency of nuclear and cytoplasmic components. The intensity of staining was graded 0 (negative), 1 (weak), 2 (moderate), and 3 (strong). The percentage of positive cells was rated as follows: 0 points, 0%; 1 point, 1-10%; 2 points, 11-25%; 3 points, 25-50%; 4 points, >51%. Points for expression and percentage of positive cells were added (total score of 0-7), and classified as negative (score 0), weak (total 1-2), moderate (total 3-4), or strong (total 5-7).
| 5-Year EFS | P-value | 5-Year OS | P-value | |
|---|---|---|---|---|
| Age | ||||
| <4 years (n:46) | 81.2±6.0 | 0.002* | 90.6±4.5 | 0.004* |
| ≥4 years (n:14) | 36.9±13.8 | 53.8±13.8 | ||
| Tumor size | ||||
| <100 g (n:40) | 88.3±5.5 | <0.001* | 97.2±2.7 | <0.001* |
| ≥100 g (n:20) | 36.7 ±11.6 | 48.5± 12.1 | ||
| <200 cm3 (n:44) | 84.4±5.9 | <0.001* | 92.5±4.2 | <0.001* |
| ≥200 cm3 (n:16) | 35.2 ±12.4 | 47.6±13.8 | ||
| Pathologyª | ||||
| Weiss score <3 (n:13) | 100.0±0.0 | 0.01* | 100.0±0.0 | 0.05* |
| Weiss score ≥3 (n:46) | 60.7±7.9 | 74.8±7.0 | ||
| TP53 p.R337H status | ||||
| Negative (n:8) | 85.7 ±13.2 | 0.24 | 100.0±0.0 | 0.16 |
| Positive (n:52) | 68.0±6.9 | 78.1±6.1 | ||
| Classificationb | ||||
| Stage 1 (n:35) | 86.8±6.2 | 0.003* | 96.9±3.1 | 0.001* |
| Stage II (n:10) | 54.9±17.2 | 68.6±15.1 | ||
| Stage III (n:8) | 62.5 ±17.1 | 68.6±18.6 | ||
| Stage IV (n:7) | 28.6±17.1 | 33.3±19.2 |
ªOne patient with an indeterminate score; bAccording Sandrini’s classification; * Bold-statistically significant (P ≤ 0.05). Pediatr Blood Cancer DOI 10.1002/pbc
Cell Primary Culture
A primary culture was obtained from a functioning adrenocor- tical carcinoma (stage IV with p.R337H TP53 mutation) diagnosed in a 1.1-year-old girl. Cells were cultured in HAM F10 (Gibco BRL, Life Technologies, Carlsbad, CA) supplemented with 10% fetal bovine serum, penicillin (100 U/ml) and streptomycin (100 µg/ml) at 37°℃ in a humidified 5% CO2 incubator.
Cultures were grown for 4 weeks and frozen. To experiments the cultures were frozen in SFB containing 10% DMSO and preserved in liquid nitrogen. To experiments the cultures were thawed and expanded for one additional passage. All the in vitro experiments were performed between the third and fifth passage. This culture overexpressed both AURKA (2.6-fold) and AURKB (14.9-fold) when compared to the median values of the non- neoplastic samples.
Aurora Kinase Inhibitor
The ZM447439 (ZM) inhibitor is an experimental selective ATP-competitive inhibitor of Aurora-A, -B, and -C in vitro [20] and was obtained from Tocris Cookson, Inc. (Ellisville, MO). The compound was diluted to 10 mmol/L using 99% DMSO and stored at -20℃.
Proliferation Assay
For the proliferation assay, cells were seeded on 96-well plates (3 × 103/well). After 24 hours, the medium was replaced with fresh medium containing the treatment and then cultured for 24, 48, and 72 hours. After treatment, the culture medium was removed and replaced with medium containing 10 ul XTT dye (3 mg/ml) (XTT II; Roche Molecular Biochemicals, Indianapolis, IN) in each well. The plates were incubated for 2 hours at 37℃ and the formazan product were measured at 450 nm using an iMarkmicroplate reader (Bio-Rad Laboratories, Hercules, CA). The experiments were performed twice in triplicate at different times. Data were analyzed by the median-effect method (Calcusyn software; Biosoft, Ferguson, MO) to calculate the dose with 50% inhibition of proliferation (Dm or IC50 values).
Clonogenic Assay
The effect of ZM447439 on clonogenic capacity was evaluated by the clonogenic assay. After trypsinization, single cell suspen- sions of 300 cells were seeded in 6-well plates and treated with ZM447439 at concentrations of 200 nM, 800 nM, and 1 µM for 48 hours. After treatment, the culture medium was removed and replaced with drug-free medium. The cell cultures were incubated for 15 days and the colonies were then rinsed with PBS, fixed with methanol and stained with Giemsa. The colonies with >50 cells were counted. Assays were performed in triplicate.
Statistical Analysis
The association between the following variables: age (<4 years vs. ≥4 years), tumor weight (<100g vs. ≥100g), tumor size (<200 cm3 vs. ≥200 cm3), TP53 p.R377H mutation (positive vs. negative), pathology (Weiss score <3 vs. ≥ 3), disease stage, event- free survival (death/relapse vs. complete remission), overall survival (death vs. alive), and the expression levels of each gene
analyzed was determined by the Mann-Whitney test. Event free survival (EFS) analysis (with relapse and/or death due to any cause being considered as unfavorable events) was carried out based on Kaplan-Meier curves, using the twofold higher median values of gene expression by non-neoplastic adrenal samples as the cut-off point. Curves for different groups were compared by the log-rank test. The Cox proportional regression model was used for multivariate analysis of prognostic factors. The correlation between the expression values of the analyzed genes was determined by the Spearman correlation coefficient. Cell-line data were analyzed by ANOVA followed by the Bonferroni test, as appropriate. Data were analyzed using the SPSS Graduate Pack 20.0 software (SPSS, Inc., Chicago, IL) and GraphPad Prism 5.0 (GraphPad Software, San Diego, CA). The level of significance was set at P ≤ 0.05 in all analyses.
RESULTS
Expression of SAC-Related Genes in ACT and Non- Neoplastic Controls
Compared to non-neoplastic adrenal samples, ACT presented significantly higher expression levels of the BUB1 (6.0-fold, P< 0.0001) and BUBR1 (6.5-fold, P=0.001) genes. No significant differences were observed regarding the expression of AURKA (P=0.090) and AURKB (P =0.437; Fig. 1).
Expression of SAC-Related Genes and Clinical and Biological Characteristics
It is interesting to point out that a positive correlation was observed between AURKA and AURKB (r =0.496, P <0.001) and also between BUB1 (r=0.288, P =0.022) and BUBR1 (r = 0.418, P=0.001).
Higher expression levels of AURKA (2.0-fold, P=0.01), AURKB (7.0-fold, P=0.01), and BUBR1 (5.8-fold, P=0.007) were associated with ACT with a Weiss score ≥3. Patients with unfavorable events presented higher expression levels of AURKA (6.0-fold, P=0.034) and AURKB (17-fold, P=0.013). Death was associated with higher expression levels of AURKA (4.8-fold, P=0.050) and AURKB (11.2-fold, P=0.040). Patients with p.R337H TP53 mutation presented significantly higher expression values of AURKA (8.5-fold, P=0.014) and BUBR1 (7.5-fold, P=0.009). These data are shown in Figure 2.
In order to analyze a possible correlation between the expression levels of these genes and 5-year EFS, the patients were divided into two groups using the twofold higher than median value of the expression of the respective genes observed in non-neoplastic adrenal controls as a cut-off between low and high expression.
High expression of the AURKA gene was associated with a significantly lower 5-year EFS (34.3±13.1% vs. 82.6±6.6%, P <0.0001), with a higher risk of an unfavorable event (HR: 6.822, 95% CI: 2.400-19.394, P < 0.0001). The same was true for AURKB (40.0 ± 15.5% vs. 77.5±6.8%, P=0.006), with a higher risk of an unfavorable event (HR: 3.799, 95% CI: 1.350-10.687, P=0.01) (Fig. 3). Multivariate analysis showed that high expression of the AURKA gene was an independent unfavorable prognostic factor (P=0.002) when analyzed in combination with stage IV, age >4 years, weight >100g and tumor size ≥200 cm3. Excluding AURKA from the model, the AURKB gene (P =0.026; HR: 3.624,
Pediatr Blood Cancer DOI 10.1002/pbc
Borges et al.
A
10,00-
B
P = 0.09
5,00
cDNA relative expression (log10)
cDNA relative expression (log10)
4,00
P = 0.437
3,00
2,00
1,00-
1,00-
,00
,00-
ACT
non-neoplatic adrenal
ACT
non-neoplastic adrenal
AURKA
AURKB
C
D
cDNA relative expression (log 10)
100,00
6.0-fold, P<0.0001
Relative expression (log10)
6.5-fold, P = 0.001
10,00
10,00
1,00-
1,00-
,00
,00
ACT
non-neoplastic adrenal
ACT
non-neoplastic adrenal
BUB1
BUBR1
95% CI: 1.160-11.382) was also an prognostic factor when analyzed in combination with weight ≥100g (P=0.030, HR: 6.211, 95% CI: 1.195-32.280), age >4 years (P=0.574), stage IV (P=0.294) and tumor size ≥200 cm3 (P=0.898). No significant differences in 5-year EFS were observed for BUB1 (P=0.12) or BUBR1 (P=0.32).
When only the patients with Weiss score ≥3 were analyzed, higher expression levels of AURKA (5.3-fold, P=0.044) and AURKB (9.3-fold, P = 0.016) were observed in patients presenting an unfavorable event. In this group, high expression of the AURKA gene were associated with a lower 5-year EFS (34.3±13.1% vs. 73.4±9.6%, P=0.002) and a higher risk of an unfavorable event (HR: 4.476, 95% CI: 1.580-12.678, P = 0.005). High expression of AURKB were also associated with a lower 5-year EFS (33.3 ± 15.7% vs. 68.8 ±9.0%, P=0.020) and with a higher risk of an unfavorable event (HR: 3.184, 95% CI: 1.132-8.958, P=0.028; Fig. 3). Multivariate analysis showed that high expression of the AURKA gene unfavorable prognostic factor (P = 0.004, HR: 5.454, 95% CI: 1.734-17.152) and AURKB gene (P=0.036; HR: 3.338, 95% CI: 1.080-10.276) were also an independent when analyzed in combination with age, stage IV, weight ≥100g and tumor size ≥200. No significant differences in 5-year EFS were observed for BUB1 (P=0.26) or BUBR1 (P=0.84).
Expression of AURKA and AURKB Determined by IHC
The AURKA and AURKB protein expression was evaluated by IHC and the results were similar to those observed when using RQ-
PCR. Forty-one out of 48 (85.4%) ACT samples analyzed showed immunoreactivity for AURKA. In 12 cases (25%) immunoreactivity was classified as strong, in 17 cases (35.4%) as moderate, in 12 (25%) as weak and in 7 (14.6%) it was negative. Regarding AURKB, immunoreactivity was positive in 37 out of 48 (77.8%) ACT samples. Six (12.5%) tumors presented immunoreactivity classified as strong, 16 (35.5%) as moderate, 15 (31.2%) as weak and 8 (16.6%) showed negative or weak immunoreactivity; in three cases (6.2%) the result was inconclusive due to intense background staining and they were excluded from analysis. The cases were interpreted as AURKA or AURKB positive if the neoplastic cells showed moderate or strong staining. In the group with positivity for the AURKA protein the median relative mRNA expression detected by RQ-PCR was 1.17 (mean: 2.71, range: 0.00-16.94) and in the group with weak/negative AURKA expression the median mRNA expression was 0.50 (mean: 0.97, range: 0.00-8.67) (P =0.040). In the group with positivity for AURKB protein the median relative mRNA expression detected by RQ-PCR was 0.24 (mean: 2.44, range: 0.00-25.75) and in the group with weak/negative expression the median mRNA expression was 0.06 (mean: 0.67, range: 0.00- 8.41) (P=0.010). A representative AURKA and AURKB IHC staining is shown in Figure 4.
The 5-year EFS was lower in the patients with strong/moderate AURKA immunoreactivity (55.2±10.5% vs. 94.4±5.4%, P= 0.012) with a higher risk of an unfavorable event (HR: 8.717, 95% CI: 1.123-67.692, P=0.038). Patients with strong/moderate AUKB immunoreactivity had a lower 5-year EFS (53.2± 11.9% vs. 84.8 ± 8.1%, P=0.045) and a higher risk of an unfavorable event (HR: 3.487, 95% CI: 0.942-12.909, P=0.061).
A
10,00
E
2.0-fold, P = 0.010
7.5-fold, P = 0.009
cDNA relative expression (log10)
cDNA relative expression (log10)
10,00-
1,00
1,00-
,00-
,00
Weiss < 3
Weiss ≥3
TP53 wild type
TP53 p.R337H
AURKA
BUBR1
B
10,00-
7.0-fold, P = 0.010
F
cDNA relative expression (log10)
cDNA relative expression (log10)
6.0- fold, P = 0.034
10,00-
1,00
1,00-
,00-
,00
Weiss <3
Weiss ≥3
Remission
Event (relapse/death)
AURKB
AURKA
C
G
10,00-
17-fold, P = 0.013
cDNA relative expression (log10)
5.8-fold, P = 0.007
cDNA relative expression (log10)
10,00
1,00-
1,00
,00
,00
Weiss < 3
Weiss ≥3
Remission
Event (relapse/death)
BUBR1
AURKB
D
10,00-
H
cDNA relative expression (log10)
4.8-fold, P = 0.040
cDNA relative expression (log10)
8.5-fold, P = 0.014
10.00-
1,00-
1.00
,00
.00
TP53 wild type
TP53 p.R337H
Alive
Death
AURKA
AURKA
A
AURKA
C
1.0-
1.0-
AURKA
< 2-fold
< 2-fold
..: """≥ 2-fold
..: ""≥ 2-fold
< 2-fold-censored
0.8
+< 2-fold-censored
× ≥ 2-fold-censored
0.8
>2-fold-censored
Cumulative Survival
n = 43
Cumulative Survival
n = 30
0.6-
0.6
0.4-
0.4-
n = 17
n = 16
0.2-
0.2
0.0
P < 0.0001
0.0
P = 0.002
00
50.00
100.00
150.00
200.00
00
50.00
100.00
150.00
200.00
Event free survival (months)
Event free survival (months)
B
D
1.0-
AURKB
1.0
AURKB
< 2-fold
2-fold
..! ”:≥ 2-fold
..: ""≥ 2-fold
+ < 2-fold-censored
< 2-fold-censored
0.8
≥ 2 fold-censored
0.8
≥ 2 fold-censored
Cumulative Survival
n= 47
Cumulative Survival
0.6
n = 35
0.6
0.4
0.4
n = 13
n = 11
0.2
0.2
0.0-
P = 0.006
0.0
P = 0.020
.00
50.00
100.00
150.00
200.00
.00
50.00
100.00
150.00
200.00
Event free survival (months)
Event free survival (months)
Functional Studies
Primary culture of ACT was sensitive to ZM, showing a dose- and time-dependent inhibition of proliferation after treatment (P < 0.05) (Fig. 5), with an IC value of 14.9±0.2 M. ZM also inhibited colony formation in primary culture in a dose-dependent manner, reaching a maximum effect at the dose of 1 p.M (P < 0.05) (Fig. 5).
A
C
8
B
D
DISCUSSION
The current treatment strategies for advanced pediatric ACT or a high risk of disease recurrence have a small impact on survival [3]. Inhibition of the Aurora and SAC kinases has been suggested as a potential target to treat human cancers but its therapeutic potential for childhood ACT has not been considered before. In this study we demonstrate the overexpression of BUB1 and BUBR1 in ACT compared to non-neoplastic adrenal as well as the association of overexpression of Aurora kinases A and B with a poor outcome in these tumors. Moreover, we present the first in vitro evidence that Aurora inhibitors may be effective for the treatment of these patients.
Adrenocortical carcinomas are characterized by high levels of CIN [9]. In childhood adrenal tumors chromosomal imbalances were associated with overall survival and tumor progression [10,11]. . Among the genes associated in maintaining chromosomal integrity are the SAC kinases such as BUB1 and BUBR1. They play a central role in preventing aneuploidy by missegregation of sister chromatids during mitosis [14,21]. BUB1 and BUBR1 have been found expressed at high levels in many human cancers and it has been suggested that they can be responsible for neoplastic transformation [22,23].
In the present study we found a highly significant over- expression of BUB1 and BUBR1 in ACT compared to non- neoplastic adrenal tissue and also of BUBR1 in tumors with a Weiss score ≥3. These results are similar to those reported for other human cancers when compared to their normal counterparts, such as clear cell kidney carcinomas [22] and gastric cancer [23],
Pediatr Blood Cancer DOI 10.1002/pbc
A
1.5-
24h
Proliferation (% DMSO)
48h
1.0-
T
T
I
I
I
72h
I
0.5-
0.0
DMSO
2uM
5UM
10μΜ
20μΜ
[ZM]
B
Surviving Fraction
1.2
0.8
0.4
*
0
DMSO
200nM
800nM
[ZM]
1 µM
suggesting that BUB1 and BUBR1 could be involved in the development of childhood ACT. The present data also agrees with previous microarray study on adult ACT where BUBR1 has been reported to be overexpressed in carcinomas when compared to adenomas and non-neoplastic adrenals [17]. Moreover, combined expression of BUBR1 with the PINK1 gene has been described as a predictor of survival in adult ACT [16]. In our study, no significant association between the expression of BUB1 and BUBR1 and survival was found in childhood ACT. This result is in agreement with a study showing no association of BUB1 expression with overall survival in childhood ACT [24].
The Aurora kinase family comprises three structurally related serine/threonine kinases (Aurora A-C) that ensure proper spindle assembly and accurate chromosome segregation in mitosis [25]. Studies have shown that the ectopic expression of AURKA in mouse NIH3T3 cells and Rat-1 fibroblasts causes centrosome amplifica- tion and transformation in vitro as well as tumorigenesis in vivo [26], but the same was not true for AURKB [15]. However, both AURKA and AURKB can induce tetraploidy and enhance chromo- some instability and tumor invasiveness [15].
The present study showed that AURKA and AURKB over- expression was significantly associated with more advanced as well as lower tumor stages. Microrarray studies of adult ACT also demonstrated that AURKA and AURKB expression was increased and associated with a poor prognosis [16]. A correlation between AURKA and AURKB up-regulation and clinical aggres- siveness has been described for several human cancers including acute myelogenous leukemia [27], head and neck squamous Pediatr Blood Cancer DOI 10.1002/pbc
cell carcinoma [28], squamous cell cervical cancer [29], non- small cell lung carcinoma [30], glioblastoma [31], ovarian carcinoma [32], bladder carcinoma [33], and hepatocellular carcinoma [34]. These finding suggest that AURKA and AURKB may be involved in the molecular pathways of childhood ACT and their inhibition could be a new approach to its treatment.
It has been described that TP53 is a negative regulator of AURKA and its deficiency leads to overexpression of AURKA [35]. This observation is inagreementwith our data which showed significant overexpression of AURKA in patients carrying p.R337H TP53 mutation, suggesting that this mutation could be associated with AURKA overexpression in ACT. The role of AURKs in ACT without or with TP53 mutation other than p.R337H needs further investigation.
Preclinical studies have demonstrated potent effects of the specific inhibitory activity of the Aurora kinases [36-38] including preclinical models of pediatric cancer [39]. Results of clinical phase I studies using different Aurora kinase inhibitors have demonstrated that in general these drugs are well-tolerated and have limited non- hematological toxicity [36,37]. Several phase II clinical trials are currently being conducted in different human cancers (http:// clinicaltrials.org).
ZM447439 is an experimental Aurora kinase inhibitor with effects very related to others Aurora kinase inhibitors currently in clinical trial such as VX-680, PF-03814735 and AT9283 [15,20]. In the present study we showed that ZM447439 reduced the colony formation of a childhood ACT primary culture specimen presenting the p.R337H TP53 mutation. Moreover, ZM447439 effectively inhibited the proliferation of this ACT culture. These results were similar to those observed in other studies targeting Aurora kinases in different cancer cells and in xenograft models [38-40], suggesting a potential therapeutic effect also on childhood ACT.
The IC50 value of ZM in ACT primary culture was similar to that previously reported in studies that also treated primary cultures with Aurora kinase inhibitors such as ZM and R763/ AS703569 [38,40]. Primary tumor cultures are believed to be more similar to in vivo tumors than cancer cell lines [38], thereby, it is interesting that ZM caused a growth-inhibiting activity on primary culture of ACT and diminished effectively its survival after long term exposure as showed through clonogenic assay. Although all these functional findings have been interesting, others should be conduced using human cell model of ACT and xenograft model to validate the potential effects of Aurora kinase inhibition in ACT.
In summary, we have shown that AURKA and AURKB overexpression was significantly associated with more aggressive disease and risk of relapse and death in childhood ACT, especially in those with Weiss criteria ≥3. In addition, we demonstrated that Aurora kinase inhibition suppressed growth of a childhood ACT primary culture specimen presenting the p.R337H TP53 mutation. On this basis, controlled preclinical and clinical studies are necessary to evaluate the therapeutic potential of Aurora kinase inhibitors in childhood ACT.
ACKNOWLEDGMENTS
This study was supported by public grants from Fundação de Amparo a Pesquisa do Estado de São Paulo (FAPESP) (grant 2010/ 7020-9). FAPESP fellowship to KSB (2010/08699-5).
1. Sandrini R, Ribeiro RC, DeLacerda L. Childhood adrenocortical tumors. J Clin Endocrinol Metab 1997;82:2027-2203.
2. Michalkiewicz E, Sandrini R, Figueiredo B, et al. Clinical and outcome characteristics of children with adrenocortical tumors: A report from the International Pediatric Adrenocortical Tumor Registry. J Clin Oncol 2004;22:838-845.
3. Rodriguez-Galindo C, Figueiredo BC, Zambetti GP, et al. Biology, clinical characteristics, and management of adrenocortical tumors in children. Pediatr Blood Cancer 2005;45:265-273.
4. Latronico AC, Pinto EM, Domenice S, et al. An inherited mutation outside the highly conserved DNA- binding domain of the p53 tumor suppressor protein in children and adults with sporadic adrenocortical tumors. J Clin Endocrinol Metab 2001;86:4970-4993.
5. Sandrini F, Villani DP, Tucci S, et al. Inheritance of R337H p53 gene mutation in children with sporadic adrenocortical tumor. Horm Metab Res 2005;37:231-235.
6. Leal LF, Mermejo LM, Ramalho LZ, et al. Wnt/beta-catenin pathway deregulation in childhood adrenocortical tumors. J Clin Endocrinol Metab 2011;96:3106-3114.
7. Carter SL, Eklund AC, Kohane IS, et al. A signature of chromosomal instability inferred from gene expression profiles predicts clinical outcome in multiple human cancers. Nat Genet 2006;38:1043-1048.
8. Dohna M, Reincke M, Mincheva A, et al. Adrenocortical carcinoma is characterized by a high frequency of chromosomal gains and high-level amplifications. Genes Chromosomes Cancer 2000;28:145-152.
9. Stephan EA, Chung TH, Grant CS, et al. Adrenocortical carcinoma survival rates correlated to genomic copy number variants. Mol Cancer Ther 2008;7:425-431.
10. Figueiredo BC, Stratakis CA, Sandrini R, et al. Comparative genomic hybridization analysis of adrenocortical tumors of childhood. J Clin Endocrinol Metab 1999;84:1116-1121.
11. Loncarevic IF, Hering A, Posorski N, et al. Number of genomic imbalances correlates with the overall survival for adrenocortical cancer in childhood. Pediatr Blood Cancer 2008;51:356-362.
12. Silva P, Barbosa J, Nascimento AV, et al. Monitoring the fidelity of mitotic chromosome segregation by the spindle assembly checkpoint. Cell Prolif 2011;44:391-400.
13. Malumbres M, Barbacid M. Cell cycle kinases in cancer. Curr Opin Genet Dev 2007;17:60-65.
14. Bolanos-Garcia VM, Blundell TL. BUB1 and BUBR1: Multifaceted kinases of the cell cycle. Trends Biochem Sci 2010;36:141-150.
15. Green MR, Woolery JE, Mahadevan D. Update on aurora kinase targeted therapeutics in oncology. Expert Opin Drug Discov 2011;6:291-307.
16. de Reyniès A, Assié G, Rickman DS, et al. Gene expression profiling reveals a new classification of adrenocortical tumors and identifies molecular predictors of malignancy and survival. J Clin Oncol 2009;27:1108-1115.
17. Giordano TJ, Thomas DG, Kuick R, et al. Distinct transcriptional profiles of adrenocortical tumors uncovered by DNA microarray analysis. Am J Pathol 2003;162:521-531.
18. Weiss LM, Medeiros LJ, Vickery AL, Jr., Pathologic features of prognostic significance in adrenocortical carcinoma. Am J Surg Pathol 1989;13:202-206.
19. 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;25:402-408.
20. Ditchfield C, Johnson VL, Tighe A, et al. Aurora B couples chromosome alignment with anaphase by targeting BubR1, Mad2, and Cenp-E to kinetochores. J Cell Biol 2003;161:267-280.
21. Ricke RM, Jeganathan KB, van Deursen JM. Bub1 overexpression induces aneuploidy and tumor formation through Aurora B kinase hyperactivation. J Cell Biol 2011;193:1049-1064.
22. Pinto M, Vieira J, Ribeiro FR, et al. Overexpression of the mitotic checkpoint genes BUB1 and BUBR1 is associated with genomic complexity in clear cell kidney carcinomas. Cell Oncol 2008;30:389-395.
23. Grabsch H, Takeno S, Parsons WJ, et al. Overexpression of the mitotic checkpoint genes BUB1, BUBR1, and BUB3 in gastric cancer-association with tumour cell proliferation. J Pathol 2003;200:16- 22.
24. Fragoso MC, Almeida MQ, Mazzuco TL, et al. Combined expression of BUB1B, DLGAP5, and PINK1 as predictors of poor outcome in adrenocortical tumors: Validation in a Brazilian cohort of adult and pediatric patients. Eur J Endocrinol 2012;166:61-67.
25. Boss DS, Beijnen JH, Schellens JH. Clinical experience with aurora kinase inhibitors: A review. Oncologist 2009;14:780-793.
26. Zhou H, Kuang J, Zhong L, et al. Tumour amplified kinase STK15/BTAK induces centrosome amplification, aneuploidy and transformation. Nat Genet 1998;20:189-193.
27. Lucena-Araujo AR, de Oliveira FM, Leite-Cueva SD, et al. High expression of AURKA and AURKB is associated with unfavorable cytogenetic abnormalities and high white blood cell count in patients with acute myeloid leukemia. Leuk Res 2011;35:260-264.
28. Reiter R, Gais P, Jutting U, et al. Aurora kinase A messenger RNA overexpression is correlated with tumor progression and shortened survival in head and neck squamous cell carcinoma. Clin Cancer Res 2006;12:5136-5141.
29. Twu NF, Yuan CC, Yen MS. Expression of Aurora kinase A and B in normal and malignant cervical tissue: High Aurora A kinase expression in squamous cervical cancer. Eur J Obstet Gynecol Reprod Biol 2009;142:57-63.
30. Smith SL, Bowers NL, Betticher DC, et al. Overexpression of aurora B kinase (AURKB) in primary non- small cell lung carcinoma is frequent, generally driven from one allele, and correlates with the level of genetic instability. Br J Cancer 2005;93:719-729.
31. Scrideli CA, Carlotti CG, Jr., Okamoto OK, et al. Gene expression profile analysis of primary glioblastomas and non-neoplastic brain tissue: Identification of potential target genes by oligonucleotide microarray and real-time quantitative PCR. J Neurooncol 2008;88:281-291.
32. Yang F, Guo X, Yang G, et al. AURKA and BRCA2 expression highly correlate with prognosis of endometrioid ovarian carcinoma. Mod Pathol 2011;24:836-845.
33. Bufo P, Sanguedolce F, Tortorella S, et al. Expression of mitotic kinases phospho-aurora A and aurora B correlates with clinical and pathological parameters in bladder neoplasms. Histol Histopathol 2010;25:1371-1377.
34. Lin ZZ, Jeng YM, Hu FC, et al. Significance of Aurora B overexpression in hepatocellular carcinoma. Aurora B overexpression in HCC. BMC Cancer 2010;10:461.
35. Wu CC, Yang TY, Yu CT, et al. p53 negatively regulates Aurora A via both transcriptional and posttranslational regulation. Cell Cycle 2012;11:3433-3442.
36. Steeghs N, Eskens FA, Gelderblom H, et al. Phase I pharmacokinetic and pharmacodynamic study of the Aurora kinase inhibitor danusertib in patients with advanced or metastatic solid tumors. J Clin Oncol 2009;27:5094-5101.
37. Diamond JR, Bastos BR, Hansen RJ, et al. Phase I safety, pharmacokinetic, and pharmacodynamic study of ENMD-2076, a novel angiogenic and Aurora kinase inhibitor, in patients with advanced solid tumors. Clin Cancer Res 2011;17:849-860.
38. Borges KS, Castro-Gamero AM, Moreno DA, et al. Inhibition of Aurora kinases enhances chemosensitivity to temozolomide and causes radiosensitization in glioblastoma cells. J Cancer Res Clin Oncol 2012;138:405-414.
39. Carol H, Boehm I, Reynolds CP, et al. Efficacy and pharmacokinetic/pharmacodynamic evaluation of the Aurora kinase A inhibitor MLN8237 against preclinical models of pediatric cancer. Cancer Chemother Pharmacol 2011;68:1291-1304.
40. Mclaughlin J, Markovtsov V, Li H, et al. Preclinical characterization of Aurora kinase inhibitor R763/ AS703569 identified through an image-based phenotypic screen. J Cancer Res Clin Oncol 2010;136:99- 113.