Mol Biol Rep (2014) 41:6125–6131 DOI 10.1007/s11033-014-3491-0
Expression profile of MAGI2 gene as a novel biomarker in combination with major deregulated genes in prostate cancer Reza Mahdian • Vahideh Nodouzi • Mojgan Asgari • Mitra Rezaie • Javad Alizadeh • Behzad Yousefi • Hossein Shahrokh • Maryam Abolhasani Mohamadreza Nowroozi
•
Received: 9 September 2013 / Accepted: 17 June 2014 / Published online: 2 July 2014 Ó Springer Science+Business Media Dordrecht 2014
Abstract Complex molecular changes that occur during prostate cancer (PCa) progression have been described recently. Whole genome sequencing of primary PCa samples has identified recurrent gene deletions and rearrangements in PCa. Specifically, these molecular events disrupt the gene loci of phosphatase and tensin homolog (PTEN) and membrane-associated guanylate kinase inverted-2 (MAGI2). In the present study, we analyzed the expression profile of MAGI2 gene in a cohort of clinical PCa (n = 45) and benign prostatic hyperplasia (BPH) samples (n = 36) as well as three PCa cell lines. We also studied the expression of PCa-related genes, including PTEN, NKX3.1, SPINK1, DD3, AMACR, ERG, and TMPRSS2-ERG fusion in the same samples. The expression of MAGI2 mRNA was significantly down-regulated in PC3, LNCaP and DU-145 PCa cell lines (p = 0.000), and also in clinical tumor samples (Relative expression = 0.307, p = 0.002, [95 %
CI 0.002–12.08]). The expression of PTEN, NKX3.1, SPINK1, DD3, and AMACR genes was significantly deregulated in prostate tumor samples (p range 0.000–0.044). A significant correlation was observed between MAGI2 and NKX3.1 expression in tumor samples (p = 0.006). Furthermore, the inclusion of MAGI2 in the gene panel improved the accuracy for discrimination between PCa and BPH samples with the sensitivity and specificity of 0.88 [CI 0.76–0.95] and 0.83 [CI 0.68–0.92], respectively. The data presented here suggest that MAGI2 gene can be considered as a novel component of gene signatures for the detection of PCa. Keywords MAGI2 PTEN Prostate cancer Gene expression BPH
Introduction R. Mahdian (&) V. Nodouzi J. Alizadeh Molecular Medicine Department, Biotechnology Research Center, Pasteur Institute of Iran, Tehran 1316943551, Iran e-mail: [email protected] M. Asgari M. Abolhasani Oncopathology Research Center, Iran University of Medical Science (IUMS), Tehran, Iran M. Asgari H. Shahrokh M. Abolhasani Hasheminejad Clinical Research Developing Center (HCRDC), Iran University of Medical Science (IUMS), Tehran, Iran M. Rezaie Flowcytometry Department, Medical Diagnosis and Reference Laboratories of Iranian Blood Transfusion, Tehran, Iran B. Yousefi M. Nowroozi Department of Urology, Uro-oncology Research Center (UORC), Imam Khomeini Hospital, Tehran, Iran
Prostate cancer (PCa) is the second most common cause of male cancer deaths in the United States. However, the full range of molecular changes associated with PCa development has not been characterized yet. Nearly half of all PCa samples harbor various gene deletions and rearrangements. These gene rearrangements may affect the expression of the corresponding mRNAs and proteins [1, 2]. Many PCa cases can be characterized by the status of E26 transformation-specific and/or phosphatase and tensin homolog (PTEN) genes. Nevertheless, more genomic events may occur in smaller subsets of PCa cases. Recently, two reports in the Journal ‘‘Nature’’ have shown complex molecular changes occur during PCa progression [3, 4]. Berger et al [3] completed whole genome sequencing of seven primary PCa samples which identified novel somatic genomic events in PCa [3]. They identified rearrangements
123
6126
that disrupted the PTEN and membrane-associated guanylate kinase inverted-2 (MAGI2) gene loci [3]. MAGI2, known as synaptic scaffolding molecule, belongs to membrane-associated guanylate kinases superfamily. MAGI2 acts as a scaffold protein in assembling and anchoring cellular signaling proteins, such as Atrophin-1, glutamate receptors, neuroligins-1, b1-adrenergic receptor, b-catenin, and PTEN [5]. As a scaffold protein for PTEN, MAGI2 interacts with the C-terminus of PTEN through its PDZ domain and enables PTEN to convert PIP3 into PIP2 [6–8]. Moreover, MAGI2 enhances PTEN activity via decreasing its protein degradation [5–7]. MAGI2 gene rearrangement has been shown in the genome of a melanoma cell line, another cancer type in which PTEN loss is prevalent [3]. Therefore, deregulation of MAGI2 expression either via somatic genomic events or post-transcriptional regulation (e.g. by micro-RNAs) can affect PTEN activity [9]. This fact suggests the involvement of the phosphatidylinositol 3-kinase pathway as a driver of prostate carcinogenesis and a potential therapeutic target. Mutations in PTEN and MAGI2 within the genome of PCa cells appear to be mutually exclusive. MAGI2 was found to be recurrently affected by a copy-neutral rearrangement in PCa [3]. This status makes MAGI2 gene rearrangement invisible to detection methods other than whole genome sequencing or fluorescent in situ hybridization (FISH). Alternatively, gene expression analysis can indirectly detect disruptive genomic events that have affected the mRNA expression [10, 11]. In the present study, for the first time, we analyzed the expression profile of MAGI2 mRNA in a cohort of clinical PCa samples as well as PCa cell lines. We also studied the expression of other PCa-related genes, including PTEN, NKX3.1, SPINK1, DD3, AMACR, ERG, and TMPRSS2-ERG fusion in the same clinical samples.
Patients and methods Tissue sample collection PCa and BPH patients were referred to the urology department at Uro-oncology Research Center (UORC) at Imam Khomeini Hospital or Hasheminejad Clinical Research Development Center (Tehran, Iran) from June 2011 to February 2013. All the patients were new cases with no medical history of surgery or chemotherapy. The median age of the patients was 63 years ranged from 47 to 75. The patients were examined by an expert urologist and evaluated according to the standard imaging procedures and laboratory analyses for PCa. Each patient contributed to the study signed a written informed consent approved by the Ethics Committee of UORC and Pasteur Institute of Iran.
123
Mol Biol Rep (2014) 41:6125–6131
Tissue samples were obtained via open radical prostatectomy. Each tissue sample was sectioned into two replicates. One replicate was examined by a pathologist for the detection of malignant changes and the determination of Gleason score. The other replicate was microdissected to obtain PCa tumor tissue and its matched normal tissue. The tissue samples were instantly immersed in RNAlater solution (Qiagen, Germany) and kept at room temperature for 24 h. Then, the samples were transferred into liquid nitrogen containers for long term storage. To rule out the effect of tumor grade variation on the result of the gene expression analysis, only tumor samples with intermediate Gleason score (Mean ± SEM; 6.8 ± 0.1, Median = 7) were included in the study. The samples were assigned as tumor (T, n = 45) harboring at least 80 % tumor cell content, tumor adjacent normal tissue (N, n = 33), or BPH (n = 36). Mean plasma level of prostate specific antigen (PSA) was 17.82 ± 3.71 ng/ml and 7.71 ± 1.28 ng/ml (Mean ± SEM) in PCa and BPH group, respectively. Total RNA extraction and cDNA synthesis Total RNA containing small RNAs (e.g. miRNAs) was extracted and purified from tissue samples (50 mg) using miRNeasyÒ Mini kit (Qiagen, Germany) according to the kit instruction. High quality RNA samples (A260/ 280 [ 1.8) were used as templates for cDNA synthesis. Fifteen samples were excluded from the study due to the low quality of their RNAs. cDNA synthesis was performed using oligo(dT) or random hexamers (ProtoscriptÒ kit, New England BioLabs, USA) to convert mRNA or noncoding RNAs, respectively. Development of quantitative TaqMan probe real-time PCR assays Exon-exon junction spanning primers and TaqMan probes were designed using Primer ExpressÒ V.3 software (Applied Biosystems, USA) and verified to be specific for their targets by BLAST analysis (Table 1). MAGI2, PTEN, NKX3.1, ERG, AMACR, SPINK1, DD3, and TMPRSS2ERG gene fusion were assigned as target genes. GAPDH and PSA genes were used to normalize the gene expression variations as described previously [12–14]. According to the preliminary experiments, b-actin mRNA was selected as the most stable housekeeping gene in the PCa cell lines and used to normalize the gene expression levels in the cell line experiments. The primers of TMPRSS2-ERG gene fusion were designed to amplify the most common form of the rearrangement occurred by joining of exon 1 of TMPRSS2 and exon 4 of ERG gene. To determine the dynamic range and amplification efficiency of each target, specific PCR
Mol Biol Rep (2014) 41:6125–6131
6127
Table 1 The sequence of the primers and TaqMan probes used in the gene expression analysis Gene
Forward primer
Reverse primer
TaqMan Probe
PSA
TCTGCGGCGGTGTTCTGG
GCTGTGGCTGACCTGAAATAC
TGTGCCGACCCAGCAAGATCAC
DD3
GGTGGGAAGGACCTGATGA
GGGCGAGGCTCATCGAT
AGAAATGCCCGGCCGCCATC
AMACR
GCTGAATCTCCTGGCTGACT
TGCTGTTCCTTCCACCATATTT
TGACCGCACACGCACTGGCAAG
SPINK1
AGGTAACAGGCATCTTTCTT
CCCACAGACAGGGTCATATAT
TGGCCTCTCTTCCCAGGGAGTCA
ERG
GACGACTTCCAGAGGCTCAC
GACGACTTCCAGAGGCTCACC
GACGACTTCCAGAGGCTCACC
TMPRSS2-ERG
AGTAGGCGCGAGCTAAGCA
AGTAGGCGCGAGCTAAGCAG
AGTAGGCGCGAGCTAAGCAG
PTEN
CACACGACGGGAAGACAAG
CCTCTGGTCCTGGTATGAAGA
AGTTCCCTCAGCCGTTACCTGTG
MAGI2
TGTGAGAAGAAAGGTGCTAT
CGTGGTTGCTGTTGGTGTAGG
AAGTCCAGGCTCTGTATCCACC
NKX3.1
CCAGAGCCAGAGCCAGAGG
TCCAACAGATAAGACCCCAA
CTCGGTCTCTGCCAGCGTCTCGG
GAPDH
ACACCCACTCCTCCACCTTT
TCCACCACCCTGTTGCTGTAG
TGGCATTGCCCTCAACGACCAC
b-actin
CAGAGCCTCGCCTTTGCC
CACGATGGAGGGGAAGACG
ACACCCGCCGCCAGCTCACCA
Table 2 The relative expression of the PCa-related genes in the clinical tumor samples Gene
Type
PCR Efficiency
Expression
Std. error
95 % CI.
p value
Result
GAPDH
REF
0.91
1.000
PTEN
TRG
0.94
NKX3.1
TRG
0.89
0.186
0.035–1.06
0.005–10.56
0.000
DOWN
0.003
0.000–0.26
0.000–0.28
0.000
ERG
TRG
DOWN
0.93
0.371
0.006–30.46
0.000–27.05
0.254
AMACR
TRG
0.92
6.918
0.124–44.54
0.000–23.55
0.031
UP
SPINK1
TRG
0.96
5.130
0.088–14.06
0.003–24.19
0.044
UP
DD3
TRG
0.91
1.815–603.14
0.066–595.86
0.000
UP
MAGI2
TRG
0.99
0.307
0.047–3.02
0.002–12.08
0.002
DOWN
PSA
TRG
0.90
1.139
1.000–1.64
0.216–1.94
0.128
104.24
The expression ratio of each gene in tumor group was compared to their expression in the BPH group. The expression of the target genes (TRG) was normalized to the reference genes GAPDH (REF) and PSA. p values less than 0.05 were considered statistically significant for the data analysis
products were purified, sequenced and ligated into TAvector plasmids (Thermo Scientific, USA). The recombinant plasmids were serially diluted and used as templates for the generation of Real-time PCR standard curves. Data analysis Relative mRNA expression in each particular tumor tissue and its adjacent normal tissue was normalized to the geometrical mean of the CT values determined for GAPDH and PSA genes as described previously [12–14]. Three BPH samples were used as external controls in each experiment. Gene expression ratio in each sample was determined relative to the mean DCT value of the BPH samples included in the experiments [15]. Gene expression variations with more than two-fold change were considered significant. Pooled RNA sample from three PBH cases was used as calibrator sample (Relative expression ratio = 1) for the gene expression analysis in the PCa cell lines [16].
The comparison between mean gene expression levels in PCa and BPH group was performed using Relative Expression Software Tool (RESTÓ 2009, Qiagen, Germany) [17]. RESTÓ software compares two or more treatment groups or conditions with data points (CT) in sample or control group for multiple reference and target genes [17]. Chi square analysis (SPSS software, v 16.0) was performed to evaluate the correlation between gene expression variations in the study groups. In comparison and correlation analyses of the gene expression data, p values \ 0.05 were considered statistically significant.
Results Validation of quantitative real-time PCR assays The Real-time PCR experiments were set up using PC3, LNCaP, and DU-145 PCa cell lines. The specificity of each
123
6128
Mol Biol Rep (2014) 41:6125–6131
PCR amplification reaction was verified by agarose gel electrophoresis and sequencing of the amplified fragments through the experiments. The accuracy of the quantitative assays was also verified by the generation of standard curves, which were drawn using template plasmids containing the desired amplicons. The slope and R2 coefficient of the standard curves were determined for each gene. The PCR efficiency of the studied genes is indicated in Table 2.
RELATIVE EXPRESSION (Normalized to BPH)
PSA mRNA expression
A
The quantitative expression analysis of MAGI2 in PCa cell lines showed significant (p = 0.000) down-regulation of the gene expression in PC3, LNCaP, and DU-145 cell lines. The expression of MAGI2 mRNA was almost lost in PC3 and LNCaP cell lines (Fig. 1). PTEN mRNA was not detectable in androgen-independent PC3 cells which harbor homologous deletions of the PTEN gene. The PTEN expression profiles observed in PC3, LNCaP, and DU-145 cells were consistent with the previously published data [18]. MAGI2 expression was down-regulated in clinical tumor sample cohort
∗
∗
MAGI-2 mRNA expression RELATIVE EXPRESSION (Normalized to BPH)
MAGI2 expression was significantly down-regulated in PCa cell lines
B
∗
∗
∗
The expression of MAGI2 was down-regulated in tumor sample group (p = 0.002) suggesting probable involvement of MAGI2 in PCa tumorigenesis (Table 2 and Fig. 2). MAGI2 expression was down-regulated in 72 % of the tumor samples but remained unchanged (18 %) or upregulated (10 %) in the other cases. The mean expression of MAGI2 gene in tumor tissues relative to PBH samples was 0.307 (S.E. 0.047–3.02, [95 % CI 0.002–12.08]). Chi square analysis of the gene expression results showed a significant correlation between MAGI2 and NKX3.1 mRNA expression in tumor samples (p = 0.006). The inclusion of MAGI2 in the gene expression profile increased the assay accuracy
RELATIVE EXPRESSION (Normalized to BPH)
PTEN mRNA expression
C
∗
Fig. 1 Expression profile of MAGI2 and PTEN mRNA in prostate cancer cell lines. PC3, LNCaP, and DU-145 PCa cell lines were cultured in RMPI-1640 medium containing 10 % FBS. Total RNA was extracted from 106 cells and was converted to cDNA using a commercial kit. The relative expression of MAGI2 and PTEN normalized to b-actin gene expression was determined by quantitative TaqMan real-time PCR assay. The mean expression level of the analyzed genes in three BPH samples (calibrator sample) and three prostate cancer samples (positive control) are also indicated. a. PSA mRNA expression was detectable only in LNCap cells, PCa and BPH samples. b. MAGI2 was significantly down-regulated in all the cell lines and PCa sample compared to PBH control samples (p = 0.000). c. PTEN mRNA was not detectable in PC3 cell line which harbor homozygous PTEN gene deletion. (*p \ 0.01)
123
The landscape of the data achieved by the quantitative gene expression assay on each patient tumor sample is illustrated in Fig. 2. NKX3.1 gene was down-regulated in the most tumor tissue samples (Mean relative expression = 0.003, [95 % CI 0.00–0.28], p = 0.000). NKX3.1 expression was almost lost in most of tumor samples and was also significantly decreased in tumor adjacent normal tissues (Mean relative expression = 0.302, [95 % CI 0.011–5.272], p = 0.001). The expression of PTEN, SPINK1, DD3, and AMACR genes was also deregulated in prostate tumor samples compared to the BPH group (p range 0.000–0.044, Table 2 and Fig. 3). However, despite different gene expression levels of these genes in tumor tissue (T) vs. their matched pathologically normal tissue (N) in each PCa patient (Fig. 2), there was no statistically significant difference between the mean expression ratios of these genes in T versus N group. TMPRSS2-ERG gene fusion was detected in 39 % (14/33) of PCa tumor samples. Interestingly, this gene fusion was present in all PCa samples obtained from the patients with distant metastasis (n = 4). The expression of SPINK1 was
Mol Biol Rep (2014) 41:6125–6131
6129
Fig. 2 The landscape of the PCa-related gene expression in the clinical prostate cancer tissue samples. The color intensity of each cell of the diagram demonstrates the gene expression changes as: upregulated ( , fold change [2), unchanged ( , 0.5\ fold change \2) and down-regulated ( , fold change \0.5). The expression ratio of the target genes in tumor tissue (T) and its matched normal tissue
(N) is indicated relative to the mean expression of those genes in the PBH group. The samples with undetectable levels of the target genes are indicated by blank cells ( ). The status of TMPRSS2-ERG gene fusion is indicated as either positive ( ) or negative ( ). The data are indicated for 33 out of 45 PCa patients for whom both tumor and matched normal tissue samples was available
inversely correlated to the presence of the TMPRSS2-ERG fusion discriminating a subset of TMPRSS2-ERG negative tumors. Collectively, the gene expression profile could distinguish between PCa (n = 45) and BPH samples (n = 36) with the sensitivity and specificity of 0.88 [CI 0.76–0.95] and 0.83 [CI 0.68–0.92], respectively. The exclusion of MAGI2 from the gene panel decreased the accuracy of the assay for discrimination between PCa and BPH cases (with the sensitivity and specificity of 0.78 [CI 0.63–0.87] and 0.69 [CI 0.53–0.82], respectively) (Table 3).
during PCa progression [3, 4]. Various gene deletions and rearrangements in the genome of PCa cells have been described by Berger et al. [3]. These gene mutations and rearrangements eventually affect gene expression at either mRNA or protein level. In this study, we examined MAGI2 gene expression in clinical PCa samples. Interestingly, we observed that MAGI2 expression was significantly downregulated in primary PCa tissue as well as in PCa cell lines. Decreased MAGI2 expression would negatively affect PTEN function and enhance the activation of Akt. This was shown in a study on breast cancer by Sachdeva et al. [9]. They showed that the suppression of MAGI2 expression by miR-101 reduces PTEN activity and Akt activation. However, the genomic rearrangements are likely to be the main cause of MAGI2 down regulation in PCa, because miR-101 has been reported to be down-regulated in PCa
Discussion Recent findings about the genomic aberrations in PCa have revealed the complexity of molecular changes that occur
123
6130
Mol Biol Rep (2014) 41:6125–6131
Table 3 The sensitivity and specificity of the gene expression profile for the discrimination between prostate cancer and BPH samples Assay statistical measures
MAGI2 included
MAGI2 excluded
Sensitivity
0.88 [CI 0.76–0.95]
Specificity
0.83 [CI 0.68–0.92]
0.69 [CI 0.53–0.82]
Positive likelihood ratio
5.33 [CI 2.55–11.15]
2.54 [CI 1.5–4.26]
Negative likelihood ratio
0.13 [CI 0.05–0.30]
0.32 [CI 0.17–0.57]
Diagnostic odds ratio
0.78 [CI 0.63–0.87]
40.00 [CI 11.14–143.54]
7.95 [CI 2.93–21.58]
The tumor samples (n = 45) and BPH samples (n = 36) were analyzed in parallel by pathologic assessment (the standard method) and quantitative TaqMan Real-time PCR assay for the gene expression signature, including MAGI2, PTEN, NKX3.1, AMACR, SPINK1, DD3, and TMPRSS2-ERG gene fusion (the alternative method). The assay statistical measures were calculated either when MAGI2 was included or excluded from the gene panel. The inclusion of MAGI2 in the gene panel enhanced the assay accuracy for the discrimination between prostate cancer and BPH samples. The statistical analysis was performed with Wilson score method (Elie et al. BMC Medical Research Methodology 2008, 8:7)
10 2
(Relative to BPH)
Gene expression ratio
10 3
10 1 10 0 10
-1
10-2 10-3
PTEN NKX3.1 ERG
AMACR SPINK1
DD3
MAGI2 SHARPIN PSA
Fig. 3 Box-plot charts demonstrating median expression level of the PCa-related genes in the clinical PCa samples. The quantitative TaqMan probe Real-time PCR assay was performed for the gene expression analysis in PCa and BPH samples. The CT values were transferred to gene expression analysis software (RESTÓ 2009). The gene expression data are shown relative to the BPH group. The statistical analysis results are illustrated as box-plot graph for each gene. Dashed line in each box represents the median value of the data. MAGI2, PTEN, and NKX3.1 genes were significantly down-regulated, whereas DD3, SPINK1, and AMACR genes were up-regulated in tumor tissue samples
MAGI2 at mRNA level in tumor samples further confirmed that molecular events at genomic level (i.e. rearrangements) but not post-transcriptional regulations (i.e. translational repression) might be involved in PCa. In our study on PCa cell lines harboring different genomic alterations in PTEN gene alleles, there was a significant decrease in the expression of MAGI2 while PTEN gene expression was lost only in PC3 cell line. In fact, despite the unchanged expression of PTEN, the expression of MAGI2 was significantly down-regulated in the LNCaP and DU-145 cells, indicating that distinct molecular events might be involved in each gene alteration. Also, whole genome sequencing of primary PCa showed that the genomic events affecting these two genes may occur in a mutually exclusive manner [20]. Our investigation demonstrates a significant correlation between the expression profile of MAGI2 and NKX3.1 in clinical primary PCa samples. This correlation may be mediated via the reduced PTEN protein function as PTEN controls the activity of NKX3.1 through regulation of its expression [21]. PTEN function loss correlates with the decreased expression of NKX3.1 and PCa progression in both mice and humans [22]. In the present study, the down regulation of MAGI2 was not correlated to PTEN mRNA expression in the PCa samples. This may further confirm that different genomic events cause distinct chromosomal aberrations in each gene. The discovery of MAGI2 genomic rearrangements in PCa proposes that interrogating both the PTEN and MAGI2 loci might improve prognostication and patient stratification for clinical trials of PI3 kinase pathway inhibitors [3, 7]. However, the genomic alteration in MAGI2 locus (i.e. balanced rearrangements) are undetectable by conventional molecular methods other than whole genome sequencing or FISH analysis. Here, we showed that the analysis of MAGI2 gene expression, as previously established for other PCa biomarkers [10, 13, 23, 24], may reflect the genomic defects and improve the molecular assays for PCa diagnosis. Acknowledgments This work was funded by Pasteur Institute of Iran (Grant No. 562). The authors would like to thank Ms. M. Saffari at Iranian Biomedical Journal editorial office for her contribution in the final editing of the manuscript.
References [19]. On the other hand, despite the fact that miR-101 suppressed the endogenous MAGI2 protein levels, Realtime PCR assay showed no significant effect of miR-101 on MAGI2 mRNA expression. This finding suggests that the miR-101-mediated suppression of MAGI2 takes place through translational repression instead of mRNA degradation [9]. In the present study, the down regulation of
123
1. Carver BS, Tran J, Gopalan A, Chen Z, Shaikh S, Carracedo A, Alimonti A, Nardella C, Varmeh S, Scardino PT, Cordon-Cardo C, Gerald W, Pandolfi PP (2009) Aberrant ERG expression cooperates with loss of PTEN to promote cancer progression in the prostate. Nat Genet 41(5):619–624 2. King JC, Xu J, Wongvipat J, Hieronymus H, Carver BS, Leung DH, Taylor BS, Sander C, Cardiff RD, Couto SS, Gerald WL, Sawyers CL (2009) Cooperativity of TMPRSS2-ERG with PI3-
Mol Biol Rep (2014) 41:6125–6131
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
kinase pathway activation in prostate oncogenesis. Nat Genet 41(5):524–526 Berger MF, Lawrence MS, Demichelis F, Drier Y, Cibulskis K, Sivachenko AY, Sboner A, Esgueva R, Pflueger D, Sougnez C, Onofrio R, Carter SL, Park K, Habegger L, Ambrogio L, Fennell T, Parkin M, Saksena G, Voet D, Ramos AH, Pugh TJ, Wilkinson J, Fisher S, Winckler W, Mahan S, Ardlie K, Baldwin J, Simons JW, Kitabayashi N, MacDonald TY, Kantoff PW, Chin L, Gabriel SB, Gerstein MB, Golub TR, Meyerson M, Tewari A, Lander ES, Getz G, Rubin MA, Garraway LA (2011) The genomic complexity of primary human prostate cancer. Nature 470(7333):214–220 Ding Z, Wu CJ, Chu GC, Xiao Y, Ho D, Zhang J, Perry SR, Labrot ES, Wu X, Lis R, Hoshida Y, Hiller D, Hu B, Jiang S, Zheng H, Stegh AH, Scott KL, Signoretti S, Bardeesy N, Wang YA, Hill DE, Golub TR, Stampfer MJ, Wong WH, Loda M, Mucci L, Chin L, DePinho RA (2011) SMAD4-dependent barrier constrains prostate cancer growth and metastatic progression. Nature 470(7333):269–273 Hu Y, Li Z, Guo L, Wang L, Zhang L, Cai X, Zhao H, Zha X (2007) MAGI-2 Inhibits cell migration and proliferation via PTEN in human hepatocarcinoma cells. Arch Biochem Biophys 467(1):1–9 Tolkacheva T, Boddapati M, Sanfiz A, Tsuchida K, Kimmelman AC, Chan AM (2001) Regulation of PTEN binding to MAGI-2 by two putative phosphorylation sites at threonine 382 and 383. Cancer Res 61(13):4985–4989 Engelman JA (2009) Targeting PI3 K signalling in cancer: opportunities, challenges and limitations. Nat Rev Cancer 9(8):550–562 Wu X, Hepner K, Castelino-Prabhu S, Do D, Kaye MB, Yuan XJ, Wood J, Ross C, Sawyers CL, Whang YE (2000) Evidence for regulation of the PTEN tumor suppressor by a membrane-localized multi-PDZ domain containing scaffold protein MAGI-2. Proc Natl Acad Sci USA 97(8):4233–4238 Sachdeva M, Wu H, Ru P, Hwang L, Trieu V, Mo YY (2011) MicroRNA-101-mediated Akt activation and estrogen-independent growth. Oncogene 30(7):822–831 Shen MM, Abate-Shen C (2010) Molecular genetics of prostate cancer: new prospects for old challenges. Genes Dev 24(18):1967–2000 Kumar A, Shendure J, Nelson PS (2011) Genome interrupted: sequencing of prostate cancer reveals the importance of chromosomal rearrangements. Genome Med 3(4):23 Groskopf J, Aubin SM, Deras IL, Blase A, Bodrug S, Clark C, Brentano S, Mathis J, Pham J, Meyer T, Cass M, Hodge P, Macairan ML, Marks LS, Rittenhouse H (2006) APTIMA PCA3 molecular urine test: development of a method to aid in the diagnosis of prostate cancer. Clin Chem 52(6):1089–1095
6131 13. Laxman B, Morris DS, Yu J, Siddiqui J, Cao J, Mehra R, Lonigro RJ, Tsodikov A, Wei JT, Tomlins SA, Chinnaiyan AM (2008) A first-generation multiplex biomarker analysis of urine for the early detection of prostate cancer. Cancer Res 68(3):645–649 14. Rice KR, Chen Y, Ali A, Whitman EJ, Blase A, Ibrahim M, Elsamanoudi S, Brassell S, Furusato B, Stingle N, Sesterhenn IA, Petrovics G, Miick S, Rittenhouse H, Groskopf J, McLeod DG, Srivastava S (2010) Evaluation of the ETS-related gene mRNA in urine for the detection of prostate cancer. Clin Cancer Res 16(5):1572–1576 15. Livak KJ, Schmittgen TD (2001) Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) Method. Methods 25(4):402–408 16. Dhanasekaran SM, Barrette TR, Ghosh D, Shah R, Varambally S, Kurachi K, Pienta KJ, Rubin MA, Chinnaiyan AM (2001) Delineation of prognostic biomarkers in prostate cancer. Nature 412(6849):822–826 17. Pfaffl MW, Horgan GW, Dempfle L (2002) Relative expression software tool (REST) for group-wise comparison and statistical analysis of relative expression results in real-time PCR. Nucleic Acids Res 30(9):e36 18. Bastola DR, Pahwa GS, Lin MF, Cheng PW (2002) Downregulation of PTEN/MMAC/TEP1 expression in human prostate cancer cell line DU145 by growth stimuli. Mol Cell Biochem 236(1–2):75–81 19. Varambally S, Cao Q, Mani RS, Shankar S, Wang X, Ateeq B, Laxman B, Cao X, Jing X, Ramnarayanan K, Brenner JC, Yu J, Kim JH, Han B, Tan P, Kumar-Sinha C, Lonigro RJ, Palanisamy N, Maher CA, Chinnaiyan AM (2008) Genomic loss of microRNA-101 leads to overexpression of histone methyltransferase EZH2 in cancer. Science 322(5908):1695–1699 20. Brenner JC, Chinnaiyan AM (2011) Disruptive events in the life of prostate cancer. Cancer Cell 19(3):301–303 21. Lei Q, Jiao J, Xin L, Chang CJ, Wang S, Gao J, Gleave ME, Witte ON, Liu X, Wu H (2006) NKX3.1 stabilizes p53, inhibits AKT activation, and blocks prostate cancer initiation caused by PTEN loss. Cancer Cell 9(5):367–378 22. Kim MJ, Cardiff RD, Desai N, Banach-Petrosky WA, Parsons R, Shen MM, Abate-Shen C (2002) Cooperativity of Nkx3.1 and Pten loss of function in a mouse model of prostate carcinogenesis. Proc Natl Acad Sci USA 99(5):2884–2889 23. Reynolds MA, Kastury K, Groskopf J, Schalken JA, Rittenhouse H (2007) Molecular markers for prostate cancer. Cancer Lett 249(1):5–13 24. Madu CO, Lu Y (2010) Novel diagnostic biomarkers for prostate cancer. J Cancer 1:150–177
123
Expression profile of MAGI2 gene as a novel biomarker in combination with major deregulated genes in prostate cancer Reza Mahdian • Vahideh Nodouzi • Mojgan Asgari • Mitra Rezaie • Javad Alizadeh • Behzad Yousefi • Hossein Shahrokh • Maryam Abolhasani Mohamadreza Nowroozi
•
Received: 9 September 2013 / Accepted: 17 June 2014 / Published online: 2 July 2014 Ó Springer Science+Business Media Dordrecht 2014
Abstract Complex molecular changes that occur during prostate cancer (PCa) progression have been described recently. Whole genome sequencing of primary PCa samples has identified recurrent gene deletions and rearrangements in PCa. Specifically, these molecular events disrupt the gene loci of phosphatase and tensin homolog (PTEN) and membrane-associated guanylate kinase inverted-2 (MAGI2). In the present study, we analyzed the expression profile of MAGI2 gene in a cohort of clinical PCa (n = 45) and benign prostatic hyperplasia (BPH) samples (n = 36) as well as three PCa cell lines. We also studied the expression of PCa-related genes, including PTEN, NKX3.1, SPINK1, DD3, AMACR, ERG, and TMPRSS2-ERG fusion in the same samples. The expression of MAGI2 mRNA was significantly down-regulated in PC3, LNCaP and DU-145 PCa cell lines (p = 0.000), and also in clinical tumor samples (Relative expression = 0.307, p = 0.002, [95 %
CI 0.002–12.08]). The expression of PTEN, NKX3.1, SPINK1, DD3, and AMACR genes was significantly deregulated in prostate tumor samples (p range 0.000–0.044). A significant correlation was observed between MAGI2 and NKX3.1 expression in tumor samples (p = 0.006). Furthermore, the inclusion of MAGI2 in the gene panel improved the accuracy for discrimination between PCa and BPH samples with the sensitivity and specificity of 0.88 [CI 0.76–0.95] and 0.83 [CI 0.68–0.92], respectively. The data presented here suggest that MAGI2 gene can be considered as a novel component of gene signatures for the detection of PCa. Keywords MAGI2 PTEN Prostate cancer Gene expression BPH
Introduction R. Mahdian (&) V. Nodouzi J. Alizadeh Molecular Medicine Department, Biotechnology Research Center, Pasteur Institute of Iran, Tehran 1316943551, Iran e-mail: [email protected] M. Asgari M. Abolhasani Oncopathology Research Center, Iran University of Medical Science (IUMS), Tehran, Iran M. Asgari H. Shahrokh M. Abolhasani Hasheminejad Clinical Research Developing Center (HCRDC), Iran University of Medical Science (IUMS), Tehran, Iran M. Rezaie Flowcytometry Department, Medical Diagnosis and Reference Laboratories of Iranian Blood Transfusion, Tehran, Iran B. Yousefi M. Nowroozi Department of Urology, Uro-oncology Research Center (UORC), Imam Khomeini Hospital, Tehran, Iran
Prostate cancer (PCa) is the second most common cause of male cancer deaths in the United States. However, the full range of molecular changes associated with PCa development has not been characterized yet. Nearly half of all PCa samples harbor various gene deletions and rearrangements. These gene rearrangements may affect the expression of the corresponding mRNAs and proteins [1, 2]. Many PCa cases can be characterized by the status of E26 transformation-specific and/or phosphatase and tensin homolog (PTEN) genes. Nevertheless, more genomic events may occur in smaller subsets of PCa cases. Recently, two reports in the Journal ‘‘Nature’’ have shown complex molecular changes occur during PCa progression [3, 4]. Berger et al [3] completed whole genome sequencing of seven primary PCa samples which identified novel somatic genomic events in PCa [3]. They identified rearrangements
123
6126
that disrupted the PTEN and membrane-associated guanylate kinase inverted-2 (MAGI2) gene loci [3]. MAGI2, known as synaptic scaffolding molecule, belongs to membrane-associated guanylate kinases superfamily. MAGI2 acts as a scaffold protein in assembling and anchoring cellular signaling proteins, such as Atrophin-1, glutamate receptors, neuroligins-1, b1-adrenergic receptor, b-catenin, and PTEN [5]. As a scaffold protein for PTEN, MAGI2 interacts with the C-terminus of PTEN through its PDZ domain and enables PTEN to convert PIP3 into PIP2 [6–8]. Moreover, MAGI2 enhances PTEN activity via decreasing its protein degradation [5–7]. MAGI2 gene rearrangement has been shown in the genome of a melanoma cell line, another cancer type in which PTEN loss is prevalent [3]. Therefore, deregulation of MAGI2 expression either via somatic genomic events or post-transcriptional regulation (e.g. by micro-RNAs) can affect PTEN activity [9]. This fact suggests the involvement of the phosphatidylinositol 3-kinase pathway as a driver of prostate carcinogenesis and a potential therapeutic target. Mutations in PTEN and MAGI2 within the genome of PCa cells appear to be mutually exclusive. MAGI2 was found to be recurrently affected by a copy-neutral rearrangement in PCa [3]. This status makes MAGI2 gene rearrangement invisible to detection methods other than whole genome sequencing or fluorescent in situ hybridization (FISH). Alternatively, gene expression analysis can indirectly detect disruptive genomic events that have affected the mRNA expression [10, 11]. In the present study, for the first time, we analyzed the expression profile of MAGI2 mRNA in a cohort of clinical PCa samples as well as PCa cell lines. We also studied the expression of other PCa-related genes, including PTEN, NKX3.1, SPINK1, DD3, AMACR, ERG, and TMPRSS2-ERG fusion in the same clinical samples.
Patients and methods Tissue sample collection PCa and BPH patients were referred to the urology department at Uro-oncology Research Center (UORC) at Imam Khomeini Hospital or Hasheminejad Clinical Research Development Center (Tehran, Iran) from June 2011 to February 2013. All the patients were new cases with no medical history of surgery or chemotherapy. The median age of the patients was 63 years ranged from 47 to 75. The patients were examined by an expert urologist and evaluated according to the standard imaging procedures and laboratory analyses for PCa. Each patient contributed to the study signed a written informed consent approved by the Ethics Committee of UORC and Pasteur Institute of Iran.
123
Mol Biol Rep (2014) 41:6125–6131
Tissue samples were obtained via open radical prostatectomy. Each tissue sample was sectioned into two replicates. One replicate was examined by a pathologist for the detection of malignant changes and the determination of Gleason score. The other replicate was microdissected to obtain PCa tumor tissue and its matched normal tissue. The tissue samples were instantly immersed in RNAlater solution (Qiagen, Germany) and kept at room temperature for 24 h. Then, the samples were transferred into liquid nitrogen containers for long term storage. To rule out the effect of tumor grade variation on the result of the gene expression analysis, only tumor samples with intermediate Gleason score (Mean ± SEM; 6.8 ± 0.1, Median = 7) were included in the study. The samples were assigned as tumor (T, n = 45) harboring at least 80 % tumor cell content, tumor adjacent normal tissue (N, n = 33), or BPH (n = 36). Mean plasma level of prostate specific antigen (PSA) was 17.82 ± 3.71 ng/ml and 7.71 ± 1.28 ng/ml (Mean ± SEM) in PCa and BPH group, respectively. Total RNA extraction and cDNA synthesis Total RNA containing small RNAs (e.g. miRNAs) was extracted and purified from tissue samples (50 mg) using miRNeasyÒ Mini kit (Qiagen, Germany) according to the kit instruction. High quality RNA samples (A260/ 280 [ 1.8) were used as templates for cDNA synthesis. Fifteen samples were excluded from the study due to the low quality of their RNAs. cDNA synthesis was performed using oligo(dT) or random hexamers (ProtoscriptÒ kit, New England BioLabs, USA) to convert mRNA or noncoding RNAs, respectively. Development of quantitative TaqMan probe real-time PCR assays Exon-exon junction spanning primers and TaqMan probes were designed using Primer ExpressÒ V.3 software (Applied Biosystems, USA) and verified to be specific for their targets by BLAST analysis (Table 1). MAGI2, PTEN, NKX3.1, ERG, AMACR, SPINK1, DD3, and TMPRSS2ERG gene fusion were assigned as target genes. GAPDH and PSA genes were used to normalize the gene expression variations as described previously [12–14]. According to the preliminary experiments, b-actin mRNA was selected as the most stable housekeeping gene in the PCa cell lines and used to normalize the gene expression levels in the cell line experiments. The primers of TMPRSS2-ERG gene fusion were designed to amplify the most common form of the rearrangement occurred by joining of exon 1 of TMPRSS2 and exon 4 of ERG gene. To determine the dynamic range and amplification efficiency of each target, specific PCR
Mol Biol Rep (2014) 41:6125–6131
6127
Table 1 The sequence of the primers and TaqMan probes used in the gene expression analysis Gene
Forward primer
Reverse primer
TaqMan Probe
PSA
TCTGCGGCGGTGTTCTGG
GCTGTGGCTGACCTGAAATAC
TGTGCCGACCCAGCAAGATCAC
DD3
GGTGGGAAGGACCTGATGA
GGGCGAGGCTCATCGAT
AGAAATGCCCGGCCGCCATC
AMACR
GCTGAATCTCCTGGCTGACT
TGCTGTTCCTTCCACCATATTT
TGACCGCACACGCACTGGCAAG
SPINK1
AGGTAACAGGCATCTTTCTT
CCCACAGACAGGGTCATATAT
TGGCCTCTCTTCCCAGGGAGTCA
ERG
GACGACTTCCAGAGGCTCAC
GACGACTTCCAGAGGCTCACC
GACGACTTCCAGAGGCTCACC
TMPRSS2-ERG
AGTAGGCGCGAGCTAAGCA
AGTAGGCGCGAGCTAAGCAG
AGTAGGCGCGAGCTAAGCAG
PTEN
CACACGACGGGAAGACAAG
CCTCTGGTCCTGGTATGAAGA
AGTTCCCTCAGCCGTTACCTGTG
MAGI2
TGTGAGAAGAAAGGTGCTAT
CGTGGTTGCTGTTGGTGTAGG
AAGTCCAGGCTCTGTATCCACC
NKX3.1
CCAGAGCCAGAGCCAGAGG
TCCAACAGATAAGACCCCAA
CTCGGTCTCTGCCAGCGTCTCGG
GAPDH
ACACCCACTCCTCCACCTTT
TCCACCACCCTGTTGCTGTAG
TGGCATTGCCCTCAACGACCAC
b-actin
CAGAGCCTCGCCTTTGCC
CACGATGGAGGGGAAGACG
ACACCCGCCGCCAGCTCACCA
Table 2 The relative expression of the PCa-related genes in the clinical tumor samples Gene
Type
PCR Efficiency
Expression
Std. error
95 % CI.
p value
Result
GAPDH
REF
0.91
1.000
PTEN
TRG
0.94
NKX3.1
TRG
0.89
0.186
0.035–1.06
0.005–10.56
0.000
DOWN
0.003
0.000–0.26
0.000–0.28
0.000
ERG
TRG
DOWN
0.93
0.371
0.006–30.46
0.000–27.05
0.254
AMACR
TRG
0.92
6.918
0.124–44.54
0.000–23.55
0.031
UP
SPINK1
TRG
0.96
5.130
0.088–14.06
0.003–24.19
0.044
UP
DD3
TRG
0.91
1.815–603.14
0.066–595.86
0.000
UP
MAGI2
TRG
0.99
0.307
0.047–3.02
0.002–12.08
0.002
DOWN
PSA
TRG
0.90
1.139
1.000–1.64
0.216–1.94
0.128
104.24
The expression ratio of each gene in tumor group was compared to their expression in the BPH group. The expression of the target genes (TRG) was normalized to the reference genes GAPDH (REF) and PSA. p values less than 0.05 were considered statistically significant for the data analysis
products were purified, sequenced and ligated into TAvector plasmids (Thermo Scientific, USA). The recombinant plasmids were serially diluted and used as templates for the generation of Real-time PCR standard curves. Data analysis Relative mRNA expression in each particular tumor tissue and its adjacent normal tissue was normalized to the geometrical mean of the CT values determined for GAPDH and PSA genes as described previously [12–14]. Three BPH samples were used as external controls in each experiment. Gene expression ratio in each sample was determined relative to the mean DCT value of the BPH samples included in the experiments [15]. Gene expression variations with more than two-fold change were considered significant. Pooled RNA sample from three PBH cases was used as calibrator sample (Relative expression ratio = 1) for the gene expression analysis in the PCa cell lines [16].
The comparison between mean gene expression levels in PCa and BPH group was performed using Relative Expression Software Tool (RESTÓ 2009, Qiagen, Germany) [17]. RESTÓ software compares two or more treatment groups or conditions with data points (CT) in sample or control group for multiple reference and target genes [17]. Chi square analysis (SPSS software, v 16.0) was performed to evaluate the correlation between gene expression variations in the study groups. In comparison and correlation analyses of the gene expression data, p values \ 0.05 were considered statistically significant.
Results Validation of quantitative real-time PCR assays The Real-time PCR experiments were set up using PC3, LNCaP, and DU-145 PCa cell lines. The specificity of each
123
6128
Mol Biol Rep (2014) 41:6125–6131
PCR amplification reaction was verified by agarose gel electrophoresis and sequencing of the amplified fragments through the experiments. The accuracy of the quantitative assays was also verified by the generation of standard curves, which were drawn using template plasmids containing the desired amplicons. The slope and R2 coefficient of the standard curves were determined for each gene. The PCR efficiency of the studied genes is indicated in Table 2.
RELATIVE EXPRESSION (Normalized to BPH)
PSA mRNA expression
A
The quantitative expression analysis of MAGI2 in PCa cell lines showed significant (p = 0.000) down-regulation of the gene expression in PC3, LNCaP, and DU-145 cell lines. The expression of MAGI2 mRNA was almost lost in PC3 and LNCaP cell lines (Fig. 1). PTEN mRNA was not detectable in androgen-independent PC3 cells which harbor homologous deletions of the PTEN gene. The PTEN expression profiles observed in PC3, LNCaP, and DU-145 cells were consistent with the previously published data [18]. MAGI2 expression was down-regulated in clinical tumor sample cohort
∗
∗
MAGI-2 mRNA expression RELATIVE EXPRESSION (Normalized to BPH)
MAGI2 expression was significantly down-regulated in PCa cell lines
B
∗
∗
∗
The expression of MAGI2 was down-regulated in tumor sample group (p = 0.002) suggesting probable involvement of MAGI2 in PCa tumorigenesis (Table 2 and Fig. 2). MAGI2 expression was down-regulated in 72 % of the tumor samples but remained unchanged (18 %) or upregulated (10 %) in the other cases. The mean expression of MAGI2 gene in tumor tissues relative to PBH samples was 0.307 (S.E. 0.047–3.02, [95 % CI 0.002–12.08]). Chi square analysis of the gene expression results showed a significant correlation between MAGI2 and NKX3.1 mRNA expression in tumor samples (p = 0.006). The inclusion of MAGI2 in the gene expression profile increased the assay accuracy
RELATIVE EXPRESSION (Normalized to BPH)
PTEN mRNA expression
C
∗
Fig. 1 Expression profile of MAGI2 and PTEN mRNA in prostate cancer cell lines. PC3, LNCaP, and DU-145 PCa cell lines were cultured in RMPI-1640 medium containing 10 % FBS. Total RNA was extracted from 106 cells and was converted to cDNA using a commercial kit. The relative expression of MAGI2 and PTEN normalized to b-actin gene expression was determined by quantitative TaqMan real-time PCR assay. The mean expression level of the analyzed genes in three BPH samples (calibrator sample) and three prostate cancer samples (positive control) are also indicated. a. PSA mRNA expression was detectable only in LNCap cells, PCa and BPH samples. b. MAGI2 was significantly down-regulated in all the cell lines and PCa sample compared to PBH control samples (p = 0.000). c. PTEN mRNA was not detectable in PC3 cell line which harbor homozygous PTEN gene deletion. (*p \ 0.01)
123
The landscape of the data achieved by the quantitative gene expression assay on each patient tumor sample is illustrated in Fig. 2. NKX3.1 gene was down-regulated in the most tumor tissue samples (Mean relative expression = 0.003, [95 % CI 0.00–0.28], p = 0.000). NKX3.1 expression was almost lost in most of tumor samples and was also significantly decreased in tumor adjacent normal tissues (Mean relative expression = 0.302, [95 % CI 0.011–5.272], p = 0.001). The expression of PTEN, SPINK1, DD3, and AMACR genes was also deregulated in prostate tumor samples compared to the BPH group (p range 0.000–0.044, Table 2 and Fig. 3). However, despite different gene expression levels of these genes in tumor tissue (T) vs. their matched pathologically normal tissue (N) in each PCa patient (Fig. 2), there was no statistically significant difference between the mean expression ratios of these genes in T versus N group. TMPRSS2-ERG gene fusion was detected in 39 % (14/33) of PCa tumor samples. Interestingly, this gene fusion was present in all PCa samples obtained from the patients with distant metastasis (n = 4). The expression of SPINK1 was
Mol Biol Rep (2014) 41:6125–6131
6129
Fig. 2 The landscape of the PCa-related gene expression in the clinical prostate cancer tissue samples. The color intensity of each cell of the diagram demonstrates the gene expression changes as: upregulated ( , fold change [2), unchanged ( , 0.5\ fold change \2) and down-regulated ( , fold change \0.5). The expression ratio of the target genes in tumor tissue (T) and its matched normal tissue
(N) is indicated relative to the mean expression of those genes in the PBH group. The samples with undetectable levels of the target genes are indicated by blank cells ( ). The status of TMPRSS2-ERG gene fusion is indicated as either positive ( ) or negative ( ). The data are indicated for 33 out of 45 PCa patients for whom both tumor and matched normal tissue samples was available
inversely correlated to the presence of the TMPRSS2-ERG fusion discriminating a subset of TMPRSS2-ERG negative tumors. Collectively, the gene expression profile could distinguish between PCa (n = 45) and BPH samples (n = 36) with the sensitivity and specificity of 0.88 [CI 0.76–0.95] and 0.83 [CI 0.68–0.92], respectively. The exclusion of MAGI2 from the gene panel decreased the accuracy of the assay for discrimination between PCa and BPH cases (with the sensitivity and specificity of 0.78 [CI 0.63–0.87] and 0.69 [CI 0.53–0.82], respectively) (Table 3).
during PCa progression [3, 4]. Various gene deletions and rearrangements in the genome of PCa cells have been described by Berger et al. [3]. These gene mutations and rearrangements eventually affect gene expression at either mRNA or protein level. In this study, we examined MAGI2 gene expression in clinical PCa samples. Interestingly, we observed that MAGI2 expression was significantly downregulated in primary PCa tissue as well as in PCa cell lines. Decreased MAGI2 expression would negatively affect PTEN function and enhance the activation of Akt. This was shown in a study on breast cancer by Sachdeva et al. [9]. They showed that the suppression of MAGI2 expression by miR-101 reduces PTEN activity and Akt activation. However, the genomic rearrangements are likely to be the main cause of MAGI2 down regulation in PCa, because miR-101 has been reported to be down-regulated in PCa
Discussion Recent findings about the genomic aberrations in PCa have revealed the complexity of molecular changes that occur
123
6130
Mol Biol Rep (2014) 41:6125–6131
Table 3 The sensitivity and specificity of the gene expression profile for the discrimination between prostate cancer and BPH samples Assay statistical measures
MAGI2 included
MAGI2 excluded
Sensitivity
0.88 [CI 0.76–0.95]
Specificity
0.83 [CI 0.68–0.92]
0.69 [CI 0.53–0.82]
Positive likelihood ratio
5.33 [CI 2.55–11.15]
2.54 [CI 1.5–4.26]
Negative likelihood ratio
0.13 [CI 0.05–0.30]
0.32 [CI 0.17–0.57]
Diagnostic odds ratio
0.78 [CI 0.63–0.87]
40.00 [CI 11.14–143.54]
7.95 [CI 2.93–21.58]
The tumor samples (n = 45) and BPH samples (n = 36) were analyzed in parallel by pathologic assessment (the standard method) and quantitative TaqMan Real-time PCR assay for the gene expression signature, including MAGI2, PTEN, NKX3.1, AMACR, SPINK1, DD3, and TMPRSS2-ERG gene fusion (the alternative method). The assay statistical measures were calculated either when MAGI2 was included or excluded from the gene panel. The inclusion of MAGI2 in the gene panel enhanced the assay accuracy for the discrimination between prostate cancer and BPH samples. The statistical analysis was performed with Wilson score method (Elie et al. BMC Medical Research Methodology 2008, 8:7)
10 2
(Relative to BPH)
Gene expression ratio
10 3
10 1 10 0 10
-1
10-2 10-3
PTEN NKX3.1 ERG
AMACR SPINK1
DD3
MAGI2 SHARPIN PSA
Fig. 3 Box-plot charts demonstrating median expression level of the PCa-related genes in the clinical PCa samples. The quantitative TaqMan probe Real-time PCR assay was performed for the gene expression analysis in PCa and BPH samples. The CT values were transferred to gene expression analysis software (RESTÓ 2009). The gene expression data are shown relative to the BPH group. The statistical analysis results are illustrated as box-plot graph for each gene. Dashed line in each box represents the median value of the data. MAGI2, PTEN, and NKX3.1 genes were significantly down-regulated, whereas DD3, SPINK1, and AMACR genes were up-regulated in tumor tissue samples
MAGI2 at mRNA level in tumor samples further confirmed that molecular events at genomic level (i.e. rearrangements) but not post-transcriptional regulations (i.e. translational repression) might be involved in PCa. In our study on PCa cell lines harboring different genomic alterations in PTEN gene alleles, there was a significant decrease in the expression of MAGI2 while PTEN gene expression was lost only in PC3 cell line. In fact, despite the unchanged expression of PTEN, the expression of MAGI2 was significantly down-regulated in the LNCaP and DU-145 cells, indicating that distinct molecular events might be involved in each gene alteration. Also, whole genome sequencing of primary PCa showed that the genomic events affecting these two genes may occur in a mutually exclusive manner [20]. Our investigation demonstrates a significant correlation between the expression profile of MAGI2 and NKX3.1 in clinical primary PCa samples. This correlation may be mediated via the reduced PTEN protein function as PTEN controls the activity of NKX3.1 through regulation of its expression [21]. PTEN function loss correlates with the decreased expression of NKX3.1 and PCa progression in both mice and humans [22]. In the present study, the down regulation of MAGI2 was not correlated to PTEN mRNA expression in the PCa samples. This may further confirm that different genomic events cause distinct chromosomal aberrations in each gene. The discovery of MAGI2 genomic rearrangements in PCa proposes that interrogating both the PTEN and MAGI2 loci might improve prognostication and patient stratification for clinical trials of PI3 kinase pathway inhibitors [3, 7]. However, the genomic alteration in MAGI2 locus (i.e. balanced rearrangements) are undetectable by conventional molecular methods other than whole genome sequencing or FISH analysis. Here, we showed that the analysis of MAGI2 gene expression, as previously established for other PCa biomarkers [10, 13, 23, 24], may reflect the genomic defects and improve the molecular assays for PCa diagnosis. Acknowledgments This work was funded by Pasteur Institute of Iran (Grant No. 562). The authors would like to thank Ms. M. Saffari at Iranian Biomedical Journal editorial office for her contribution in the final editing of the manuscript.
References [19]. On the other hand, despite the fact that miR-101 suppressed the endogenous MAGI2 protein levels, Realtime PCR assay showed no significant effect of miR-101 on MAGI2 mRNA expression. This finding suggests that the miR-101-mediated suppression of MAGI2 takes place through translational repression instead of mRNA degradation [9]. In the present study, the down regulation of
123
1. Carver BS, Tran J, Gopalan A, Chen Z, Shaikh S, Carracedo A, Alimonti A, Nardella C, Varmeh S, Scardino PT, Cordon-Cardo C, Gerald W, Pandolfi PP (2009) Aberrant ERG expression cooperates with loss of PTEN to promote cancer progression in the prostate. Nat Genet 41(5):619–624 2. King JC, Xu J, Wongvipat J, Hieronymus H, Carver BS, Leung DH, Taylor BS, Sander C, Cardiff RD, Couto SS, Gerald WL, Sawyers CL (2009) Cooperativity of TMPRSS2-ERG with PI3-
Mol Biol Rep (2014) 41:6125–6131
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
kinase pathway activation in prostate oncogenesis. Nat Genet 41(5):524–526 Berger MF, Lawrence MS, Demichelis F, Drier Y, Cibulskis K, Sivachenko AY, Sboner A, Esgueva R, Pflueger D, Sougnez C, Onofrio R, Carter SL, Park K, Habegger L, Ambrogio L, Fennell T, Parkin M, Saksena G, Voet D, Ramos AH, Pugh TJ, Wilkinson J, Fisher S, Winckler W, Mahan S, Ardlie K, Baldwin J, Simons JW, Kitabayashi N, MacDonald TY, Kantoff PW, Chin L, Gabriel SB, Gerstein MB, Golub TR, Meyerson M, Tewari A, Lander ES, Getz G, Rubin MA, Garraway LA (2011) The genomic complexity of primary human prostate cancer. Nature 470(7333):214–220 Ding Z, Wu CJ, Chu GC, Xiao Y, Ho D, Zhang J, Perry SR, Labrot ES, Wu X, Lis R, Hoshida Y, Hiller D, Hu B, Jiang S, Zheng H, Stegh AH, Scott KL, Signoretti S, Bardeesy N, Wang YA, Hill DE, Golub TR, Stampfer MJ, Wong WH, Loda M, Mucci L, Chin L, DePinho RA (2011) SMAD4-dependent barrier constrains prostate cancer growth and metastatic progression. Nature 470(7333):269–273 Hu Y, Li Z, Guo L, Wang L, Zhang L, Cai X, Zhao H, Zha X (2007) MAGI-2 Inhibits cell migration and proliferation via PTEN in human hepatocarcinoma cells. Arch Biochem Biophys 467(1):1–9 Tolkacheva T, Boddapati M, Sanfiz A, Tsuchida K, Kimmelman AC, Chan AM (2001) Regulation of PTEN binding to MAGI-2 by two putative phosphorylation sites at threonine 382 and 383. Cancer Res 61(13):4985–4989 Engelman JA (2009) Targeting PI3 K signalling in cancer: opportunities, challenges and limitations. Nat Rev Cancer 9(8):550–562 Wu X, Hepner K, Castelino-Prabhu S, Do D, Kaye MB, Yuan XJ, Wood J, Ross C, Sawyers CL, Whang YE (2000) Evidence for regulation of the PTEN tumor suppressor by a membrane-localized multi-PDZ domain containing scaffold protein MAGI-2. Proc Natl Acad Sci USA 97(8):4233–4238 Sachdeva M, Wu H, Ru P, Hwang L, Trieu V, Mo YY (2011) MicroRNA-101-mediated Akt activation and estrogen-independent growth. Oncogene 30(7):822–831 Shen MM, Abate-Shen C (2010) Molecular genetics of prostate cancer: new prospects for old challenges. Genes Dev 24(18):1967–2000 Kumar A, Shendure J, Nelson PS (2011) Genome interrupted: sequencing of prostate cancer reveals the importance of chromosomal rearrangements. Genome Med 3(4):23 Groskopf J, Aubin SM, Deras IL, Blase A, Bodrug S, Clark C, Brentano S, Mathis J, Pham J, Meyer T, Cass M, Hodge P, Macairan ML, Marks LS, Rittenhouse H (2006) APTIMA PCA3 molecular urine test: development of a method to aid in the diagnosis of prostate cancer. Clin Chem 52(6):1089–1095
6131 13. Laxman B, Morris DS, Yu J, Siddiqui J, Cao J, Mehra R, Lonigro RJ, Tsodikov A, Wei JT, Tomlins SA, Chinnaiyan AM (2008) A first-generation multiplex biomarker analysis of urine for the early detection of prostate cancer. Cancer Res 68(3):645–649 14. Rice KR, Chen Y, Ali A, Whitman EJ, Blase A, Ibrahim M, Elsamanoudi S, Brassell S, Furusato B, Stingle N, Sesterhenn IA, Petrovics G, Miick S, Rittenhouse H, Groskopf J, McLeod DG, Srivastava S (2010) Evaluation of the ETS-related gene mRNA in urine for the detection of prostate cancer. Clin Cancer Res 16(5):1572–1576 15. Livak KJ, Schmittgen TD (2001) Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) Method. Methods 25(4):402–408 16. Dhanasekaran SM, Barrette TR, Ghosh D, Shah R, Varambally S, Kurachi K, Pienta KJ, Rubin MA, Chinnaiyan AM (2001) Delineation of prognostic biomarkers in prostate cancer. Nature 412(6849):822–826 17. Pfaffl MW, Horgan GW, Dempfle L (2002) Relative expression software tool (REST) for group-wise comparison and statistical analysis of relative expression results in real-time PCR. Nucleic Acids Res 30(9):e36 18. Bastola DR, Pahwa GS, Lin MF, Cheng PW (2002) Downregulation of PTEN/MMAC/TEP1 expression in human prostate cancer cell line DU145 by growth stimuli. Mol Cell Biochem 236(1–2):75–81 19. Varambally S, Cao Q, Mani RS, Shankar S, Wang X, Ateeq B, Laxman B, Cao X, Jing X, Ramnarayanan K, Brenner JC, Yu J, Kim JH, Han B, Tan P, Kumar-Sinha C, Lonigro RJ, Palanisamy N, Maher CA, Chinnaiyan AM (2008) Genomic loss of microRNA-101 leads to overexpression of histone methyltransferase EZH2 in cancer. Science 322(5908):1695–1699 20. Brenner JC, Chinnaiyan AM (2011) Disruptive events in the life of prostate cancer. Cancer Cell 19(3):301–303 21. Lei Q, Jiao J, Xin L, Chang CJ, Wang S, Gao J, Gleave ME, Witte ON, Liu X, Wu H (2006) NKX3.1 stabilizes p53, inhibits AKT activation, and blocks prostate cancer initiation caused by PTEN loss. Cancer Cell 9(5):367–378 22. Kim MJ, Cardiff RD, Desai N, Banach-Petrosky WA, Parsons R, Shen MM, Abate-Shen C (2002) Cooperativity of Nkx3.1 and Pten loss of function in a mouse model of prostate carcinogenesis. Proc Natl Acad Sci USA 99(5):2884–2889 23. Reynolds MA, Kastury K, Groskopf J, Schalken JA, Rittenhouse H (2007) Molecular markers for prostate cancer. Cancer Lett 249(1):5–13 24. Madu CO, Lu Y (2010) Novel diagnostic biomarkers for prostate cancer. J Cancer 1:150–177
123
