Up-Regulation of Estrogen Responsive Genes in Hypospadias: Microarray Analysis

Up-Regulation of Estrogen Responsive Genes in Hypospadias: Microarray Analysis Zhong Wang, Ben Chun Liu, Gui Ting Lin, Ching-Shwun Lin, Tom F. Lue, Emily Willingham and Laurence S. Baskin* From the Departments of Urology, Ninth People’s Hospital, Shanghai Jiaotong University, Shanghai, People’s Republic of China, and University of California-San Francisco (BCL, GTL, CSL, TFL, EW, LSB), San Francisco, California

Purpose: An unexplained increase in the incidence of hypospadias has been reported, and yet to our knowledge the molecular events and their regulation leading to hypospadias remain unknown, although environmental compounds capable of endocrine activity are suspected. We screened on a global scale abnormalities in gene expression in human hypospadiac tissue compared to those in nonhypospadiac tissue. Additionally, microarray analysis of tissue from a pair of fraternal twins, including 1 with and 1 without hypospadias, served as a control for genetic variability. We hypothesized that gene expression would differ between hypospadiac vs nonhypospadiac tissue and fraternal twin data would show patterns similar to those of group data on hypospadiac and nonhypospadiac tissue. Materials and Methods: Microarray analysis was performed on tissue from patients with and without hypospadias, and from a pair of fraternal twins, including 1 with and 1 without hypospadias. Analysis incorporated the expression of 22,000 genes. Results: We found significant differences in gene expression, specifically with a group of genes, including CYR61, CTGF, ATF3 and GADD45␤, known to be responsive to estrogen or to interact with estrogen receptor. Conclusions: Our findings provide support for the hypothesis that endocrine active environmental compounds may contribute to the development of hypospadias. Additionally, regulation of these genes may have a role in formation of the urethra. Key Words: ureter; hypospadias; microarray analysis; estrogen; twins, dizygotic

o define the etiology of hypospadias and explain the recent findings of an increase in its incidence, immunohistochemistry and microanatomy studies,1 and studies of the effects of estrogen exposure2 have been performed on mouse and human fetal tissues. Estrogen exposure was examined because hypospadias is linked to the endocrine disruption phenomenon, by which hormonally active compounds in the environment exert effects during vertebrate development, disrupting normal endocrine processes.3 Results showed that hypospadias can be directly induced by exposure to estrogenic compounds.2 Although hypospadias is one of the most frequently occurring congenital anomalies, little is understood about its molecular biology or the gene expression changes that underlie its development.4 Current findings indicate that hypospadias is largely under environmental influence but its development is multifactorial.4 We used microarray analysis to identify genes that are up-regulated in hypospadiac tissue compared to controls and may be estrogen responsive. We

T

Submitted for publication April 12, 2006. Study received approval from the University of California-San Francisco ethics committee. Supported by a grant from the Society of Pediatric Urology, National Institutes of Health Grant RO1 DK058105-01 (LSB) and a grant from the Science Foundation Shanghai (ZW). * Correspondence: Pediatric Urology, University of CaliforniaSan Francisco Children’s Medical Center, 400 Parnassus Ave., A 640, San Francisco, California 94143 (telephone: 415-476-1611; FAX: 415-476-8849; e-mail: [email protected]).

0022-5347/07/1775-1939/0 THE JOURNAL OF UROLOGY® Copyright © 2007 by AMERICAN UROLOGICAL ASSOCIATION

selected fraternal twin brothers, including 1 with and 1 without hypospadias, as a unique opportunity to focus on genetic factors. We also completed microarray analysis on a statistically viable sample of tissue from normal patients and those with hypospadias. Using the microarray approach we screened thousands of genes for changes in expression between the normal twin and the one with hypospadias, and the normal and hypospadiac subject groups. MATERIALS AND METHODS RNA Isolation We obtained foreskin from patients with hypospadias and age matched controls, including foreskin from a pair of 8-month-old fraternal twins, of whom 1 had severe hypospadias and the other underwent elective circumcision. The ethics committee at the University of California-San Francisco approved this study and participant parents provided written informed consent. Tissue was placed in Tri-Reagent RNA extraction solution (lot 81L1811, Sigma®) or RNAlater®. RNA was isolated according to product instructions and quantified by spectrophotometry. Double Strand cDNA Synthesis and Purification Synthesis of first strand cDNA was done from total RNA using protocols described in the Affymetrix® GeneChip® Expression Analysis Manual. Double strand cDNA was synthesized from each total RNA sample via oligo-dT mediated

1939

Vol. 177, 1939-1946, May 2007 Printed in U.S.A. DOI:10.1016/j.juro.2007.01.014

1940

UP-REGULATION OF ESTROGEN RESPONSIVE GENES IN HYPOSPADIAS

reverse transcription. High quality total RNA (2.5 ␮g) was used for cDNA synthesis with a SuperScript® Choice Kit. The first strand reaction was performed using the T7-(dT) 24 primer (100 pMol/␮l, high performance liquid chromatography purified DNA, 5=-GGCCAGTGAATTGTAATACGACTCACTATAGGGAGGCGG-(dT) 24-3=) (Operon, Huntsville, Alabama). Second strand cDNA synthesis from single strand cDNA followed. Purification of double strand cDNA was achieved using a GeneChip cleaning kit. cRNA Synthesis and Purification An in vitro transcription reaction was performed to obtain biotin labeled cRNA from double strand cDNA using a BioArray high yield RNA transcript labeling kit (Affymetrix). The product was purified using a GeneChip Cleanup Kit and quantified by spectrophotometry. Microarray Hybridization and Scanning The use of oligonucleotide expression microarrays (Affymetrix) was described in numerous publications, for example the study by Novak et al.5 Briefly, fragmented biotin labeled cRNA was hybridized to a human expression microarray. Purified cRNA was fragmented, biotin labeled and hybridized to an HG-133A microarray (Affymetrix). For 16 hours at 45C and rotating at 45 rpm in a GeneChip hybridization oven the probe arrays were washed, stained and scanned with a GeneArray® Scanner. GeneChip® Microarray Suite, version 4.01 (Affymetrix) was used to generate the subsequent data used for statistical evaluation. Two sets of arrays were used for all samples. The same cRNA sample was assayed 2 or 3 times in independent experiments to confirm the results. Hybridized components were detected by fluorescence laser scanning and confocal localization, and analyzed on a computer using Microarray Suite, version 5.0 and Excel®. Each microarray contained 2 groups of approximately 16, 25 residue oligonucleotides for each of the 12,422 probe sets and ESTs as well as internal standards. Each member of 1 group was a perfect complementary match for a particular segment of a given gene fragment or EST. The second group contained a single base mismatch in the middle of each oligo. Thus, 1 probe set contained approximately 32 oligonucleotides for each gene fragment or EST for a total of more than 22,000 oligonucleotides plus standards on the microarray. The oligonucleotides for each gene fragment were distributed on the microarray and collated by computer for further analysis. Cluster Analysis Cluster analysis using the CLUSFAVOR program included the 22,000 genes present in the UA-133 array. Clustering was performed using self-organizing maps, which cluster genes based on their profile of gene level changes. These analyses showed multiple clusters of genes, including genes not significantly changed in those groups, and clusters of genes that showed profiles of up-regulation or down-regulation. Data Analyses Signal intensity values of each element from chip data sets were analyzed with Microarray Suite and transformed to an Excel file. Spots with intensity values less than 2-fold that of

the local background were flagged, discarded, corrected and normalized. Using such stringent criteria the possibility of false-positive results was remote. Data were first sorted based on the p value that defined the presence or absence of the signal for each probe set. In the current study the cutoff p value was set at 0.08 to limit false-negative results. Statistical Methods Quantitative variables are expressed as the median and range (minimum to maximum). Correlation coefficients and p values were calculated according to Spearman. The impact of the stabilization procedure on housekeeping gene expression was analyzed using the Mann-Whitney test. The intensity of all genes present was analyzed using 1-way ANOVA (p ⬍0.01) and Tukey’s test. One-way ANOVA was applied with a stringent cutoff of 1 false-positive result per 1,000 genes. Thus, a significant change between any 2 groups was detected and analytical power was considerably increased. Average difference data on the fluorescent signal for each gene was analyzed with standard linear 1-way ANOVA. Genes with corrected p ⬍0.01 were subjected to Tukey’s test for contrasts among the mid shaft hypospadias, severe hypospadias and normal control preparations. RT-PCR Analysis Table 1 lists the oligonucleotide primers used in RT-PCR analyses. We used MacVector® software to search the GenBank™ database and analyze the retrieved gene sequences. RT-PCR was performed in an RT step and a PCR step. Briefly, 2.5 ␮g of each RNA were reverse transcribed in a 20 ␮l reaction volume. This RT product was diluted 100fold with TE buffer. Each dilution (1 ␮l) was used in 10 ␮l PCR to identify the optimal input in the linear amplification range. PCR was performed in a DNA Engine thermocycler (MJ Research, Watertown, Massachusetts) under calculated temperature control. The cycling program for the fraternal

TABLE 1. Oligonucleotide primers

Gene (primer name)

␤-Actin: ␤-Actin-s ␤-Actin-a CTGF: CTGF-s CTGF-a STAT13: STAT13s STAT13a NR4A1: NR4A1s NR4A1a BL34: BL34s BL34a CYR61: CYR61s CYR61a ATF3: ATF3s ATF3a GADD45b: GADD45bs GADD45ba FABP7: FABP7s FABP7a

Sequence

PCR Product Size (bp)

TCTACAATGAGCTGCGTGTG AATGTCACGCACGATTTCCC

368

GAAGGGCAAAAAGTGCATCC GACAGTTGTAATGGCAGGCA

235

CCAGCATAGGAAAGCCACAT CCAGCATAGGAAAGCCACAT

355

AAGTTCGAGGACTTCCAGGT GTTAGCCAGGCAGATGTACT

615

CTCGAGAATCTACAGCCAAG CAACTCTGCGCCTGGATATC

333

GTCAGGTTTACTTACGCTGG TACAATGAGTCCCATCACCC

371

AAGAGTCGGAGAAGCTGGAA AGAATCCAGATGAAAGGCGG

504

ACTGTCTTCCCTTCCTACAG CAGACAAGGACCAACCAAGT

355

GCTGGGAGAAGAGTTTGATG ACCTCCACACCAAGGATAAC

321

UP-REGULATION OF ESTROGEN RESPONSIVE GENES IN HYPOSPADIAS TABLE 2. RNA and cRNA evaluation

TABLE 4. GeneChip information

Hypospadias Control RNA: A260/A280 Used (mg) cRNA: A260/A280 Yield (mg) Amplification (fold) Adjusted yield (␮g) Adjusted concentration Vol (ml) Fragmented remaining (ml) Fragmented remaining (␮l)

1941

Hypospadias

Mid Shaft

Severe

2.03 5.00

2.05 5.00

2.06 4.56

2.01 50.97 657.85 55.48 1.55 41.79 67.73 27.81

1.90 61.89 618.89 50.79 1.78 28.10 99.78 32.20

1.98 48.78 640.07 48.32 1.46 44.58 72.03 28.16

twin pairs was set for 30 cycles at 94C for 10 seconds, 55C for 10 seconds and 72C for 10 seconds, followed by 1 cycle at 72C for 5 minutes For gene GADD45C the condition was set at 94C for 3 minutes, followed by 35 cycles at 94C for 20 seconds, 58C for 20 seconds and 72C for 30 seconds, followed by 1 cycle at 74C for 5 minutes. The cycling program for the larger study was set for 33 cycles at 94C for 10 seconds, 55C for 10 seconds and 72C for 10 seconds, followed by 1 cycle at 72C for 5 minutes. PCR products were electrophoresed in 1.5% agarose gel in the presence of ethidium bromide, visualized by ultraviolet fluorescence and recorded using the ChemiImager™ 4000 System. Quantification of PCR products was determined by the densitometry program of the ChemiImager 4000 System. Commercial computer software was used for statistical analysis of the densitometry data with p ⬍0.05 considered significant. RESULTS RNA Preparation A total of 28 total RNA samples were prepared for the microarray and 18 RT-PCR samples were prepared for the larger study. The 18 samples included 6 controls, and 5 mid shaft and 7 severe hypospadias preparations. All samples showed 18S and 28S rRNA peaks on agarose gel electrophoresis with no signs of RNA degradation. The 260/280 nm ratio was 2.04 (range 1.80 to 2.4).

RawQ Scaling factor Background % Present % Absent 3=/5= Ratio: Actin GAPDH

Control

Mid Shaft

Severe

3.02 5.18 68.72 53.2 45.16

2.53 4.33 55.84 55.14 43.24

2.78 3.89 62.21 56.32 42.02

2.13 1.58

1.77 1.6

1.45 1.21

0.61 to 2.88) (table 2). For the fraternal twins the cRNA A260/280 nm ratio of hypospadiac and normal tissue was 1.99 and 2.03, and the concentrations were 0.9 and 1.13 ␮g/␮l, respectively (table 3). Gene Microarray Analysis Table 4 shows gene chip quality. mRNA expression in hypospadias samples, including mid shaft and severe hypospadias preparations, was compared with that in normal samples in the larger study and in the twin study. Microarrays from different experiments were normalized to similar average signals to permit comparisons across different experiments. The average of the 3 replicated experiments was determined and used for fold-change determination by comparing the gene expression fold change relative to the nonhypospadias control (fig. 1). Genes whose expression changed more than 2-fold were selected as significant expression genes (table 5). Those genes included a number of transcription factors, signal pathway, cell cycle, metabolism, nuclear receptor family and structure proteins as well as growth factor receptors. The intensity of all significant genes that were up-regulated was built using SOM and 4 clusters were created (fig. 2). Cluster Analysis We chose select genes that showed increased or decreased expression between hypospadias samples and normal controls, including a number of transcription factors, and cell

cRNA Purification The cRNA A260/280 nm ratio of 18 samples was 1.96 (range 1.86 to 2.11) and the concentration was 1.59 ␮g/␮l (range TABLE 3. cDNA-cRNA results H Elution 1 RNA A260/A280 cRNA: A260/A280 Total yield (mg) Amplification (fold) Adjusted yield (␮g) Adjusted concentration Remaining (␮g) Vol (ml) Vol fragmented remaining (ml) Fragmented remaining (mg)

2

C Elution 3

1.91 1.09 33.87 635.8 58.52 0.9 55.38 61.6 56.76 23.38

1

2

2.00 2.03 23.69

1.99 6.02

1.98 38.63 564.5 51.43 1.13 48.62 43.1 43.24

2.03 17.82

21.62

For H and C first elutions total RNA used was 5.00 mg each and there was PASS cRNA.

FIG. 1. Gene tree of top 94 up-regulated genes whose expression significantly changed between hypospadias and control groups. Color bar (left) indicates different gene expression levels from red— highest up-regulated expression to blue— down-regulation. Each row represents single sample case. Each column represents single gene.

1942

UP-REGULATION OF ESTROGEN RESPONSIVE GENES IN HYPOSPADIAS TABLE 5. Significantly expressed genes that changed more than 2-fold between hypospadias and control groups Fold Change

GenBank Reference Sequence

Gene Name

Mid Shaft vs Control

Mid Shaft Severe vs Control

⫺2.5 4.7

5.1

RGC32 ID1

3.2 2.3

3 2.4

S59049 BF304996

RGS1 RGS16

3.3

8.9 2.9

NM 002923

RGS2

2.7

3.2

NM 001446 NM 001446 AI857639

FABP7 FABP7 PMAIP1

46.5 13.7

20.8 6.6 4.1

NM 000963

PTGS2

NM 004417 NM 006732

NM 004949 BC004490

DSC2 FOS

NM 014059 D13889

Mid Shaft Severe vs Mid Shaft 2.8

9

7.4

PTPN10 FOSB

2.6 16.2

2.5 14.7

NM 002135

NR4A1

13.1

12

AI935096

NR4A2

10.1

10.8

S77154

NR4A2

8.8

8.1

NM 006186

NR4A2

7.8

7.2

D49728

NR4A1

6.9

7.4

NM 000024 L07555

ADRB2 CD69

2.9

3.6 2.5

BG035761 NM 004418 AA675892 NM 005382

SSI-3 DUSP2 TOB1 NEF3

10.3 5.5 2.1 ⫺2.5

12.8 4.8 2.2 ⫺2.7

NM 002922 NM 001674 NM 030751

RGS1 ATF3 TCF8

8.6 4.3

4.9 10.3 4.4

NM 003407

ZFP36

4

4.2

NM 016270 NM 002229 NM 002228

LOC51713 JUNB JUN

3.2 3.2 2.9

2.5 3.3 4

BG491844

JUN

2.8

3.3

AI339541 NM 002467

JUND MYC

2.7 2.7

2.5 2.6

NM 001964 AB017493

EGR1 COPEB

2.7 2.1

2.9 2.2

AV733950

EGR1

4.1

4.6

AF000361

FBP

3

2.1

BE963370 AL556438 AF003114 NM 016140 AI817801 NM 021960

BTF DKFZP434J214 CYR61 LOC51673 BIRC1 MCL1

2.3

4.5 2.5 7.7

AK025862

FLJ22209

AK001816

MTMR1

3.2

2.4

2.5

4.1

2.4

2.6 2.8 3 3.7 2.8

Description Desmocollin 2 v-fos FBJ murine osteosarcoma viral oncogene homologue RGC32 protein Inhibitor of DNA binding 1, dominant neg helix-loop-helix protein Regulator of G-protein signaling 1 Regulator of G-protein signaling 16 Regulator of G-protein signaling 2, 24 kDa Fatty acid binding protein 7, brain Fatty acid binding protein 7, brain Phorbol-12-myristate-13-acetateinduced protein 1 Prostaglandin-endoperoxide synthase 2 (prostaglandin GH synthase ⫹ cyclooxygenase) Dual specificity phosphatase 1 FBJ murine osteosarcoma viral oncogene homologue B Nuclear receptor subfamily 4, group A, member 1 Nuclear receptor subfamily 4, group A, member 2 Nuclear receptor subfamily 4, group A, member 2 Nuclear receptor subfamily 4, group A, member 2 Nuclear receptor subfamily 4, group A, member 1 Adrenergic, ␤-2-, receptor, surface CD69 antigen (p60, early T-cell activation antigen) STAT induced STAT inhibitor 3 Dual specificity phosphatase 2 Transducer of ERBB2, 1 Neurofilament 3 (150 kDa medium) Regulator of G-protein signalling 1 Activating transcription factor 3 Homo sapiens transcription factor 8 (represses interleukin-2 expression) (TCF8), mRNA Zinc finger protein homologous to Zfp-36 in mouse Kruppel-like factor 2 (lung) jun B proto-transcriptione v-jun Avian sarcoma virus 17 transcriptione homologue v-jun Avian sarcoma virus 17 transcriptione homologue jun D proto-oncogene v-myc avian myelocytomatosis viral transcriptione homologue Early growth response 1 Core promoter element binding protein Early growth response 1 Nonfunctional folate binding protein mRNA, complete cds KIAA0164 gene product DKFZP434J214 protein Cysteine-rich, angiogenic inducer, 61 Brain specific protein Baculoviral IAP repeat-containing 1 Myeloid cell leukemia sequence 1 (BCL2-related) Homo sapiens cDNA: FLJ22209 fis, clone HRC01496 Homo sapiens cDNA FLJ10954 fis, clone PLACE1000383, highly similar to Homo sapiens mRNA for MTMR1 protein

Function

Unigene

Cell adhesion Apoptosis

Hs.239727 Hs.25647

Cell cycle DNA binding

Hs.76640 Hs.75424

G-protein coupled G-protein coupled signaling G-protein coupled signaling Metabolism Metabolism Metabolism

Hs.385701 Hs.183601

Metabolism

Hs.196384

Metabolism Nuclear factor

Hs.171695 Hs.75678

Nuclear receptor

Hs.1119

Nuclear receptor

Hs.82120

Nuclear receptor

Hs.82120

Nuclear receptor

Hs.82120

Receptor

Hs.1119

Receptor Signaling

Hs.2551 Hs.82401

Signaling Signaling Signaling Structural protein Transcription Transcription Transcription

Hs.345728 Hs.1183 Hs.178137 Hs.71346

Transcription

Hs.343586

Transcription Transcription Transcription

Hs.107740 Hs.400124 Hs.78465

Transcription

Hs.78465

Transcription Transcription

Hs.2780 Hs.79070

Transcription Transcription

Hs.326035 Hs.285313

Transcription factor Transport

Hs.326035

Unknown protein Unknown protein Cell proliferation Unknown protein Apoptosis Apoptotic program cDNA clone

Hs.80338 Hs.12813 Hs.8867 Hs.279772 Hs.79019 Hs.86386

cDNA clone

Hs.372428

Hs.78944 Hs.26770 Hs.26770 Hs.96

Hs.75258 Hs.460 Hs.232068

Hs.73769

Hs.351438

(table continues)

UP-REGULATION OF ESTROGEN RESPONSIVE GENES IN HYPOSPADIAS

1943

TABLE 5. Continued Fold change GenBank Reference Sequence

Gene Name

Mid Shaft vs control

Mid Shaft Severe vs Control

Mid Shaft Severe vs Mid Shaft

2.6 2.3

Description

NM 001264 L20615 U83981

CDSN CDSN GADD34

NM 014330

GADD34

2.8

NM 015675

GADD45B

3.8

NM 002616 AF112857 NM 001554 AJ224869

PER1 CCNE2 CYR61 CXCR4

2.3

L01639

CXCR4

4.4

NM 005572 NM 005524 NM 017734

LMNA HRY PALMD

2.1 2.4 2.5

NM 030952

DKFZP434J037

2.7

NM 030755

TMX

NM 022350 NM 000287 NM 021127

LOC64167 PEX6 PMAIP1

AF162666 AA530892 NM 002065

TLK1 PTPN10 GLUL

NM 005887

BRF2

X15132

SOD2

NM 000361 AI582238 AI472757 NM 006664

THBD TRA1 SYNCRIP SCYA27

BF576710

PTP4A1

2.6

AF199015 BF971587

VIL2 TUBB

3 2.6

Homo sapiens hypothetical protein DKFZp434J037 (DKFZP434J037), mRNA Homo sapiens thioredoxin-related transmembrane protein (TMX), mRNA Aminopeptidase Peroxisomal biogenesis factor 6 Phorbol-12-myristate-13-acetateinduced protein 1 Tousled-like kinase 1 Dual specificity phosphatase 1 Glutamate-ammonia ligase (glutamine synthase) Butyrate response factor 2 (EGFresponse factor 2) Superoxide dismutase 2, mitochondrial Thrombomodulin Tumor rejection antigen (gp96) 1 NS1-associated protein 1 Small inducible cytokine subfamily A (Cys-Cys), member 27 Protein tyrosine phosphatase type IVA, member 1 Villin 2 (ezrin) Tubulin, ␤ polypeptide

BF663141

VIL2

4.3

Villin 2 (ezrin)

NM 031308

EPPK1

2.5

AL537457

CMT2E

⫺2.4

J03223

PRG1

2.5

Homo sapiens epiplakin 1 (EPPK1), mRNA Neurofilament, light polypeptide (68 kDa) Proteoglycan 1, secretory granule

BE973687 AF294326 AF220509 BG032366

HRY CBFB TAF9L ILF3

AI762296 BE675435

JUND COPEB

2.9 2.8

AI078167

NFKBIA

2.1

NM 004907 BC002646

ETR101 JUN

2.2 6.8

NM 004235 NM 005239

KLF4 ETS2

3 2.1

AL616854 N33009 NM 014181

SUI1 APOE HSPC159

2.7

2.6 6.7 2.8

2.1 ⫺2.8 11.6 4 2.8 24.9 2.2 2.8 3.1 2.1 2.5 2.1 2.5

2.4 2.1 5.2 4.5

3.5 2.5 2.6

Comeodesmosin Comeodesmosin Apoptosis associated protein (GADD34) mRNA Growth arrest ⫹ DNA-damageinducible 34 Growth arrest ⫹ DNA-damageinducible, ␤ Period (Drosophila) homologue 1 Cyclin E2 Cysteine-rich, angiogenic inducer, 61 Chemokine (C-X-C motif), receptor 4 (fusin) Chemokine (C-X-C motif), receptor 4 (fusin) Lamin AC Hairy (Drosophila)-homologue Hypothetical protein FLJ20271

Hairy (Drosophila)-homologue Core-binding factor, ␤ subunit Neuronal cell death-related protein Interleukin enhancer binding factor 3, 90 kDa Jun D proto-oncogene Core promoter element binding protein Nuclear factor of ␬ light polypeptide gene enhancer in B-cells inhibitor, ␣ Immediate early protein v-jun Avian sarcoma virus 17 oncogene homologue Kruppel-like factor 4 (gut) v-ets Avian erythroblastosis virus E26 transcriptione homologue 2 Putative translation initiation factor Apolipoprotein E HSPC159 protein

Function

Unigene

Cell adhesion Cell adhesion Cell cycle

Hs.507 Hs.507 Hs.76556

Cell cycle

Hs.76556

Cell cycle

Hs.110571

Cell cycle Cell cycle Cell proliferation Chemokine receptor Chemokine receptor Disease related DNA binding Hypothetical protein Hypothetical protein

Hs.68398 Hs.30464 Hs.8867 Hs.69414

Membrane

Hs.24766

Metabolism Metabolism Metabolism

Hs.280380 Hs.301636 Hs.96

Metabolism Metabolism Neurogenesis

Hs.16895 Hs.171695 Hs.170171

Oncogenesis

Hs.78909

Organelle metabolism Receptor Stress response RNA binding Secreted

Hs.372783

Hs.89414 Hs.377973 Hs.250666 Hs.14606 Hs.172012

Hs.2030 Hs.82689 Hs.373499 Hs.225948

Signal transduction Signaling Structural protein Structural protein Structural protein Structural protein Structural protein Tissue specific Transcription Transcription Transcription

Hs.227777

Hs.250666 Hs.179881 Hs.171723 Hs.256583

Transcription Transcription

Hs.2780 Hs.285313

Transcription

Hs.81328

Transcription Transcription

Hs.737 Hs.78465

Transcription Transcription factor Translation Transport Unknown protein

Hs.356370 Hs.85146

Hs.155191 Hs.336780 Hs.155191 Hs.200412 Hs.211584 Hs.1908

Hs.150580 Hs.169401 Hs.372208

1944

UP-REGULATION OF ESTROGEN RESPONSIVE GENES IN HYPOSPADIAS Of the genes that we investigated 3 up-regulated genes were common to the fraternal twin analysis and to the larger study analysis, that is ATF3, CTGF and Cyr61. As it turns out, all 3 of these genes are estrogen responsive.9,10 In addition, GADD45␤ was up-regulated in hypospadiac tissue and it was shown to interact with estrogen receptor.11 These findings are of clinical interest because endocrine disruption was proposed as a mechanism of hypospadias.3

FIG. 2. Total of 94 significant genes in hypospadias samples were divided into 4 clusters. Gene expression pattern in hypospadias was compared with that in controls. Microarray signals from different experiments were normalized to similar average signals to permit comparisons across different experiments.

cycle, cell proliferation, nuclear receptor family and structural proteins (table 5 and fig. 2). Cluster analysis revealed several patterns of genes clustering in the cell proliferation and development areas. RT-PCR A total of 43 RNA samples from age matched boys, including 16 with severe hypospadias, 8 with mid shaft hypospadias and 19 with circumcision serving as controls, were used in RT-PCR to detect ATF3, CYR61, CTGF and GADD45␤. Results demonstrated the expression of CYR61, CTGF and GADD45 compared to expression of the internal control ␤-actin (fig. 3, A). In hypospadias expression was higher than in controls and expression increased with the severity of the anomaly for all of these genes (fig. 3, B and C). In the twin study mRNA specimens from the fraternal twins were analyzed using RT-PCR, targeting the 6 genes that we identified as being of interest and ␤-actin as the internal control (table 1). Results indicated that the expression of all of these genes in hypospadias was up-regulated (fig. 4). For several genes, including ATF3, CYR61 and CTGF, expression was increased in the fraternal twin with hypospadias compared to the twin without hypospadias (fig. 4, B). Figure 4, C shows the ratios of these genes in the twin with vs the twin without hypospadias. DISCUSSION We investigated the interaction of genetics and environment using a pair of fraternal twins, including 1 with and 1 without hypospadias. In addition, we further clarified gene expression patterns in patients with and without hypospadias. Results identified a group of estrogen responsive genes that are a possible risk factor for hypospadias. Most studies in the field focused on chromosome anomalies, such as deletions and alterations,6,7 steroidal disruptions via androgen or estrogen and their receptors, or environmental factors.3 Most patients who present for the evaluation of hypospadias, chordee and undescended testes have a normal karyotype.8 These investigations do not delineate the relative contribution of genetic and environmental factors in the development of hypospadias.

FIG. 3. A, RT-PCR analysis of CYR61, CTCF and GADD45␤ genes, which were up-regulated in hypospadias vs normal age matched preparations. RNA was subjected to RT-PCR with every sense and antisense primer pair, and primer pair ␤-actin (b-ACT)-s and ␤-actin-a. B, median IDV ratio for every gene was used to derive values, eg value of 1.827 for CYR61 under CYR61 column was average of ratios between hypospadias bands of CYR61 and ␤-actin bands with CYR61 vs ␤-actin ratios in hypospadias and control of 1.827 vs 0.747. C, commercial computer software was used for statistical analysis of densitometry data. For CYR61 gene expression difference in hypospadias was significant compared with normal tissue but no difference was noted between mid shaft and severe or mid shaft and normal preparations. For ATF3 there was significant difference among normal, mid shaft and severe hypospadias samples. For GADD45␤ gene there was significant difference between severe hypospadias and normal tissue but no difference between normal and mid shaft hypospadias or mid shaft and severe hypospadias groups. For CTGF there was significant difference between normal and severe hypospadias, and mid shaft and severe hypospadias groups but not between normal and mid shaft hypospadias groups.

UP-REGULATION OF ESTROGEN RESPONSIVE GENES IN HYPOSPADIAS

1945

interaction with these receptors appears to affect receptor transactivation function.11 CONCLUSIONS The gene profile from the twin brothers with and without hypospadias, and the microarray data from a larger group of normal vs hypospadiac tissue demonstrate that the genes discussed were strongly up-regulated in hypospadiac tissue and they were also estrogen responsive. These findings provide support for the idea that estrogenic endocrine active compounds have a role in the etiology of hypospadias and their increase may correlate with the increase in the incidence of this anomaly. FIG. 4. Expression of 6 genes CYR61, ATF3, BL34, FABP7, STAT1 and CTGF in twin brothers, including 1 with hypospadias and 1 control. A, RT-PCR. ACT, ␤-actin. B, qualification of PCR products by ChemiImager 4000 System densitometry program. C, IDV value ratio of 6 genes, corresponding to PCR product qualification.

ATF3 is an estrogen responsive gene.9,10 To our knowledge our study is the first to demonstrate a relationship between ATF3 and hypospadias. We report that ATF3 is up-regulated in the skin of patients with hypospadias compared to normal prepuce. Other studies at our laboratory of ATF3 expression at the mRNA level in fetal mouse tissues demonstrated that its mRNA is expressed significantly more in genital tubercles from fetal mice exposed in utero to estrogens than in those of unexposed fetal mice.12 We also completed a study of ATF3 expression in human tissue, comparing circumcision and hypospadiac tissue in age matched controls, and we found that ATF3 was expressed significantly more in hypospadiac tissue samples.13 We completed another study of fetal hypospadiac tissue and found that ATF3 was expressed in the urethral plate of a fetus with hypospadias but not in the urethral plate of a fetus without hypospadias. Cyr61 and CTGF are another 2 estrogen responsive genes that were up-regulated in hypospadias. Each is a member of the CCN gene family. CCNs in general appear to be estrogen responsive and they were shown to strongly up-regulate in response to estrogen stimulation.14 Cyr61 (CCN1) is a heparin binding, extracellular matrix associated protein related to connective tissue growth factors (CCN2 and CTGF) that are also in the CCN family. These proteins may function in matrix remodeling through the activation of metalloproteinases during angiogenesis and wound healing.15 In expression profile of genes involved in endometriosis Cyr61 was up-regulated in the secretory (estrogen increasing) phase of the menstrual cycle and in fact it was the most consistently up-regulated gene.16 This up-regulation could be suppressed with antiestrogen exposure. Studies of uterine tissues exposed to estradiol showed that Cyr61 protein levels increase in response to such exposure.17 Cyr61 may also have a role in estrogen dependent and growth factor dependent breast cancer proliferation.18 CTGF is part of the TGF␤ signaling pathway and its transcription in the uterus is regulated by steroid hormones, eg estrogen.19 It also appears to mediate the effects of steroid hormones in the endometrium. GADD45 proteins share motifs that were identified as co-activators of nuclear receptors, including estrogen receptor ␣, and their

ACKNOWLEDGMENTS Donglei Hu, Shirley Zhu and Lisa Wang, Affymetrix GeneChip Array Core Laboratory, University of CaliforniaSan Francisco provided assistance.

Abbreviations and Acronyms dT EST IDV PCR RT

⫽ ⫽ ⫽ ⫽ ⫽

deoxythymidine Expressed Sequence Tag indinavir polymerase chain reaction reverse transcriptase

REFERENCES 1. 2.

3.

4. 5.

6. 7.

8.

9.

10.

11.

12.

Baskin LS, Erol A, Li Y, Liu W and Cunha GR: Urethral seam formation and hypospadias. Cell Tissue Res 2001; 305: 379. Kim KS, Torres C, Yucel S, Raimondo K, Cunha GR and Baskin LS: Induction of hypospadias in a murine model by maternal exposure to synthetic estrogens. Environ Res 2004; 94: 267. Baskin LS, Colborn T and Himes K: Hypospadias and endocrine disruption: is there a connection? Environ Health Perspect 2001; 109: 1175. Baskin LS: Hypospadias. Adv Exp Med Biol 2004; 545: 3. Novak JP, Sladik R and Hudson TJ: Characterization of variability in large-scale gene expression data: implication for study design. Genomics 2002; 79: 104. Moreno-Garcia M and Miranda EB: Chromosomal anomalies in cryptorchidism and hypospadias. J Urol 2002; 168: 2170. Tateno T, Sasagawa I, Ashida J, Nakada T and Ogata T: Absence of Y-chromosome microdeletions in patients with isolated hypospadias. Fertil Steril 2000; 72: 299. McAleer IM and Kaplan G: Is routine karyotyping necessary in the evaluation of hypospadias and cryptorchidism? J Urol 2001; 165: 2029. Pedram A, Razandi M, Aitkenhead M, Hughes CC and Levin ER: Integration of the non-genomic and genomic actions of estrogen. Membrane-initiated signaling by steroid to transcription and cell biology. J Biol Chem 2002; 277: 50768. Inoue A, Yoshida N, Omoto Y, Oguchi S, Yamori T, Kiyama R et al: Development of cDNA microarray for expression profiling of estrogen-responsive genes. J Mol Endocrinol 2002; 29: 175. Yi YW, Kim D, Jung N, Hong SS, Lee HS and Bae I: Gadd45 family proteins are coactivators of nuclear hormone receptors. Biochem Biophys Res Commun 2000; 272: 193. Liu B, Agras K, Willingham EJ, Vilela MCB and Baskin LS: Activating transcription factor 3 is estrogen responsive in utero and upregulated during sexual differentiation. Hormon Res 2006; 65: 217.

1946 13.

UP-REGULATION OF ESTROGEN RESPONSIVE GENES IN HYPOSPADIAS

Liu B, Wang Z, Lin G, Agras K, Ebbers M, Willingham E et al: Activating transcription factor 3 is up-regulated in patients with hypospadias. Pediatr Res 2005; 58: 1280. 14. Mason HR, Grove-Strawser D, Rubin BS, Nowak RA and Castellot JJ Jr: Estrogen induces CCN5 expression in the rat uterus in vivo. Endocrinology 2004; 145: 976. 15. Mo FE, Muntean AG, Chen CC, Stolz DB, Watkins SC and Lau LF: CYR61 (CCN1) is essential for placental development and vascular integrity. Mol Cell Biol 2002; 22: 8709. 16. Absenger Y, Hess-Stumpp H, Kreft B, Kratzschmar J, Haendler B, Schutze N et al: Cyr61, a deregulated gene in endometriosis. Mol Hum Reprod 2004; 10: 399.

17.

Hewitt SC, Deroo BJ, Hansen K, Collins J, Grissom S et al: Estrogen receptor-dependent genomic responses in the uterus mirror the biphasic physiological response to estrogen. Mol Endocrinol 2003; 17: 2070. 18. Sampath D, Winneker RC and Zhang ZM: Cyr61, a member of the CCN family, is required for MCF-7 cell proliferation: regulation by 17␤-estradiol and overexpression in human breast cancer. Endocrinology 2001; 142: 2540. 19. Rageh MA, Moussad EE, Wilson AK and Brigstock DR: Steroidal regulation of connective tissue growth factor (CCN2; CTGF) synthesis in the mouse uterus. Clin Pathol Mol Pathol 2001; 54: 338.